Journal articles on the topic 'Machine Learning, Graphical Models, Kernel Methods, Optimization'
To see the other types of publications on this topic, follow the link: Machine Learning, Graphical Models, Kernel Methods, Optimization.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 journal articles for your research on the topic 'Machine Learning, Graphical Models, Kernel Methods, Optimization.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse journal articles on a wide variety of disciplines and organise your bibliography correctly.
1
Deist, Timo M., Andrew Patti, Zhaoqi Wang, David Krane, Taylor Sorenson, and David Craft. "Simulation-assisted machine learning." Bioinformatics 35, no. 20 (March 23, 2019): 4072–80. http://dx.doi.org/10.1093/bioinformatics/btz199.
Abstract:
Abstract Motivation In a predictive modeling setting, if sufficient details of the system behavior are known, one can build and use a simulation for making predictions. When sufficient system details are not known, one typically turns to machine learning, which builds a black-box model of the system using a large dataset of input sample features and outputs. We consider a setting which is between these two extremes: some details of the system mechanics are known but not enough for creating simulations that can be used to make high quality predictions. In this context we propose using approximate simulations to build a kernel for use in kernelized machine learning methods, such as support vector machines. The results of multiple simulations (under various uncertainty scenarios) are used to compute similarity measures between every pair of samples: sample pairs are given a high similarity score if they behave similarly under a wide range of simulation parameters. These similarity values, rather than the original high dimensional feature data, are used to build the kernel. Results We demonstrate and explore the simulation-based kernel (SimKern) concept using four synthetic complex systems—three biologically inspired models and one network flow optimization model. We show that, when the number of training samples is small compared to the number of features, the SimKern approach dominates over no-prior-knowledge methods. This approach should be applicable in all disciplines where predictive models are sought and informative yet approximate simulations are available. Availability and implementation The Python SimKern software, the demonstration models (in MATLAB, R), and the datasets are available at https://github.com/davidcraft/SimKern. Supplementary information Supplementary data are available at Bioinformatics online.
2
Özöğür Akyüz, Süreyya, Gürkan Üstünkar, and Gerhard Wilhelm Weber. "Adapted Infinite Kernel Learning by Multi-Local Algorithm." International Journal of Pattern Recognition and Artificial Intelligence 30, no. 04 (April 12, 2016): 1651004. http://dx.doi.org/10.1142/s0218001416510046.
Abstract:
The interplay of machine learning (ML) and optimization methods is an emerging field of artificial intelligence. Both ML and optimization are concerned with modeling of systems related to real-world problems. Parameter selection for classification models is an important task for ML algorithms. In statistical learning theory, cross-validation (CV) which is the most well-known model selection method can be very time consuming for large data sets. One of the recent model selection techniques developed for support vector machines (SVMs) is based on the observed test point margins. In this study, observed margin strategy is integrated into our novel infinite kernel learning (IKL) algorithm together with multi-local procedure (MLP) which is an optimization technique to find global solution. The experimental results show improvements in accuracy and speed when comparing with multiple kernel learning (MKL) and semi-infinite linear programming (SILP) with CV.
3
Lu, Shengfu, Sa Liu, Mi Li, Xin Shi, and Richeng Li. "Depression Classification Model Based on Emotionally Related Eye-Movement Data and Kernel Extreme Learning Machine." Journal of Medical Imaging and Health Informatics 10, no. 11 (November 1, 2020): 2668–74. http://dx.doi.org/10.1166/jmihi.2020.3198.
Abstract:
The paper constructed a depression classification model based on emotionally related eye-movement data and kernel extreme learn machine (ELM). In order to improve the classification ability of the model, we use particle swarm optimization (PSO) to optimize the model parameters (regularization coefficient C and the parameter σ in the kernel function). At the same time, in order to avoid to be caught in the local optimum and improve PSO's searching ability, we use improved chaotic PSO optimization algorithm and Gauss mutation strategy to increase PSO's particle diversity. The classification results show that the accuracy, sensitivity and specificity of classification models without parameter optimization and Gauss mutation strategy are 80.23%, 80.31% and 79.43%, respectively, while those results of classification model using improved chaotic projection model and Gauss mutation strategy are improved to 88.55%, 87.71% and 89.42%, respectively. Compared with other classification methods of depression, the proposed classification method has better performance on depression recognition.
4
SEEGER, MATTHIAS. "GAUSSIAN PROCESSES FOR MACHINE LEARNING." International Journal of Neural Systems 14, no. 02 (April 2004): 69–106. http://dx.doi.org/10.1142/s0129065704001899.
Abstract:
Gaussian processes (GPs) are natural generalisations of multivariate Gaussian random variables to infinite (countably or continuous) index sets. GPs have been applied in a large number of fields to a diverse range of ends, and very many deep theoretical analyses of various properties are available. This paper gives an introduction to Gaussian processes on a fairly elementary level with special emphasis on characteristics relevant in machine learning. It draws explicit connections to branches such as spline smoothing models and support vector machines in which similar ideas have been investigated. Gaussian process models are routinely used to solve hard machine learning problems. They are attractive because of their flexible non-parametric nature and computational simplicity. Treated within a Bayesian framework, very powerful statistical methods can be implemented which offer valid estimates of uncertainties in our predictions and generic model selection procedures cast as nonlinear optimization problems. Their main drawback of heavy computational scaling has recently been alleviated by the introduction of generic sparse approximations.13,78,31 The mathematical literature on GPs is large and often uses deep concepts which are not required to fully understand most machine learning applications. In this tutorial paper, we aim to present characteristics of GPs relevant to machine learning and to show up precise connections to other "kernel machines" popular in the community. Our focus is on a simple presentation, but references to more detailed sources are provided.
5
Abdelhamid, Abdelaziz A., El-Sayed M. El El-Kenawy, Abdelhameed Ibrahim, and Marwa M. Eid. "Intelligent Wheat Types Classification Model Using New Voting Classifier." Journal of Intelligent Systems and Internet of Things 7, no. 1 (2022): 30–39. http://dx.doi.org/10.54216/jisiot.070103.
Abstract:
When assessing the quality of the grain supply chain's quality, it is essential to identify and authenticate wheat types, as this is where the process begins with the examination of seeds. Manual inspection by eye is used for both grain identification and confirmation. High-speed, low-effort options became available thanks to automatic classification methods based on machine learning and computer vision. To this day, classifying at the varietal level is still challenging. Classification of wheat seeds was performed using machine learning techniques in this work. Wheat area, wheat perimeter, compactness, kernel length, kernel width, asymmetry coefficient, and kernel groove length are the 7 physical parameters used to categorize the seeds. The dataset includes 210 separate instances of wheat kernels, and was compiled from the UCI library. The 70 components of the dataset were selected randomly and included wheat kernels from three different varieties: Kama, Rosa, and Canadian. In the first stage, we use single machine learning models for classification, including multilayer neural networks, decision trees, and support vector machines. Each algorithm's output is measured against that of the machine learning ensemble method, which is optimized using the whale optimization and stochastic fractal search algorithms. In the end, the findings show that the proposed optimized ensemble is achieving promising results when compared to single machine learning models.
6
Ramasamy, Lakshmana Kumar, Seifedine Kadry, and Sangsoon Lim. "Selection of optimal hyper-parameter values of support vector machine for sentiment analysis tasks using nature-inspired optimization methods." Bulletin of Electrical Engineering and Informatics 10, no. 1 (February 1, 2021): 290–98. http://dx.doi.org/10.11591/eei.v10i1.2098.
Abstract:
Sentiment analysis and classification task is used in recommender systems to analyze movie reviews, tweets, Facebook posts, online product reviews, blogs, discussion forums, and online comments in social networks. Usually, the classification is performed using supervised machine learning methods such as support vector machine (SVM) classifier, which have many distinct parameters. The selection of the values for these parameters can greatly influence the classification accuracy and can be addressed as an optimization problem. Here we analyze the use of three heuristics, nature-inspired optimization techniques, cuckoo search optimization (CSO), ant lion optimizer (ALO), and polar bear optimization (PBO), for parameter tuning of SVM models using various kernel functions. We validate our approach for the sentiment classification task of Twitter dataset. The results are compared using classification accuracy metric and the Nemenyi test.
7
Zhao, Xutao, Desheng Zhang, Renhui Zhang, and Bin Xu. "A comparative study of Gaussian process regression with other three machine learning approaches in the performance prediction of centrifugal pump." Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science 236, no. 8 (December 30, 2021): 3938–49. http://dx.doi.org/10.1177/09544062211050542.
Abstract:
Accurate prediction of performance indices using impeller parameters is of great importance for the initial and optimal design of centrifugal pump. In this study, a kernel-based non-parametric machine learning method named with Gaussian process regression (GPR) was proposed, with the purpose of predicting the performance of centrifugal pump with less effort based on available impeller parameters. Nine impeller parameters were defined as model inputs, and the pump performance indices, that is, the head and efficiency, were determined as model outputs. The applicability of three widely used nonlinear kernel functions of GPR including squared exponential (SE), rational quadratic (RQ) and Matern5/2 was investigated, and it was found by comparing with the experimental data that the SE kernel function is more suitable to capture the relationship between impeller parameters and performance indices because of the highest R square and the lowest values of max absolute relative error (MARE), mean absolute proportional error (MAPE), and root mean square error (RMSE). In addition, the results predicted by GPR with SE kernel function were compared with the results given by other three machine learning models. The comparison shows that the GPR with SE kernel function is more accurate and robust than other models in centrifugal pump performance prediction, and its prediction errors and uncertainties are both acceptable in terms of engineering applications. The GPR method is less costly in the performance prediction of centrifugal pump with sufficient accuracy, which can be further used to effectively assist the design and manufacture of centrifugal pump and to speed up the optimization design process of impeller coupled with stochastic optimization methods.
8
Alarfaj, Fawaz Khaled, Naveed Ahmad Khan, Muhammad Sulaiman, and Abdullah M. Alomair. "Application of a Machine Learning Algorithm for Evaluation of Stiff Fractional Modeling of Polytropic Gas Spheres and Electric Circuits." Symmetry 14, no. 12 (November 23, 2022): 2482. http://dx.doi.org/10.3390/sym14122482.
