To see the other types of publications on this topic, follow the link: Deep neural decision forest.
Journal articles on the topic 'Deep neural decision forest'
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 'Deep neural decision forest.'
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
Zhou, Zhi-Hua, and Ji Feng. "Deep forest." National Science Review 6, no. 1 (2018): 74–86. http://dx.doi.org/10.1093/nsr/nwy108.
Full textAbstract:
Abstract Current deep-learning models are mostly built upon neural networks, i.e. multiple layers of parameterized differentiable non-linear modules that can be trained by backpropagation. In this paper, we explore the possibility of building deep models based on non-differentiable modules such as decision trees. After a discussion about the mystery behind deep neural networks, particularly by contrasting them with shallow neural networks and traditional machine-learning techniques such as decision trees and boosting machines, we conjecture that the success of deep neural networks owes much to three characteristics, i.e. layer-by-layer processing, in-model feature transformation and sufficient model complexity. On one hand, our conjecture may offer inspiration for theoretical understanding of deep learning; on the other hand, to verify the conjecture, we propose an approach that generates deep forest holding these characteristics. This is a decision-tree ensemble approach, with fewer hyper-parameters than deep neural networks, and its model complexity can be automatically determined in a data-dependent way. Experiments show that its performance is quite robust to hyper-parameter settings, such that in most cases, even across different data from different domains, it is able to achieve excellent performance by using the same default setting. This study opens the door to deep learning based on non-differentiable modules without gradient-based adjustment, and exhibits the possibility of constructing deep models without backpropagation.
2
Kumano, So, and Tatsuya Akutsu. "Comparison of the Representational Power of Random Forests, Binary Decision Diagrams, and Neural Networks." Neural Computation 34, no. 4 (2022): 1019–44. http://dx.doi.org/10.1162/neco_a_01486.
Full textAbstract:
Abstract In this letter, we compare the representational power of random forests, binary decision diagrams (BDDs), and neural networks in terms of the number of nodes. We assume that an axis-aligned function on a single variable is assigned to each edge in random forests and BDDs, and the activation functions of neural networks are sigmoid, rectified linear unit, or similar functions. Based on existing studies, we show that for any random forest, there exists an equivalent depth-3 neural network with a linear number of nodes. We also show that for any BDD with balanced width, there exists an equivalent shallow depth neural network with a polynomial number of nodes. These results suggest that even shallow neural networks have the same or higher representation power than deep random forests and deep BDDs. We also show that in some cases, an exponential number of nodes are required to express a given random forest by a random forest with a much fewer number of trees, which suggests that many trees are required for random forests to represent some specific knowledge efficiently.
3
Nandi Tultul, Ahana, Romana Afroz, and Md Alomgir Hossain. "Comparison of the efficiency of machine learning algorithms for phishing detection from uniform resource locator." Indonesian Journal of Electrical Engineering and Computer Science 28, no. 3 (2022): 1640. http://dx.doi.org/10.11591/ijeecs.v28.i3.pp1640-1648.
Full textAbstract:
We are using cyberspace for completing our daily life activities because of the growth of Internet. Attackers use some approachs, such as phishing, with the use of false websites to collect personal information of users. Although, software companies launch products to prevent phishing attacks, identifying a webpage as legitimate or phishing, is a very defficult and these products cannot protect from attacks. In this paper, an anti-phishing system has been introduced that can extract feature from website’s URL as instant basis and use four classification algorithms named as K-Nearest neighbor, decision tree, support vector machine, random forest on these features. According to the comparison of the experimental results from these algorithms, random forest algorithm with the selected features gives the highest performance with the 95.67% accuracy rate. Then we have used one deep learning algorithm as enhanced of our experiment named as deep neural decision forests which have given performance with the 92.67% accuracy rate. Then we have created a system which can extract the features from raw URL and pass the features to our deep neural decision forest trained model and can classify the URL as Phishing or legitimate.
4
Tultul, Ahana Nandi, Romana Afroz, and Md Alomgir Hossain. "Comparison of the efficiency of machine learning algorithms for phishing detection from uniform resource locator." Indonesian Journal of Electrical Engineering and Computer Science 28, no. 3 (2022): 1640–48. https://doi.org/10.11591/ijeecs.v28.i3.pp1640-1648.
Full textAbstract:
We are using cyberspace for completing our daily life activities because of the growth of Internet. Attackers use some approachs, such as phishing, with the use of false websites to collect personal information of users. Although, software companies launch products to prevent phishing attacks, identifying a webpage as legitimate or phishing, is a very defficult and these products cannot protect from attacks. In this paper, an anti-phishing system has been introduced that can extract feature from website’s URL as instant basis and use four classification algorithms named as K-Nearest neighbor, decision tree, support vector machine, random forest on these features. According to the comparison of the experimental results from these algorithms, random forest algorithm with the selected features gives the highest performance with the 95.67% accuracy rate. Then we have used one deep learning algorithm as enhanced of our experiment named as deep neural decision forests which have given performance with the 92.67% accuracy rate. Then we have created a system which can extract the features from raw URL and pass the features to our deep neural decision forest trained model and can classify the URL as Phishing or legitimate.
5
M, Sudharshan. "Pneumonia Prediction and Decision Support System." INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT 09, no. 05 (2025): 1–9. https://doi.org/10.55041/ijsrem47649.