Abstract:
Fractional polytropic gas sphere problems and electrical engineering models typically simulated with interconnected circuits have numerous applications in physical, astrophysical phenomena, and thermionic currents. Generally, most of these models are singular-nonlinear, symmetric, and include time delay, which has increased attention to them among researchers. In this work, we explored deep neural networks (DNNs) with an optimization algorithm to calculate the approximate solutions for nonlinear fractional differential equations (NFDEs). The target data-driven design of the DNN-LM algorithm was further implemented on the fractional models to study the rigorous impact and symmetry of different parameters on RL, RC circuits, and polytropic gas spheres. The targeted data generated from the analytical and numerical approaches in the literature for different cases were utilized by the deep neural networks to predict the numerical solutions by minimizing the differences in mean square error using the Levenberg–Marquardt algorithm. The numerical solutions obtained by the designed technique were contrasted with the multi-step reproducing kernel Hilbert space method (MS-RKM), Laplace transformation method (LTM), and Padé approximations. The results demonstrate the accuracy of the design technique as the DNN-LM algorithm overlaps with the actual results with minimum percentage absolute errors that lie between 10−8 and 10−12. The extensive graphical and statistical analysis of the designed technique showed that the DNN-LM algorithm is dependable and facilitates the examination of higher-order nonlinear complex problems due to the flexibility of the DNN architecture and the effectiveness of the optimization procedure.
9
Mei, Wenjuan, Zhen Liu, Yuanzhang Su, Li Du, and Jianguo Huang. "Evolved-Cooperative Correntropy-Based Extreme Learning Machine for Robust Prediction." Entropy 21, no. 9 (September 19, 2019): 912. http://dx.doi.org/10.3390/e21090912.
Abstract:
In recent years, the correntropy instead of the mean squared error has been widely taken as a powerful tool for enhancing the robustness against noise and outliers by forming the local similarity measurements. However, most correntropy-based models either have too simple descriptions of the correntropy or require too many parameters to adjust in advance, which is likely to cause poor performance since the correntropy fails to reflect the probability distributions of the signals. Therefore, in this paper, a novel correntropy-based extreme learning machine (ELM) called ECC-ELM has been proposed to provide a more robust training strategy based on the newly developed multi-kernel correntropy with the parameters that are generated using cooperative evolution. To achieve an accurate description of the correntropy, the method adopts a cooperative evolution which optimizes the bandwidths by switching delayed particle swarm optimization (SDPSO) and generates the corresponding influence coefficients that minimizes the minimum integrated error (MIE) to adaptively provide the best solution. The simulated experiments and real-world applications show that cooperative evolution can achieve the optimal solution which provides an accurate description on the probability distribution of the current error in the model. Therefore, the multi-kernel correntropy that is built with the optimal solution results in more robustness against the noise and outliers when training the model, which increases the accuracy of the predictions compared with other methods.
10
Correa-Jullian, Camila, Sergio Cofre-Martel, Gabriel San Martin, Enrique Lopez Droguett, Gustavo de Novaes Pires Leite, and Alexandre Costa. "Exploring Quantum Machine Learning and Feature Reduction Techniques for Wind Turbine Pitch Fault Detection." Energies 15, no. 8 (April 11, 2022): 2792. http://dx.doi.org/10.3390/en15082792.
Abstract:
Driven by the development of machine learning (ML) and deep learning techniques, prognostics and health management (PHM) has become a key aspect of reliability engineering research. With the recent rise in popularity of quantum computing algorithms and public availability of first-generation quantum hardware, it is of interest to assess their potential for efficiently handling large quantities of operational data for PHM purposes. This paper addresses the application of quantum kernel classification models for fault detection in wind turbine systems (WTSs). The analyzed data correspond to low-frequency SCADA sensor measurements and recorded SCADA alarm logs, focused on the early detection of pitch fault failures. This work aims to explore potential advantages of quantum kernel methods, such as quantum support vector machines (Q-SVMs), over traditional ML approaches and compare principal component analysis (PCA) and autoencoders (AE) as feature reduction tools. Results show that the proposed quantum approach is comparable to conventional ML models in terms of performance and can outperform traditional models (random forest, k-nearest neighbors) for the selected reduced dimensionality of 19 features for both PCA and AE. The overall highest mean accuracies obtained are 0.945 for Gaussian SVM and 0.925 for Q-SVM models.
11
Li, Jianhong, Ken Cai, Huazhou Chen, Lili Xu, Qinyong Lin, and Feng Xu. "Machine Learning Framework for Intelligent Detection of Wastewater Pollution by IoT-Based Spectral Technology." Wireless Communications and Mobile Computing 2022 (February 11, 2022): 1–10. http://dx.doi.org/10.1155/2022/9203335.
Abstract:
Industrial wastewater contains excessive micro insoluble solids (MIS) that probably cause environmental pollutions. Near-infrared (NIR) spectroscopy is an advanced technology for rapid detection of the complex targets in wastewater. An Internet of Things (IoT) platform would support intelligent application of the NIR technologies. The studies of intelligent chemometric methods mainly contribute to improve the NIR calibration model based on the IoT platform. With the development of artificial intelligence, the backward interval and synergy interval techniques were proposed in combination use with the least square support vector machine (LSSVM) method, for adaptive selection of the informative spectral wavelength variables. The radial basis function (RBF) kernel is applied for nonlinear mapping. The regulation parameter and the kernel width are fused together for smart optimization. In the design for waveband autofittings, the total of digital wavelengths in the full scanning range was split into 43 equivalent subintervals, and then, the back interval LSSVM (biLSSVM) and the synergy interval LSSVM (siLSSVM) models were both established for the improvement of prediction results based on the adaptive selection of quasidiscrete variable combination. In comparison with some common linear and nonlinear models, the best training model was acquired with the siLSSVM method while the best testing model was obtained with biLSSVM. The intelligent optimization of model parameters indicated that the proposed biLSSVM and siLSSVM deep learning methodologies are feasible to improve the model prediction results in rapid determination of the wastewater MIS content by the IoT-based NIR technology. The machine learning framework is prospectively applied to the fast assessment of the environmental risk of industrial pollutions and water safety.
12
Cai, Zhennao, Jianhua Gu, Caiyun Wen, Dong Zhao, Chunyu Huang, Hui Huang, Changfei Tong, Jun Li, and Huiling Chen. "An Intelligent Parkinson’s Disease Diagnostic System Based on a Chaotic Bacterial Foraging Optimization Enhanced Fuzzy KNN Approach." Computational and Mathematical Methods in Medicine 2018 (June 21, 2018): 1–24. http://dx.doi.org/10.1155/2018/2396952.
Abstract:
Parkinson's disease (PD) is a common neurodegenerative disease, which has attracted more and more attention. Many artificial intelligence methods have been used for the diagnosis of PD. In this study, an enhanced fuzzy k-nearest neighbor (FKNN) method for the early detection of PD based upon vocal measurements was developed. The proposed method, an evolutionary instance-based learning approach termed CBFO-FKNN, was developed by coupling the chaotic bacterial foraging optimization with Gauss mutation (CBFO) approach with FKNN. The integration of the CBFO technique efficiently resolved the parameter tuning issues of the FKNN. The effectiveness of the proposed CBFO-FKNN was rigorously compared to those of the PD datasets in terms of classification accuracy, sensitivity, specificity, and AUC (area under the receiver operating characteristic curve). The simulation results indicated the proposed approach outperformed the other five FKNN models based on BFO, particle swarm optimization, Genetic algorithms, fruit fly optimization, and firefly algorithm, as well as three advanced machine learning methods including support vector machine (SVM), SVM with local learning-based feature selection, and kernel extreme learning machine in a 10-fold cross-validation scheme. The method presented in this paper has a very good prospect, which will bring great convenience to the clinicians to make a better decision in the clinical diagnosis.
13
Harafani, Hani. "Support Vector Machine Parameter Optimization to Improve Liver Disease Estimation with Genetic Algorithm." SinkrOn 4, no. 2 (April 2, 2020): 106. http://dx.doi.org/10.33395/sinkron.v4i2.10524.
Abstract:
Liver disease is an important public health problem. Over the past few decades, machine learning has developed rapidly and it has been introduced for application in medical-related. In this study we propose Support Vector Machine optimization parameter with genetic algorithm to get a higher performance of Root Mean Square Error value of SVM in order to estimate the liver disorder. The experiment was carried out in three stages, the first step was to try the three SVM kernels with different combination of parameters manually, The second step was to try some combination of range parameters in the genetic algorithm to find the optimal value in the SVM kernel. The third step is comparing the results of the GA-SVM experiment with other regression methods. The results prove that GA has an influence on improving the performance of GA-SVM which has the lowest RMSE value compared to another regression models.
14
Rahchamani, Ghodrat, Seyed Mojtaba Movahedifar, and Amin Honarbakhsh. "Fusion-Learning-Based Optimization: A Modified Metaheuristic Method for Lightweight High-Performance Concrete Design." Complexity 2022 (March 31, 2022): 1–15. http://dx.doi.org/10.1155/2022/6322834.
Abstract:
In order to build high-quality concrete, it is imperative to know the raw materials in advance. It is possible to accurately predict the quality of concrete and the amount of raw materials used using machine learning-enhanced methods. An automated process based on machine learning strategies is proposed in this paper for predicting the compressive strength of concrete. Fusion-learning-based optimization is used in the proposed approach to generate a strong learner by pooling support vector regression models. The SVR technique proposes an optimization method for finding the kernel radial basis function (RBF) parameters based on improving the innovative gunner algorithm (AIG). As a result of AIG's diverse solutions, local optima are effectively avoided. Therefore, the novelty of our research is that, in solving the uncertainty of predicted outputs based on integrated models, we use fusion-learning-based optimization to improve regression discrimination. We also collected a standard dataset to analyze the proposed algorithm, and subsequently, the dataset was designed from concrete laboratory tests on 244 samples, seven features, and three outputs. Different regression intensities are determined by correlation analysis of responses. Regression fusion is sufficiently accurate to estimate the number of desired outcomes examined based on the appropriate input data sample. The best quality concrete can be achieved with an error rate of less than 5%.
15
Shchelochkov, Oleg A., Irini Manoli, Paul Juneau, Jennifer L. Sloan, Susan Ferry, Jennifer Myles, Megan Schoenfeld, et al. "Severity modeling of propionic acidemia using clinical and laboratory biomarkers." Genetics in Medicine 23, no. 8 (May 18, 2021): 1534–42. http://dx.doi.org/10.1038/s41436-021-01173-2.
Abstract:
Abstract Purpose To conduct a proof-of-principle study to identify subtypes of propionic acidemia (PA) and associated biomarkers. Methods Data from a clinically diverse PA patient population (https://clinicaltrials.gov/ct2/show/NCT02890342) were used to train and test machine learning models, identify PA-relevant biomarkers, and perform validation analysis using data from liver-transplanted participants. k-Means clustering was used to test for the existence of PA subtypes. Expert knowledge was used to define PA subtypes (mild and severe). Given expert classification, supervised machine learning (support vector machine with a polynomial kernel, svmPoly) performed dimensional reduction to define relevant features of each PA subtype. Results Forty participants enrolled in the study; five underwent liver transplant. Analysis with k-means clustering indicated that several PA subtypes may exist on the biochemical continuum. The conventional PA biomarkers, plasma total 2-methylctirate and propionylcarnitine, were not statistically significantly different between nontransplanted and transplanted participants motivating us to search for other biomarkers. Unbiased dimensional reduction using svmPoly revealed that plasma transthyretin, alanine:serine ratio, GDF15, FGF21, and in vivo 1-13C-propionate oxidation, play roles in defining PA subtypes. Conclusion Support vector machine prioritized biomarkers that helped classify propionic acidemia patients according to severity subtypes, with important ramifications for future clinical trials and management of PA. Graphical Abstract
16
Frisoni, Giacomo, Gianluca Moro, Giulio Carlassare, and Antonella Carbonaro. "Unsupervised Event Graph Representation and Similarity Learning on Biomedical Literature." Sensors 22, no. 1 (December 21, 2021): 3. http://dx.doi.org/10.3390/s22010003.
Abstract:
The automatic extraction of biomedical events from the scientific literature has drawn keen interest in the last several years, recognizing complex and semantically rich graphical interactions otherwise buried in texts. However, very few works revolve around learning embeddings or similarity metrics for event graphs. This gap leaves biological relations unlinked and prevents the application of machine learning techniques to promote discoveries. Taking advantage of recent deep graph kernel solutions and pre-trained language models, we propose Deep Divergence Event Graph Kernels (DDEGK), an unsupervised inductive method to map events into low-dimensional vectors, preserving their structural and semantic similarities. Unlike most other systems, DDEGK operates at a graph level and does not require task-specific labels, feature engineering, or known correspondences between nodes. To this end, our solution compares events against a small set of anchor ones, trains cross-graph attention networks for drawing pairwise alignments (bolstering interpretability), and employs transformer-based models to encode continuous attributes. Extensive experiments have been done on nine biomedical datasets. We show that our learned event representations can be effectively employed in tasks such as graph classification, clustering, and visualization, also facilitating downstream semantic textual similarity. Empirical results demonstrate that DDEGK significantly outperforms other state-of-the-art methods.
17
Contardi, Uriel Abe, Paulo Rogério Scalassara, and Douglas Vieira Thomaz. "Benign and Malign Breast Cancer Classification Using Support Vector Machines Optimized with Particle Swarm and Genetic Algorithms." Learning and Nonlinear Models 20, no. 2 (December 31, 2022): 21–33. http://dx.doi.org/10.21528/lnlm-vol20-no2-art2.
Abstract:
Breast cancer is a neoplastic disease that can be diagnosed either as benign or malign according to the growth-rate of the neoplastic lesion. Owing to the relevance of obtaining better detection tools, this work describes the development and optimization of support vector machines for the classification of the types of such cancer. Tests were performed using the breast cancer dataset of the University of Wisconsin Hospitals, USA, available at the Machine Learning Repository of the University of California Irvine. The radial basis function kernel was selected for the classifier and its hyperparameters were refined using two methods: particle swarm optimization and genetic algorithms. The results for the first method exhibited 97.71% accuracy, 96.30% sensitivity, and 98.65% of selectivity. On the other hand, using the second method, the accuracy was 95.78%, with sensitivity and selectivity of 96.73% and 95.25%, respectively. Therefore, there is an indication that these search algorithms are viable tools to optimize machine learning models for the purpose of breast cancer classification.
18
Isabona, Joseph, Agbotiname Lucky Imoize, Stephen Ojo, Dinh-Thuan Do, and Cheng-Chi Lee. "Machine Learning-Based GPR with LBFGS Kernel Parameters Selection for Optimal Throughput Mining in 5G Wireless Networks." Sustainability 15, no. 2 (January 15, 2023): 1678. http://dx.doi.org/10.3390/su15021678.
Abstract:
Considering the ever-growing demand for an efficient method of deductive mining and extrapolative analysis of large-scale dimensional datasets, it is very critical to explore advanced machine learning models and algorithms that can reliably meet the demands of modern cellular networks, satisfying computational efficiency and high precision requirements. One non-parametric supervised machine learning model that finds useful applications in cellular networks is the Gaussian process regression (GPR). The GPR model holds a key controlling kernel function whose hyperparameters can be tuned to enhance its supervised predictive learning and adaptive modeling capabilities. In this paper, the limited-memory Broyden–Fletcher–Goldfarb–Shanno (LBFGS) with kernel parameters selection (KPS) algorithm is employed to tune the GPR model kernel hyperparameters rather than using the standard Bayesian optimization (BOP), which is computationally expensive and does not guarantee substantive precision accuracy in the extrapolative analysis of a large-scale dimensional dataset. In particular, the hybrid GPR–LBFGS is exploited for adaptive optimal extrapolative learning and estimation of throughput data obtained from an operational 5G new radio network. The extrapolative learning accuracy of the proposed GPR–LBFGS with the KPS algorithm was analyzed and compared using standard performance metrics such as the mean absolute error, mean percentage error, root mean square error and correlation coefficient. Generally, results revealed that the GPR model combined with the LBFGS kernel hyperparameter selection is superior to the Bayesian hyperparameter selection method. Specifically, at a 25 m distance, the proposed GPR–LBFGS with the KPS method attained 0.16 MAE accuracy in throughput data prediction. In contrast, the other methods attained 46.06 and 53.68 MAE accuracies. Similarly, at 50 m, 75 m, 100 m, and 160 m measurement distances, the proposed method attained 0.24, 0.18, 0.25, and 0.11 MAE accuracies, respectively, in throughput data prediction, while the two standard methods attained 47.46, 49.93, 29.80, 53.92 and 47.61, 52.54, 53.43, 54.97, respectively. Overall, the GPR–LBFGS with the KPS method would find valuable applications in 5G and beyond 5 G wireless communication systems.
19
Song, Jinze, Yuhao Li, Shuai Liu, Youming Xiong, Weixin Pang, Yufa He, and Yaxi Mu. "Comparison of Machine Learning Algorithms for Sand Production Prediction: An Example for a Gas-Hydrate-Bearing Sand Case." Energies 15, no. 18 (September 6, 2022): 6509. http://dx.doi.org/10.3390/en15186509.
Abstract:
This paper demonstrates the applicability of machine learning algorithms in sand production problems with natural gas hydrate (NGH)-bearing sands, which have been regarded as a grave concern for commercialization. The sanding problem hinders the commercial exploration of NGH reservoirs. The common sand production prediction methods need assumptions for complicated mathematical derivations. The main contribution of this paper was to introduce machine learning into the prediction sand production by using data from laboratory experiments. Four main machine learning algorithms were selected, namely, K-Nearest Neighbor, Support Vector Regression, Boosting Tree, and Multi-Layer Perceptron. Training datasets for machine learning were collected from a sand production experiment. The experiment considered both the geological parameters and the sand control effect. The machine learning algorithms were mainly evaluated according to their mean absolute error and coefficient of determination. The evaluation results showed that the most accurate results under the given conditions were from the Boosting Tree algorithm, while the K-Nearest Neighbor had the worst prediction performance. Considering an ensemble prediction model, the Support Vector Regression and Multi-Layer Perceptron could also be applied for the prediction of sand production. The tuning process revealed that the Gaussian kernel was the proper kernel function for improving the prediction performance of SVR. In addition, the best parameters for both the Boosting Tree and Multi-Layer Perceptron were recommended for the accurate prediction of sand production. This paper also involved one case study to compare the prediction results of the machine learning models and classic numerical simulation, which showed the capability of machine learning of accurately predicting sand production, especially under stable pressure conditions.
20
Alamaniotis, Miltiadis, and Lefteri H. Tsoukalas. "Fusion of Gaussian Process Kernel Regressors for Fault Prediction in Intelligent Energy Systems." International Journal on Artificial Intelligence Tools 25, no. 04 (August 2016): 1650023. http://dx.doi.org/10.1142/s0218213016500238.
Abstract:
Integration of energy systems with machine intelligence technologies advances the new generation of intelligent energy systems. One feature of intelligent energy systems is their ability to predict a future fault state (prognosis), and thus support control actions. This paper introduces a prognostic framework based on concepts originating from the machine learning universe and proceeds to assess the performance of the prognostics algorithms with statistical methods aiming to formulate a linear predictor whose coefficients are the solution of a multi-objective optimization problem. Prediction is achieved through independent Gaussian process kernel regressors put together as terms of a linear forecaster. In this novel framework the available data is used to train the regression models whose degradation predictions cover a predetermined time period and have the form of a predictive distribution whose mean and variance values are computed for future moments. Thus, given the observed data points one may search for the most probable values of other quantities of interest, or the values at different points from those measured. The feasibility of the cascading prognostics methodology is demonstrated via a turbine blade degradation example. This implementation is characterized by advantages that include the utilization of an optimization process that simultaneously determines the lowest possible values in a set of different statistical measures and the employment of a set of kernels for modeling various data features and capturing the system dynamics.
21
Gbémou, Shab, Julien Eynard, Stéphane Thil, Emmanuel Guillot, and Stéphane Grieu. "A Comparative Study of Machine Learning-Based Methods for Global Horizontal Irradiance Forecasting." Energies 14, no. 11 (May 29, 2021): 3192. http://dx.doi.org/10.3390/en14113192.
Abstract:
The proliferation of photovoltaic (PV) power generation in power distribution grids induces increasing safety and service quality concerns for grid operators. The inherent variability, essentially due to meteorological conditions, of PV power generation affects the power grid reliability. In order to develop efficient monitoring and control schemes for distribution grids, reliable forecasting of the solar resource at several time horizons that are related to regulation, scheduling, dispatching, and unit commitment, is necessary. PV power generation forecasting can result from forecasting global horizontal irradiance (GHI), which is the total amount of shortwave radiation received from above by a surface horizontal to the ground. A comparative study of machine learning methods is given in this paper, with a focus on the most widely used: Gaussian process regression (GPR), support vector regression (SVR), and artificial neural networks (ANN). Two years of GHI data with a time step of 10 min are used to train the models and forecast GHI at varying time horizons, ranging from 10 min to 4 h. Persistence on the clear-sky index, also known as scaled persistence model, is included in this paper as a reference model. Three criteria are used for in-depth performance estimation: normalized root mean square error (nRMSE), dynamic mean absolute error (DMAE) and coverage width-based criterion (CWC). Results confirm that machine learning-based methods outperform the scaled persistence model. The best-performing machine learning-based methods included in this comparative study are the long short-term memory (LSTM) neural network and the GPR model using a rational quadratic kernel with automatic relevance determination.
22
Allam, Omar, Hyun-Myung Woo, Graham Brantley, Robert Kuramshin, Zlatomir Stoichev, Byung-Jun Yoon, and Seung Soon Jang. "Uncovering Molecular Structure – Redox Potential Relationships for Organic Electrode Materials: A Hybrid DFT – Machine Learning Approach." ECS Meeting Abstracts MA2022-02, no. 2 (October 9, 2022): 165. http://dx.doi.org/10.1149/ma2022-022165mtgabs.
Abstract:
Density functional theory and machine learning are used to investigate the structure-electrochemical performance relationships of organic moieties for use in Li-ion batteries. Namely, DFT calculations are performed to predict the redox potential of several novel organic molecules with an accuracy within ~0.1 V of experimental measurements. However, despite its ability to provide valuable insight regarding the electrochemical properties of novel organic molecules, our high efficacy DFT modeling protocol demands significant computational time and is therefore impractical for the vast screening of novel material candidate. As a result, we explore machine learning as a strategy for the accelerated discovery of novel organic materials. More critically, we use machine learning as a method for assessing the various structure-electrochemical relationships which can provide a more general guideline for the design of organic electrode materials. We are employing different learning models, including artificial neural networks, gradient-boosting regression, and kernel methods (such as kernel ridge regression and Gaussian process regression), via three different pipelines with varying sophistication with the aim of generating an advanced ML scheme for the accurate prediction and analysis of electrochemical activity. In addition to incorporating structural fingerprints, we are exploring an active learning framework, namely using the efficient global optimization scheme, to explore the materials space strategically using publicly available datasets. Through this approach, we can discover new electrode materials that could 1) have a higher probability for achieving enhanced electrochemical properties and 2) increase our learning model’s performance by increasing our dataset’s representation of the material space. Additionally, we are implementing a high-throughput virtual screening (HTVS) pipeline which consists of several surrogate learning models with increasing levels of fidelity. The material candidates are pruned at each stage to discard samples that are unlikely to possess a desirable redox potential for cathodic application, and the remaining samples move on to the next model. In this way, only a subset of the original dataset which has a higher probability for lying within the desired redox potential range is pursued using the most computationally expensive model.
23
Liu, Xiaobo, Su Yang, and Zhengxian Liu. "Predicting Fluid Intelligence via Naturalistic Functional Connectivity Using Weighted Ensemble Model and Network Analysis." NeuroSci 2, no. 4 (December 17, 2021): 427–42. http://dx.doi.org/10.3390/neurosci2040032.
Abstract:
Objectives: Functional connectivity triggered by naturalistic stimuli (e.g., movie clips), coupled with machine learning techniques provide great insight in exploring brain functions such as fluid intelligence. However, functional connectivity is multi-layered while traditional machine learning is based on individual model, which is not only limited in performance, but also fails to extract multi-dimensional and multi-layered information from the brain network. Methods: In this study, inspired by multi-layer brain network structure, we propose a new method, namely weighted ensemble model and network analysis, which combines machine learning and graph theory for improved fluid intelligence prediction. Firstly, functional connectivity analysis and graphical theory were jointly employed. The functional connectivity and graphical indices computed using the preprocessed fMRI data were then all fed into an auto-encoder parallelly for automatic feature extraction to predict the fluid intelligence. In order to improve the performance, tree regression and ridge regression models were stacked and fused automatically with weighted values. Finally, layers of auto-encoder were visualized to better illustrate the connectome patterns, followed by the evaluation of the performance to justify the mechanism of brain functions. Results: Our proposed method achieved the best performance with a 3.85 mean absolute deviation, 0.66 correlation coefficient and 0.42 R-squared coefficient; this model outperformed other state-of-the-art methods. It is also worth noting that the optimization of the biological pattern extraction was automated though the auto-encoder algorithm. Conclusion: The proposed method outperforms the state-of-the-art reports, also is able to effectively capture the biological patterns of functional connectivity during a naturalistic movie state for potential clinical explorations.
24
Liu, Hongyu, Fuheng Qu, Yong Yang, Wanting Li, and Zhonglin Hao. "Soybean Variety Identification Based on Improved ResNet18 Hyperspectral Image." Journal of Physics: Conference Series 2284, no. 1 (June 1, 2022): 012017. http://dx.doi.org/10.1088/1742-6596/2284/1/012017.
Abstract:
Abstract Aiming at the problems of insufficient feature extraction, slow speed and low accuracy of traditional machine learning methods, a soybean variety identification method based on improved ResNet18 hyperspectral image was proposed. This method extracts more effective detail features by decomposing the large convolution kernel, and changes the connection of residual structure and introduces the BN layer optimization network to make the feature extraction more sufficient. The perception of soybean hyperspectral image recognition is enhanced by adding multi-scale feature extraction module. The experimental results show that the recognition accuracy of this method reaches 97.36 %, which is higher than Nasnet large and Resnet18 models, and the robustness of the model is further enhanced, which can provide reference for soybean variety recognition.
25
Santosh, P. Reddy, and M. Chandra Sekhar. "An Efficient Novel Approach with Multi Class Label Classification through Machine Learning Models for Pancreatic Cancer." Scalable Computing: Practice and Experience 23, no. 4 (December 22, 2022): 193–210. http://dx.doi.org/10.12694/scpe.v23i4.2019.
Abstract:
Pancreatic cancer is right now the fourth largest cause of cancer-related deaths. Early diagnosis is one good solution for pancreatic cancer patients and reduces the mortality rate. Accurate and earlier diagnosis of the pancreatic tumor is a demanding task due to several factors such as delayed diagnosis and absence of early warning symptoms. The conventional distributed machine learning techniques such as SVM and logistic regression were not efficient to minimize the error rate and improve the classification of pancreatic cancer with higher accuracy. Therefore, a novel technique called Distributed Hybrid Elitism gene Quadratic discriminant Reinforced Learning Classifier System (DHEGQDRLCS) is developed in this paper. First, the number of data samples is collected from the repository dataset. This repository contains all the necessary files for the identification of prognostic biomarkers for pancreatic cancer. After the data collection, the separation of training and testing samples is performed for the accurate classification of pancreatic cancer samples. Then the training samples are considered and applied to Distributed Hybrid Elitism gene Quadratic discriminant Reinforced Learning Classifier System. The proposed hybrid classifier system uses the Kernel Quadratic Discriminant Function to analyze the training samples. After that, the Elitism gradient gene optimization is applied for classifying the samples into multiple classes such as non-cancerous pancreas, benign hepatobiliary disease i.e., pancreatic cancer, and Pancreatic ductal adenocarcinoma. Then the Reinforced Learning technique is applied to minimize the loss function based on target classification results and predicted classification results. Finally, the hybridized approach improves pancreatic cancer diagnosing accuracy. Experimental evaluation is carried out with pancreatic cancer dataset with Hadoop distributed system and different quantitative metrics such as Accuracy, balanced accuracy, F1-score, precision, recall, specificity, TN, TP, FN, FP, ROC_AUC, PRC_AUC, and PRC_APS. The performance analysis results indicate that the DHEGQDRLCS provides better diagnosing accuracy when compared to existing methods.
26
Zheng, Jianguo, Yilin Wang, Shihan Li, and Hancong Chen. "The Stock Index Prediction Based on SVR Model with Bat Optimization Algorithm." Algorithms 14, no. 10 (October 15, 2021): 299. http://dx.doi.org/10.3390/a14100299.
Abstract:
Accurate stock market prediction models can provide investors with convenient tools to make better data-based decisions and judgments. Moreover, retail investors and institutional investors could reduce their investment risk by selecting the optimal stock index with the help of these models. Predicting stock index price is one of the most effective tools for risk management and portfolio diversification. The continuous improvement of the accuracy of stock index price forecasts can promote the improvement and maturity of China’s capital market supervision and investment. It is also an important guarantee for China to further accelerate structural reforms and manufacturing transformation and upgrading. In response to this problem, this paper introduces the bat algorithm to optimize the three free parameters of the SVR machine learning model, constructs the BA-SVR hybrid model, and forecasts the closing prices of 18 stock indexes in Chinese stock market. The total sample comes from 15 January 2016 (the 10th trading day in 2016) to 31 December 2020. We select the last 20, 60, and 250 days of whole sample data as test sets for short-term, mid-term, and long-term forecast, respectively. The empirical results show that the BA-SVR model outperforms the polynomial kernel SVR model and sigmoid kernel SVR model without optimized initial parameters. In the robustness test part, we use the stationary time series data after the first-order difference of six selected characteristics to re-predict. Compared with the random forest model and ANN model, the prediction performance of the BA-SVR model is still significant. This paper also provides a new perspective on the methods of stock index forecasting and the application of bat algorithms in the financial field.
27
MÀRQUEZ, LLUÍS, and ALESSANDRO MOSCHITTI. "Special issue on statistical learning of natural language structured input and output." Natural Language Engineering 18, no. 2 (March 14, 2012): 147–53. http://dx.doi.org/10.1017/s135132491200006x.
Abstract:
AbstractDuring last decade, machine learning and, in particular, statistical approaches have become more and more important for research in Natural Language Processing (NLP) and Computational Linguistics. Nowadays, most stakeholders of the field use machine learning, as it can significantly enhance both system design and performance. However, machine learning requires careful parameter tuning and feature engineering for representing language phenomena. The latter becomes more complex when the system input/output data is structured, since the designer has both to (i) engineer features for representing structure and model interdependent layers of information, which is usually a non-trivial task; and (ii) generate a structured output using classifiers, which, in their original form, were developed only for classification or regression. Research in empirical NLP has been tackling this problem by constructing output structures as a combination of the predictions of independent local classifiers, eventually applying post-processing heuristics to correct incompatible outputs by enforcing global properties. More recently, some advances of the statistical learning theory, namely structured output spaces and kernel methods, have brought techniques for directly encoding dependencies between data items in a learning algorithm that performs global optimization. Within this framework, this special issue aims at studying, comparing, and reconciling the typical domain/task-specific NLP approaches to structured data with the most advanced machine learning methods. In particular, the selected papers analyze the use of diverse structured input/output approaches, ranging from re-ranking to joint constraint-based global models, for diverse natural language tasks, i.e., document ranking, syntactic parsing, sequence supertagging, and relation extraction between terms and entities. Overall, the experience with this special issue shows that, although a definitive unifying theory for encoding and generating structured information in NLP applications is still far from being shaped, some interesting and effective best practice can be defined to guide practitioners in modeling their own natural language application on complex data.
28
Piccolo, Stephen R., Avery Mecham, Nathan P. Golightly, Jérémie L. Johnson, and Dustin B. Miller. "The ability to classify patients based on gene-expression data varies by algorithm and performance metric." PLOS Computational Biology 18, no. 3 (March 11, 2022): e1009926. http://dx.doi.org/10.1371/journal.pcbi.1009926.
Abstract:
By classifying patients into subgroups, clinicians can provide more effective care than using a uniform approach for all patients. Such subgroups might include patients with a particular disease subtype, patients with a good (or poor) prognosis, or patients most (or least) likely to respond to a particular therapy. Transcriptomic measurements reflect the downstream effects of genomic and epigenomic variations. However, high-throughput technologies generate thousands of measurements per patient, and complex dependencies exist among genes, so it may be infeasible to classify patients using traditional statistical models. Machine-learning classification algorithms can help with this problem. However, hundreds of classification algorithms exist—and most support diverse hyperparameters—so it is difficult for researchers to know which are optimal for gene-expression biomarkers. We performed a benchmark comparison, applying 52 classification algorithms to 50 gene-expression datasets (143 class variables). We evaluated algorithms that represent diverse machine-learning methodologies and have been implemented in general-purpose, open-source, machine-learning libraries. When available, we combined clinical predictors with gene-expression data. Additionally, we evaluated the effects of performing hyperparameter optimization and feature selection using nested cross validation. Kernel- and ensemble-based algorithms consistently outperformed other types of classification algorithms; however, even the top-performing algorithms performed poorly in some cases. Hyperparameter optimization and feature selection typically improved predictive performance, and univariate feature-selection algorithms typically outperformed more sophisticated methods. Together, our findings illustrate that algorithm performance varies considerably when other factors are held constant and thus that algorithm selection is a critical step in biomarker studies.
29
Debjit, Kumar, Md Saiful Islam, Md Abadur Rahman, Farhana Tazmim Pinki, Rajan Dev Nath, Saad Al-Ahmadi, Md Shahadat Hossain, Khondoker Mirazul Mumenin, and Md Abdul Awal. "An Improved Machine-Learning Approach for COVID-19 Prediction Using Harris Hawks Optimization and Feature Analysis Using SHAP." Diagnostics 12, no. 5 (April 19, 2022): 1023. http://dx.doi.org/10.3390/diagnostics12051023.
Abstract:
A healthcare monitoring system needs the support of recent technologies such as artificial intelligence (AI), machine learning (ML), and big data, especially during the COVID-19 pandemic. This global pandemic has already taken millions of lives. Both infected and uninfected people have generated big data where AI and ML can use to combat and detect COVID-19 at an early stage. Motivated by this, an improved ML framework for the early detection of this disease is proposed in this paper. The state-of-the-art Harris hawks optimization (HHO) algorithm with an improved objective function is proposed and applied to optimize the hyperparameters of the ML algorithms, namely HHO-based eXtreme gradient boosting (HHOXGB), light gradient boosting (HHOLGB), categorical boosting (HHOCAT), random forest (HHORF) and support vector classifier (HHOSVC). An ensemble technique was applied to these optimized ML models to improve the prediction performance. Our proposed method was applied to publicly available big COVID-19 data and yielded a prediction accuracy of 92.38% using the ensemble model. In contrast, HHOXGB provided the highest accuracy of 92.23% as a single optimized model. The performance of the proposed method was compared with the traditional algorithms and other ML-based methods. In both cases, our proposed method performed better. Furthermore, not only the classification improvement, but also the features are analyzed in terms of feature importance calculated by SHapely adaptive exPlanations (SHAP) values. A graphical user interface is also discussed as a potential tool for nonspecialist users such as clinical staff and nurses. The processed data, trained model, and codes related to this study are available at GitHub.
30
An, Feng-Ping. "Medical Image Classification Algorithm Based on Weight Initialization-Sliding Window Fusion Convolutional Neural Network." Complexity 2019 (October 20, 2019): 1–15. http://dx.doi.org/10.1155/2019/9151670.
Abstract:
Due to the complexity of medical images, traditional medical image classification methods have been unable to meet actual application needs. In recent years, the rapid development of deep learning theory has provided a technical approach for solving medical image classification tasks. However, deep learning has the following problems in medical image classification. First, it is impossible to construct a deep learning model hierarchy for medical image properties; second, the network initialization weights of deep learning models are not well optimized. Therefore, this paper starts from the perspective of network optimization and improves the nonlinear modeling ability of the network through optimization methods. A new network weight initialization method is proposed, which alleviates the problem that existing deep learning model initialization is limited by the type of the nonlinear unit adopted and increases the potential of the neural network to handle different visual tasks. Moreover, through an in-depth study of the multicolumn convolutional neural network framework, this paper finds that the number of features and the convolution kernel size at different levels of the convolutional neural network are different. In contrast, the proposed method can construct different convolutional neural network models that adapt better to the characteristics of the medical images of interest and thus can better train the resulting heterogeneous multicolumn convolutional neural networks. Finally, using the adaptive sliding window fusion mechanism proposed in this paper, both methods jointly complete the classification task of medical images. Based on the above ideas, this paper proposes a medical classification algorithm based on a weight initialization/sliding window fusion for multilevel convolutional neural networks. The methods proposed in this study were applied to breast mass, brain tumor tissue, and medical image database classification experiments. The results show that the proposed method not only achieves a higher average accuracy than that of traditional machine learning and other deep learning methods but also is more stable and more robust.
31
Mehra, Saanvi, Binoy Shah, Ankur Sethi, Ratna Puri, and Somashekhar Nimbalkar. "Down Syndrome Detection Through Graphical Analysis of Facial Dysmorphic Features in Newborn Children With Ethnicity/Racial Slicing: An AI/ML-Based Approach." Journal of Neonatology 36, no. 3 (September 2022): 199–205. http://dx.doi.org/10.1177/09732179221113677.
Abstract:
Background Down syndrome (DS) is associated with high mortality in India, due to nondiagnosis/late-diagnosis caused by unavailability of qualified doctors and/or lack of access to expensive medical/diagnostic facilities, especially in rural India. Using artificial intelligence/machine learning graphical pattern recognition tools, relevant facial points can be extracted from children’s photographs, facial anomalies can be identified, and probability of DS affliction can be predicted. Methods: Trained Google’s Cloud Vision AutoML Image Classification model was employed with ~2,000 photographs of DS positive children and ~3,000 photographs of DS negative children. A subset of 300 images, 100 each of Asian, Caucasian, and Other-Race children, was used to train and test 3 race-specific models. These results were compared against a unified model trained and tested with same 300 images. Results: The CloudML model trained with ~5,000 images initially achieved: Sensitivity—94.6%, specificity—96.9%, and accuracy—96.0%. Upon optimizing confidence threshold to 0.1, model maximized sensitivity at 99.6%, specificity dropped to 93.8%, and accuracy maintained at 96.0%. Each of the race-specific models trained with 100 images each, after optimization, yielded perfect scores on sensitivity, specificity, and accuracy of 100% each. Against this, the unified model with 300 images yielded overall accuracy of 98% (100% sensitivity, 83% specificity for Caucasian children, and 100% sensitivity, 100% specificity for Asian/Other children). Conclusions: Post optimization, this model can be used as an effective postnatal screening tool for DS detection. Preliminary results indicate that race-specific models can achieve even higher accuracy, sensitivity, and specificity.
32
Zeng, Tan, Matsunaga, and Shirai. "Generalization of Parameter Selection of SVM and LS-SVM for Regression." Machine Learning and Knowledge Extraction 1, no. 2 (June 19, 2019): 745–55. http://dx.doi.org/10.3390/make1020043.
Abstract:
A Support Vector Machine (SVM) for regression is a popular machine learning model that aims to solve nonlinear function approximation problems wherein explicit model equations are difficult to formulate. The performance of an SVM depends largely on the selection of its parameters. Choosing between an SVM that solves an optimization problem with inequality constrains and one that solves the least square of errors (LS-SVM) adds to the complexity. Various methods have been proposed for tuning parameters, but no article puts the SVM and LS-SVM side by side to discuss the issue using a large dataset from the real world, which could be problematic for existing parameter tuning methods. We investigated both the SVM and LS-SVM with an artificial dataset and a dataset of more than 200,000 points used for the reconstruction of the global surface ocean CO2 concentration. The results reveal that: (1) the two models are most sensitive to the parameter of the kernel function, which lies in a narrow range for scaled input data; (2) the optimal values of other parameters do not change much for different datasets; and (3) the LS-SVM performs better than the SVM in general. The LS-SVM is recommended, as it has less parameters to be tuned and yields a smaller bias. Nevertheless, the SVM has advantages of consuming less computer resources and taking less time to train. The results suggest initial parameter guesses for using the models.
33
Lu, Juan, Xiaoping Liao, Steven Li, Haibin Ouyang, Kai Chen, and Bing Huang. "An Effective ABC-SVM Approach for Surface Roughness Prediction in Manufacturing Processes." Complexity 2019 (June 13, 2019): 1–13. http://dx.doi.org/10.1155/2019/3094670.
Abstract:
It is difficult to accurately predict the response of some stochastic and complicated manufacturing processes. Data-driven learning methods which can mine unseen relationship between influence parameters and outputs are regarded as an effective solution. In this study, support vector machine (SVM) is applied to develop prediction models for machining processes. Kernel function and loss function are Gaussian radial basis function and ε-insensitive loss function, respectively. To improve the prediction accuracy and reduce parameter adjustment time of SVM model, artificial bee colony algorithm (ABC) is employed to optimize internal parameters of SVM model. Further, to evaluate the optimization performance of ABC in parameters determination of SVM, this study compares the prediction performance of SVM models optimized by well-known evolutionary and swarm-based algorithms (differential evolution (DE), genetic algorithm (GA), particle swarm optimization (PSO), and ABC) and analyzes ability of these optimization algorithms from their optimization mechanism and convergence speed based on experimental datasets of turning and milling. Experimental results indicate that the selected four evaluation indicators values that reflect prediction accuracy and adjustment time for ABC-SVM are better than DE-SVM, GA-SVM, and PSO-SVM except three indicator values of DE-SVM for AISI 1045 steel under the case that training set is enough to develop the prediction model. ABC algorithm has less control parameters, faster convergence speed, and stronger searching ability than DE, GA, and PSO algorithms for optimizing the internal parameters of SVM model. These results shed light on choosing a satisfactory optimization algorithm of SVM for manufacturing processes.
34
Shakhovska, Natalya, Vitaliy Yakovyna, and Valentyna Chopyak. "A new hybrid ensemble machine-learning model for severity risk assessment and post-COVID prediction system." Mathematical Biosciences and Engineering 19, no. 6 (2022): 6102–23. http://dx.doi.org/10.3934/mbe.2022285.
Abstract:
<abstract> <p>Starting from December 2019, the COVID-19 pandemic has globally strained medical resources and caused significant mortality. It is commonly recognized that the severity of SARS-CoV-2 disease depends on both the comorbidity and the state of the patient's immune system, which is reflected in several biomarkers. The development of early diagnosis and disease severity prediction methods can reduce the burden on the health care system and increase the effectiveness of treatment and rehabilitation of patients with severe cases. This study aims to develop and validate an ensemble machine-learning model based on clinical and immunological features for severity risk assessment and post-COVID rehabilitation duration for SARS-CoV-2 patients. The dataset consisting of 35 features and 122 instances was collected from Lviv regional rehabilitation center. The dataset contains age, gender, weight, height, BMI, CAT, 6-minute walking test, pulse, external respiration function, oxygen saturation, and 15 immunological markers used to predict the relationship between disease duration and biomarkers using the machine learning approach. The predictions are assessed through an area under the receiver-operating curve, classification accuracy, precision, recall, and F1 score performance metrics. A new hybrid ensemble feature selection model for a post-COVID prediction system is proposed as an automatic feature cut-off rank identifier. A three-layer high accuracy stacking ensemble classification model for intelligent analysis of short medical datasets is presented. Together with weak predictors, the associative rules allowed improving the classification quality. The proposed ensemble allows using a random forest model as an aggregator for weak repressors' results generalization. The performance of the three-layer stacking ensemble classification model (AUC 0.978; CA 0.920; F1 score 0.921; precision 0.924; recall 0.920) was higher than five machine learning models, viz. tree algorithm with forward pruning; Naïve Bayes classifier; support vector machine with RBF kernel; logistic regression, and a calibrated learner with sigmoid function and decision threshold optimization. Aging-related biomarkers, viz. CD3+, CD4+, CD8+, CD22+ were examined to predict post-COVID rehabilitation duration. The best accuracy was reached in the case of the support vector machine with the linear kernel (MAPE = 0.0787) and random forest classifier (RMSE = 1.822). The proposed three-layer stacking ensemble classification model predicted SARS-CoV-2 disease severity based on the cytokines and physiological biomarkers. The results point out that changes in studied biomarkers associated with the severity of the disease can be used to monitor the severity and forecast the rehabilitation duration.</p> </abstract>
35
Chen, Yong, Peng Li, Wenping Ren, Xin Shen, and Min Cao. "Field data–driven online prediction model for icing load on power transmission lines." Measurement and Control 53, no. 1-2 (November 22, 2019): 126–40. http://dx.doi.org/10.1177/0020294019878872.
Abstract:
Methods for the accurate prediction of icing loads in overhead transmission lines have become an important research topic for electrical power systems as they are necessary for ensuring the safety and stability of power-grid operations. Current machine learning models for the prediction of icing loads on transmission lines are afflicted by the following issues: insufficient prediction accuracy, high randomity in the selection of kernel functions and model parameters, and a lack of generalizability. To address these issues, we propose a field data–driven online prediction model for icing loads on transmission lines. First, the effects of the type of kernel function used in the support vector regression algorithm on the prediction accuracy of the model were analyzed using micrometeorological data and icing data collected by on-site monitoring systems. The particle swarm optimization algorithm was then used to optimize and determine the model parameters such as penalty coefficients. An offline support vector regression prediction model was thus constructed. Using the accurate online support vector regression algorithm, the weighting coefficients of the samples were dynamically adjusted to satisfy the Karush–Kuhn–Tucker conditions, which allowed online updates to be made to the regression function and prediction model. Finally, a simulation analysis was performed using actual icing incidents that occurred in a transmission line of the Yunnan Power Grid, which demonstrated that our model can make online predictions for the icing load on transmission lines in actual applications. Our model proved to be superior to conventional icing-load prediction models with regard to the single-step and multi-step prediction accuracies and generalizability. Hence, our prediction model will improve the decision-making processes regarding the deicing and maintenance of power transmission and transformation systems.
36
Meaney, Christopher, Michael Escobar, Therese A. Stukel, Peter C. Austin, and Liisa Jaakkimainen. "Comparison of Methods for Estimating Temporal Topic Models From Primary Care Clinical Text Data: Retrospective Closed Cohort Study." JMIR Medical Informatics 10, no. 12 (December 19, 2022): e40102. http://dx.doi.org/10.2196/40102.
Abstract:
Background Health care organizations are collecting increasing volumes of clinical text data. Topic models are a class of unsupervised machine learning algorithms for discovering latent thematic patterns in these large unstructured document collections. Objective We aimed to comparatively evaluate several methods for estimating temporal topic models using clinical notes obtained from primary care electronic medical records from Ontario, Canada. Methods We used a retrospective closed cohort design. The study spanned from January 01, 2011, through December 31, 2015, discretized into 20 quarterly periods. Patients were included in the study if they generated at least 1 primary care clinical note in each of the 20 quarterly periods. These patients represented a unique cohort of individuals engaging in high-frequency use of the primary care system. The following temporal topic modeling algorithms were fitted to the clinical note corpus: nonnegative matrix factorization, latent Dirichlet allocation, the structural topic model, and the BERTopic model. Results Temporal topic models consistently identified latent topical patterns in the clinical note corpus. The learned topical bases identified meaningful activities conducted by the primary health care system. Latent topics displaying near-constant temporal dynamics were consistently estimated across models (eg, pain, hypertension, diabetes, sleep, mood, anxiety, and depression). Several topics displayed predictable seasonal patterns over the study period (eg, respiratory disease and influenza immunization programs). Conclusions Nonnegative matrix factorization, latent Dirichlet allocation, structural topic model, and BERTopic are based on different underlying statistical frameworks (eg, linear algebra and optimization, Bayesian graphical models, and neural embeddings), require tuning unique hyperparameters (optimizers, priors, etc), and have distinct computational requirements (data structures, computational hardware, etc). Despite the heterogeneity in statistical methodology, the learned latent topical summarizations and their temporal evolution over the study period were consistently estimated. Temporal topic models represent an interesting class of models for characterizing and monitoring the primary health care system.
37
Paidipati, Kiran Kumar, Christophe Chesneau, B. M. Nayana, Kolla Rohith Kumar, Kalpana Polisetty, and Chinnarao Kurangi. "Prediction of Rice Cultivation in India—Support Vector Regression Approach with Various Kernels for Non-Linear Patterns." AgriEngineering 3, no. 2 (April 7, 2021): 182–98. http://dx.doi.org/10.3390/agriengineering3020012.
Abstract:
The prediction of rice yields plays a major role in reducing food security problems in India and also suggests that government agencies manage the over or under situations of production. Advanced machine learning techniques are playing a vital role in the accurate prediction of rice yields in dealing with nonlinear complex situations instead of traditional statistical methods. In the present study, the researchers made an attempt to predict the rice yield through support vector regression (SVR) models with various kernels (linear, polynomial, and radial basis function) for India overall and the top five rice producing states by considering influence parameters, such as the area under cultivation and production, as independent variables for the years 1962–2018. The best-fitted models were chosen based on the cross-validation and hyperparameter optimization of various kernel parameters. The root-mean-square error (RMSE) and mean absolute error (MAE) were calculated for the training and testing datasets. The results revealed that SVR with various kernels fitted to India overall, as well as the major rice producing states, would explore the nonlinear patterns to understand the precise situations of yield prediction. This study will be helpful for farmers as well as the central and state governments for estimating rice yield in advance with optimal resources.
38
Natalia, Friska, Julio Christian Young, Nunik Afriliana, Hira Meidia, Reyhan Eddy Yunus, and Sud Sudirman. "Automated selection of mid-height intervertebral disc slice in traverse lumbar spine MRI using a combination of deep learning feature and machine learning classifier." PLOS ONE 17, no. 1 (January 13, 2022): e0261659. http://dx.doi.org/10.1371/journal.pone.0261659.
Abstract:
Abnormalities and defects that can cause lumbar spinal stenosis often occur in the Intervertebral Disc (IVD) of the patient’s lumbar spine. Their automatic detection and classification require an application of an image analysis algorithm on suitable images, such as mid-sagittal images or traverse mid-height intervertebral disc slices, as inputs. Hence the process of selecting and separating these images from other medical images in the patient’s set of scans is necessary. However, the technological progress in making this process automated is still lagging behind other areas in medical image classification research. In this paper, we report the result of our investigation on the suitability and performance of different approaches of machine learning to automatically select the best traverse plane that cuts closest to the half-height of an IVD from a database of lumbar spine MRI images. This study considers images features extracted using eleven different pre-trained Deep Convolution Neural Network (DCNN) models. We investigate the effectiveness of three dimensionality-reduction techniques and three feature-selection techniques on the classification performance. We also investigate the performance of five different Machine Learning (ML) algorithms and three Fully Connected (FC) neural network learning optimizers which are used to train an image classifier with hyperparameter optimization using a wide range of hyperparameter options and values. The different combinations of methods are tested on a publicly available lumbar spine MRI dataset consisting of MRI studies of 515 patients with symptomatic back pain. Our experiment shows that applying the Support Vector Machine algorithm with a short Gaussian kernel on full-length image features extracted using a pre-trained DenseNet201 model is the best approach to use. This approach gives the minimum per-class classification performance of around 0.88 when measured using the precision and recall metrics. The median performance measured using the precision metric ranges from 0.95 to 0.99 whereas that using the recall metric ranges from 0.93 to 1.0. When only considering the L3/L4, L4/L5, and L5/S1 classes, the minimum F1-Scores range between 0.93 to 0.95, whereas the median F1-Scores range between 0.97 to 0.99.
39
Alrobaie, Abdurahman, and Moncef Krarti. "A Review of Data-Driven Approaches for Measurement and Verification Analysis of Building Energy Retrofits." Energies 15, no. 21 (October 22, 2022): 7824. http://dx.doi.org/10.3390/en15217824.
Abstract:
Although the energy and cost benefits for retrofitting existing buildings are promising, several challenges remain for accurate measurement and verification (M&V) analysis to estimate these benefits. Due to the rapid development in advanced metering infrastructure (AMI), data-driven approaches are becoming more effective than deterministic methods in developing baseline energy models for existing buildings using historical energy consumption data. The literature review presented in this paper provides an extensive summary of data-driven approaches suitable for building energy consumption prediction needed for M&V applications. The presented literature review describes commonly used data-driven modeling approaches including linear regressions, decision trees, ensemble methods, support vector machine, deep learning, and kernel regressions. The advantages and limitations of each data-driven modeling approach and its variants are discussed, including their cited applications. Additionally, feature engineering methods used in building energy data-driven modeling are outlined and described based on reported case studies to outline commonly used building features as well as selection and processing techniques of the most relevant features. This review highlights the gap between the listed existing frameworks and recently reported case studies using data-driven models. As a conclusion, this review demonstrates the need for a flexible M&V analysis framework to identify the best data-driven methods and their associated features depending on the building type and retrofit measures.
40
Zou, Hanting, Shuai Shen, Tianmeng Lan, Xufeng Sheng, Jiezhong Zan, Yongwen Jiang, Qizhen Du, and Haibo Yuan. "Prediction Method of the Moisture Content of Black Tea during Processing Based on the Miniaturized Near-Infrared Spectrometer." Horticulturae 8, no. 12 (December 9, 2022): 1170. http://dx.doi.org/10.3390/horticulturae8121170.
Abstract:
The moisture content of black tea is an important factor affecting its suitability for processing and forming the unique flavor. At present, the research on the moisture content of black tea mainly focuses on a single withering step, but the research on the rapid detection method of moisture content of black tea applicable to the entire processing stage is ignored. This study is based on a miniaturized near-infrared spectrometer(micro−NIRS) and establishes the prediction models for black tea moisture content through machine learning algorithms. We use micro−NIRS for spectroscopic data acquisition of samples. Linear partial least squares (PLS) and nonlinear support vector regression (SVR) were combined with four spectral pre−processing methods, and principal component analysis (PCA) was applied to establish the predictive models. In addition, we combine the gray wolf optimization algorithm (GWO) with SVR for the prediction of moisture content, aiming to establish the best prediction model of black tea moisture content by optimizing the selection of key parameters (c and g) of the kernel function in SVR. The results show that SNV, as a method to correct the error of the spectrum due to scattering, can effectively extract spectral features after combining with PCA and is better than other pre−processing methods. In contrast, the nonlinear SVR model outperforms the PLS model, and the established mixed model SNV−PCA−GWO−SVR achieves the best prediction effect. The correlation coefficient of the prediction set and the root mean square error of the prediction set are 0.9892 and 0.0362, respectively, and the relative deviation is 6.5001. Experimental data show that the moisture content of black tea can be accurately and effectively determined by micro-near-infrared spectroscopy.
41
Chaudhuri, Avijit Kumar, Dilip K. Banerjee, and Anirban Das. "A Dataset Centric Feature Selection and Stacked Model to Detect Breast Cancer." International Journal of Intelligent Systems and Applications 13, no. 4 (August 8, 2021): 24–37. http://dx.doi.org/10.5815/ijisa.2021.04.03.
Abstract:
World Health Organisation declared breast cancer (BC) as the most frequent suffering among women and accounted for 15 percent of all cancer deaths. Its accurate prediction is of utmost significance as it not only prevents deaths but also stops mistreatments. The conventional way of diagnosis includes the estimation of the tumor size as a sign of plausible cancer. Machine learning (ML) techniques have shown the effectiveness of predicting disease. However, the ML methods have been method centric rather than being dataset centric. In this paper, the authors introduce a dataset centric approach(DCA) deploying a genetic algorithm (GA) method to identify the features and a learning ensemble classifier algorithm to predict using the right features. Adaboost is such an approach that trains the model assigning weights to individual records rather than experimenting on the splitting of datasets alone and perform hyper-parameter optimization. The authors simulate the results by varying base classifiers i.e, using logistic regression (LR), decision tree (DT), support vector machine (SVM), naive bayes (NB), random forest (RF), and 10-fold cross-validations with a different split of the dataset as training and testing. The proposed DCA model with RF and 10-fold cross-validations demonstrated its potential with almost 100% performance in the classification results that no research could suggest so far. The DCA satisfies the underlying principles of data mining: the principle of parsimony, the principle of inclusion, the principle of discrimination, and the principle of optimality. This DCA is a democratic and unbiased ensemble approach as it allows all features and methods in the start to compete, but filters out the most reliable chain (of steps and combinations) that give the highest accuracy. With fewer characteristics and splits of 50-50, 66-34, and 10 fold cross-validations, the Stacked model achieves 97 % accuracy. These values and the reduction of features improve upon prior research works. Further, the proposed classifier is compared with some state-of-the-art machine-learning classifiers, namely random forest, naive Bayes, support-vector machine with radial basis function kernel, and decision tree. For testing the classifiers, different performance metrics have been employed – accuracy, detection rate, sensitivity, specificity, receiver operating characteristic, area under the curve, and some statistical tests such as the Wilcoxon signed-rank test and kappa statistics – to check the strength of the proposed DCA classifier. Various splits of training and testing data – namely, 50–50%, 66–34%, 80–20% and 10-fold cross-validation – have been incorporated in this research to test the credibility of the classification models in handling the unbalanced data. Finally, the proposed DCA model demonstrated its potential with almost 100% performance in the classification results. The output results have also been compared with other research on the same dataset where the proposed classifiers were found to be best across all the performance dimensions.
42
Devi, Munisamy Shyamala, Venkatesan Dhilip Kumar, Adrian Brezulianu, Oana Geman, and Muhammad Arif. "A Novel Blunge Calibration Intelligent Feature Classification Model for the Prediction of Hypothyroid Disease." Sensors 23, no. 3 (January 18, 2023): 1128. http://dx.doi.org/10.3390/s23031128.
Abstract:
According to the Indian health line report, 12% of the population suffer from abnormal thyroid functioning. The major challenge in this disease is that the existence of hypothyroid may not propagate any noticeable symptoms in its early stages. However, delayed treatment of this disease may lead to several other health problems, such as fertility issues and obesity. Therefore, early treatment is essential for patient survival. The proposed technology could be used for the prediction of hypothyroid disease and its severity during its early stages. Though several classification and regression algorithms are available for the prediction of hypothyroid using clinical information, there exists a gap in knowledge as to whether predicted outcomes may reach a higher accuracy or not. Therefore, the objective of this research is to predict the existence of hypothyroidism with higher accuracy by optimizing the estimator list of the pycaret classifier model. With this overview, a blunge calibration intelligent feature classification model that supports the assessment of the presence of hypothyroidism with high accuracy is proposed. A hypothyroidism dataset containing 3163 patient details with 23 independent and one dependent feature from the University of California Irvine (UCI) machine-learning repository was used for this work. We undertook dataset preprocessing and determined its incomplete values. Exploratory data analysis was performed to analyze all the clinical parameters and the extent to which each feature supports the prediction of hypothyroidism. ANOVA was used to verify the F-statistic values of all attributes that might highly influence the target. Then, hypothyroidism was predicted using various classifier algorithms, and the performance metrics were analyzed. The original dataset was subjected to dimensionality reduction by using regressor and classifier feature-selection algorithms to determine the best subset components for predicting hypothyroidism. The feature-selected subset of the clinical parameters was subjected to various classifier algorithms, and its performance was analyzed. The system was implemented with python in the Spyder editor of Anaconda Navigator IDE. Investigational results show that the Gaussian naive Bayes, AdaBoost classifier, and Ridge classifier maintained the accuracy of 89.5% for the regressor feature-selection methods. The blunge calibration regression model (BCRM) was designed with naive Bayes, AdaBoost, and Ridge as the estimators with accuracy optimization and with soft blending based on the sum of predicted probabilities of classifiers. The proposed BCRM showed 99.5% accuracy in predicting hypothyroidism. The implementation results show that the Kernel SVM, KNeighbor, and Ridge classifier maintained an accuracy of 87.5% for the classifier feature-selection methods. The blunge calibration classifier model (BCCM) was developed with Kernel SVM, KNeighbor, and Ridge as the estimators, with accuracy optimization and with soft blending based on the sum of predicted probabilities of classifiers. The proposed BCCM showed 99.7% accuracy in predicting hypothyroidism. The main contribution of this research is the design of BCCM and BCRM models that were built with accuracy optimization with soft blending based on the sum of predicted probabilities of classifiers. The BCRM and BCCM models uniqueness’s are achieved by updating the estimators list with the effective classifiers and regressors that suit the application at runtime.
43
Зеленчук, Н. А., and О. К. Альсова. "DESIGN AND IMPLEMENTATION OF AGRICULTURAL CLASSIFICATION SOFTWARE." Южно-Сибирский научный вестник, no. 1(41) (February 28, 2022): 51–59. http://dx.doi.org/10.25699/sssb.2022.41.1.008.
Abstract:
В настоящее время в сельскохозяйственной отрасли наблюдается постоянное увеличение объемов получаемых данных, возрастает потребность в их качественной обработке и точных расчетах для принятия обоснованных решений. Поэтому особую актуальность приобретают задачи, связанные с разработкой алгоритмов, методов и программного обеспечения для решения задач анализа и обработки данных в области сельского хозяйства с применением современных технологий и программных средств.В статье представлены результаты проектирования и реализации программного обеспечения (ПО) для решения задачи классификации сельскохозяйственных показателей на основе применения комплекса методов интеллектуального анализа данных и машинного обучения. В рамках проектной части работы описаны функциональные и нефункциональные требования к программному обеспечению, архитектура и структура проектируемой программы, технологии и программные средства реализации. Предложена укрупненная архитектура ПО, состоящая из двух частей: пользовательского приложения на языке программирования Java и ядра выполнения R-скриптов. В результате проектирования выделено пять модулей в структуре ПО: средства взаимодействия с данными, первичная обработка данных, классификация данных, автоматический подбор параметров алгоритмов и «интеллектуальный» модуль. В качестве средств реализации ПО предложено использовать стек технологий, а именно: язык статистических вычислений R для реализации методов анализа данных и язык Java для разработки графического пользовательского интерфейса для доступа к функциям анализа данных R.Также в статье приведено описание двух разработанных модулей программного обеспечения, а именно: модуля первичной обработки данных и модуля классификации данных. В модуле первичной обработки данных реализованы расчет основных числовых характеристик показателей, исследование законов распределения показателей на основе применения критериев согласия Шапиро-Уилка, Андерсона-Дарлинга, Крамера-фон Мизеса, Лиллиефорса, исследование взаимосвязей в данных с помощью методов корреляционного и дисперсионного анализов данных. В модуле классификации реализованы методы сэмплирования для решения проблемы несбалансированности данных, а также модели классификаторов: логистическая регрессия,наивный Байес, дискриминантный анализ, нейросетевой метод (персептрон), деревья решений, реализована возможность оценки точности получаемых моделей с помощью набора метрик. Приведен пример решения задачи классификации уровня засоренности участка с помощью нейронной сети (персептрона), точность классификации составила на тестовой выборке 0,73. The agricultural industry is currently experiencing a constant increase in the data obtained, the need for their quality processing and accurate calculations to support decision-making is increasing. Hence, the tasks related to the development of algorithms, methods and software for solving problems of analysis and processing of data in the field of agriculture using modern technologies and software are of particular relevance.The research paper provides the results of design and further implementation of software for agricultural indicators classification problem solving based on the complex application of data mining and machine learning methods. In the framework of the design part the functional and non-functional software requirements, the architecture and structure of the designed software, implementation technologies, and developing tools were included. The proposed large-scale software architecture consists of two parts: a user application based on the Java programming language and a kernel of R-scripts execution. The software design was defined to consist of five modules: data interaction tools, primary data processing, data analysis, automated selection of algorithm parameters, and «intelligent» module. To implement the software, it was proposed to use the technology stack: statistical computing language R for the realization of data analysis methods and Java to develop a graphical user interface to access the R data analysis functions.Another section provides a description of two developed software modules, namely: the module of primary data processing and the module of data classification. The module of primary data processing involves calculation of the main numerical features, the examination of the distribution laws based on the application of the Shapiro-Wilk, Anderson-Darling, Cramér-von Mises, Lilliefors consent criteria and tests, the analysis of relationships in the data using methods of correlation and variance analyses. The module of classification implemented methods of sampling to solve the problem of unbalanced data as well as models of classifiers: logistic regression, naive Bayes, discriminant analysis, neural network method (perceptron), decision trees. The ability to assess the accuracy of the obtained models using a set of metrics is realized. A case of solving the problem of classifying the level of crop infestation using a neural network (perceptron) is presented, the accuracy of classification was 0.73 on the test sample.
44
Wang, Liwei, Suraj Yerramilli, Akshay Iyer, Daniel Apley, Ping Zhu, and Wei Chen. "Scalable Gaussian Processes for Data-Driven Design Using Big Data With Categorical Factors." Journal of Mechanical Design 144, no. 2 (September 15, 2021). http://dx.doi.org/10.1115/1.4052221.
Abstract:
Abstract Scientific and engineering problems often require the use of artificial intelligence to aid understanding and the search for promising designs. While Gaussian processes (GP) stand out as easy-to-use and interpretable learners, they have difficulties in accommodating big data sets, categorical inputs, and multiple responses, which has become a common challenge for a growing number of data-driven design applications. In this paper, we propose a GP model that utilizes latent variables and functions obtained through variational inference to address the aforementioned challenges simultaneously. The method is built upon the latent-variable Gaussian process (LVGP) model where categorical factors are mapped into a continuous latent space to enable GP modeling of mixed-variable data sets. By extending variational inference to LVGP models, the large training data set is replaced by a small set of inducing points to address the scalability issue. Output response vectors are represented by a linear combination of independent latent functions, forming a flexible kernel structure to handle multiple responses that might have distinct behaviors. Comparative studies demonstrate that the proposed method scales well for large data sets with over 104 data points, while outperforming state-of-the-art machine learning methods without requiring much hyperparameter tuning. In addition, an interpretable latent space is obtained to draw insights into the effect of categorical factors, such as those associated with “building blocks” of architectures and element choices in metamaterial and materials design. Our approach is demonstrated for machine learning of ternary oxide materials and topology optimization of a multiscale compliant mechanism with aperiodic microstructures and multiple materials.
45
Chen, Youliang, Xiangjun Zhang, Hamed Karimian, Gang Xiao, and Jinsong Huang. "A novel framework for prediction of dam deformation based on extreme learning machine and Lévy flight bat algorithm." Journal of Hydroinformatics, July 21, 2021. http://dx.doi.org/10.2166/hydro.2021.178.
Abstract:
Abstract Dam deformation monitoring and prediction are crucial for evaluating the safety of reservoirs. There are several elements that influence dam deformation. However, the mixed effects of these elements are not always linear. Oppose to a single-kernel extreme learning machine, which suffers from poor generalization performance and instability, in this study, we proposed an improved bat algorithm for dam deformation prediction based on a hybrid-kernel extreme learning machine. To improve the learning ability of the global kernel and the generalization ability of the local kernel, we combined the global kernel function (polynomial kernel function) and local kernel function (Gaussian kernel function). Moreover, a Lévy flight bat optimization algorithm (LBA) was proposed to overcome the shortages of bat algorithms. The results showed that our model outperformed other models. This proves that our proposed algorithm and methods can be used in dam deformation monitoring and prediction in different projects and regions.
46
Wang, Wei, Hao Wang, Chen Zhang, and Fanjiang Xu. "Transfer Feature Representation via Multiple Kernel Learning." Proceedings of the AAAI Conference on Artificial Intelligence 29, no. 1 (February 21, 2015). http://dx.doi.org/10.1609/aaai.v29i1.9586.
Abstract:
Learning an appropriate feature representation across source and target domains is one of the most effective solutions to domain adaptation problems. Conventional cross-domain feature learning methods rely on the Reproducing Kernel Hilbert Space (RKHS) induced by a single kernel. Recently, Multiple Kernel Learning (MKL), which bases classifiers on combinations of kernels, has shown improved performance in the tasks without distribution difference between domains. In this paper, we generalize the framework of MKL for cross-domain feature learning and propose a novel Transfer Feature Representation (TFR) algorithm. TFR learns a convex combination of multiple kernels and a linear transformation in a single optimization which integrates the minimization of distribution difference with the preservation of discriminating power across domains. As a result, standard machine learning models trained in the source domain can be reused for the target domain data. After rewritten into a differentiable formulation, TFR can be optimized by a reduced gradient method and reaches the convergence. Experiments in two real-world applications verify the effectiveness of our proposed method.
47
Tamura, Shunsuke, Tomoyuki Miyao, and Jürgen Bajorath. "Large-scale prediction of activity cliffs using machine and deep learning methods of increasing complexity." Journal of Cheminformatics 15, no. 1 (January 7, 2023). http://dx.doi.org/10.1186/s13321-022-00676-7.
Abstract:
AbstractActivity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers. Graphical Abstract
48
Dral, Pavlo O., Fuchun Ge, Bao-Xin Xue, Yi-Fan Hou, Max Pinheiro, Jianxing Huang, and Mario Barbatti. "MLatom 2: An Integrative Platform for Atomistic Machine Learning." Topics in Current Chemistry 379, no. 4 (June 8, 2021). http://dx.doi.org/10.1007/s41061-021-00339-5.
Abstract:
AbstractAtomistic machine learning (AML) simulations are used in chemistry at an ever-increasing pace. A large number of AML models has been developed, but their implementations are scattered among different packages, each with its own conventions for input and output. Thus, here we give an overview of our MLatom 2 software package, which provides an integrative platform for a wide variety of AML simulations by implementing from scratch and interfacing existing software for a range of state-of-the-art models. These include kernel method-based model types such as KREG (native implementation), sGDML, and GAP-SOAP as well as neural-network-based model types such as ANI, DeepPot-SE, and PhysNet. The theoretical foundations behind these methods are overviewed too. The modular structure of MLatom allows for easy extension to more AML model types. MLatom 2 also has many other capabilities useful for AML simulations, such as the support of custom descriptors, farthest-point and structure-based sampling, hyperparameter optimization, model evaluation, and automatic learning curve generation. It can also be used for such multi-step tasks as Δ-learning, self-correction approaches, and absorption spectrum simulation within the machine-learning nuclear-ensemble approach. Several of these MLatom 2 capabilities are showcased in application examples.
49
Priya, R. S. Padma, and P. Senthil Vadivu. "Bio-inspired ensemble feature selection (biefs) and kernel extreme learning machine classifier for breast cancer diagnosis." International journal of health sciences, June 14, 2022, 1404–29. http://dx.doi.org/10.53730/ijhs.v6ns5.8976.
Abstract:
Breast cancer is a major disease identified in women, affecting 2.1 million women every year, and is the reason for most cancer-related mortality in women, as per the World Health Organization (WHO). For cancer researchers, accurately forecasting the life expectancy of breast cancer patients is a serious challenge. Machine Learning (ML) has acknowledged much interest in the hope of providing correct results, but due to irrelevant features, its modelling methodologies and prediction performance are still a difficulty. To solve this issue, Feature Selection (FS) was also done to verify whether comparable accuracy can be achieved even with lesser number of features or not. Bio-Inspired Ensemble Feature Selection (BIEFS) algorithm is introduced aimed at selecting a subset of features that increase the prediction performance of subsequent classification models while also simplifying their interpretability. BIEFS algorithm uses three feature selection methods such as Adaptive Mutation Enhanced Elephant Herding Optimization (AMEHO), Adaptive Mutation Butterfly Optimization Algorithm (AMBOA), and Adaptive Salp Swarm Algorithm (ASSA) and integrates their normalized outputs for getting quantitative ensemble importance. BIEFS algorithm depends upon the aggregation of multiple FS techniques by Pearson Correlation Coefficient (PCC).This BIEFS algorithm can improve the accuracy of analysis (benign and malignant).
50
Aoun, Bachir. "Stochastic atomic modeling and optimization with fullrmc." Journal of Applied Crystallography 55, no. 6 (October 14, 2022). http://dx.doi.org/10.1107/s1600576722008536.
Abstract:
Understanding materials' atomic structure with a high level of confidence and certainty is often regarded as a very arduous and sometimes impossible task, especially for newer, emerging technology materials exhibiting limited long-range order. Nevertheless, information about atomic structural properties is very valuable for materials science and synthesis. For non-crystalline amorphous and nanoscale materials, using conventional structural determination methods is impossible. Reverse Monte Carlo (RMC) modeling is commonly used to derive models of materials from experimental diffraction data. Here, the latest developments in the fullrmc software package are discussed. Despite its name, fullrmc provides a very flexible modeling framework for solving atomic structures with many methods beyond RMC. The stochastic nature of fullrmc allows it to explore all possible dimensions and degrees of freedom for atomic modeling and create statistical solutions to match measurements. Differing versions of fullrmc are provided as open source or for cloud computing access. The latter includes a modern web-based graphical user interface that incorporates advanced computing and structure-building modules and machine-learning-based components. The main features of fullrmc are presented, including constraint types, boundary conditions, density shape functions and the two running modes: stochastic using a Monte Carlo algorithm and optimization using a genetic algorithm. Capabilities include tools for statistical, mesoscopic and nanoscopic approaches, atomic or coarse-grained models, and smart artificial-intelligence-ready loss functions.