Full textAbstract:
Abstract The rising prevalence of pneumonia demands automated diagnostic systems to enhance clinical efficiency and accuracy. Traditional diagnosis, reliant on manual chest X-ray and blood test analysis, is time-consuming and error-prone. This project introduces a deep learning-based system for pneumonia prediction, integrating chest X-ray images and blood test biomarkers to classify patients as healthy, viral, or bacterial pneumonia. It employs Convolutional Neural Networks (CNNs) for X-ray feature extraction, Random Forests for biomarker classification, and a heuristic-based fusion model for accurate predictions. Preprocessing includes image normalization and biomarker validation, with Grad-CAM enhancing interpretability. The system achieves 93.4% fusion accuracy, processes cases in 3.8 seconds, and generates structured clinical reports. Scalable and web-based, it supports paperless healthcare and hospital integration. Future enhancements include multilingual support and cloud deployment, advancing digital transformation in medical diagnostics. Keywords— Pneumonia Prediction, Deep Learning, Multimodal Fusion, CNN, Random Forest, Healthcare Automation
6
Liu, Xiaobo, Xu Yin, Min Wang, Yaoming Cai, and Guang Qi. "Emotion Recognition Based on Multi-Composition Deep Forest and Transferred Convolutional Neural Network." Journal of Advanced Computational Intelligence and Intelligent Informatics 23, no. 5 (2019): 883–90. http://dx.doi.org/10.20965/jaciii.2019.p0883.
Full textAbstract:
In human-machine interaction, facial emotion recognition plays an important role in recognizing the psychological state of humans. In this study, we propose a novel emotion recognition framework based on using a knowledge transfer approach to capture features and employ an improved deep forest model to determine the final emotion types. The structure of a very deep convolutional network is learned from ImageNet and is utilized to extract face and emotion features from other data sets, solving the problem of insufficiently labeled samples. Then, these features are input into a classifier called multi-composition deep forest, which consists of 16 types of forests for facial emotion recognition, to enhance the diversity of the framework. The proposed method does not need require to train a network with a complex structure, and the decision tree-based classifier can achieve accurate results with very few parameters, making it easier to implement, train, and apply in practice. Moreover, the classifier can adaptively decide its model complexity without iteratively updating parameters. The experimental results for two emotion recognition problems demonstrate the superiority of the proposed method over several well-known methods in facial emotion recognition.
7
Lee, Sang-Hyun. "Performance Evaluation of Machine Learning and Deep Learning-Based Models for Predicting Remaining Capacity of Lithium-Ion Batteries." Applied Sciences 13, no. 16 (2023): 9127. http://dx.doi.org/10.3390/app13169127.
Full textAbstract:
Lithium-ion batteries are widely used in electric vehicles, smartphones, and energy storage devices due to their high power and light weight. The goal of this study is to predict the remaining capacity of a lithium-ion battery and evaluate its performance through three machine learning models: linear regression, decision tree, and random forest, and two deep learning models: neural network and ensemble model. Mean squared error (MSE), mean absolute error (MAE), coefficient of determination (R-squared), and root mean squared error (RMSE) were used to measure prediction accuracy. For the evaluation of the artificial intelligence model, the dataset was downloaded and integrated with measurement data of the CS2 lithium-ion battery provided by the University of Maryland College of Engineering. As a result of the study, the RMSE of the linear regression model was 0.045, the decision tree model was 0.038, the random forest model was 0.034, the neural network model was 0.032, and the ensemble model was 0.030. According to the measured values, the ensemble model showed the best predictive performance, followed by the neural network model. Decision tree and random forest models also showed very good performance, and the linear regression model showed relatively poor predictive performance compared to the other models.
8
Naderpour, Mohsen, Hossein Mojaddadi Rizeei, and Fahimeh Ramezani. "Forest Fire Risk Prediction: A Spatial Deep Neural Network-Based Framework." Remote Sensing 13, no. 13 (2021): 2513. http://dx.doi.org/10.3390/rs13132513.
Full textAbstract:
Forest fire is one of the foremost environmental disasters that threatens the Australian community. Recognition of the occurrence patterns of fires and the identification of fire risk is beneficial to mitigate probable fire threats. Machine learning techniques are recognized as well-known approaches to solving non-linearity problems such as forest fire risk. However, assessing such environmental multivariate disasters has always been challenging as modelling may be biased from multiple uncertainty sources such as the quality and quantity of input parameters, training processes, and a default setup for hyper-parameters. In this study, we propose a spatial framework to quantify the forest fire risk in the Northern Beaches area of Sydney. Thirty-six significant key indicators contributing to forest fire risk were selected and spatially mapped from different contexts such as topography, morphology, climate, human-induced, social, and physical perspectives as input to our model. Optimized deep neural networks were developed to maximize the capability of the multilayer perceptron for forest fire susceptibility assessment. The results show high precision of developed model against accuracy assessment metrics of ROC = 95.1%, PRC = 93.8%, and k coefficient = 94.3%. The proposed framework follows a stepwise procedure to run multiple scenarios to calculate the probability of forest risk with new input contributing parameters. This model improves adaptability and decision-making as it can be adapted to different regions of Australia with a minor localization adoption requirement of the weighting procedure.
9
Du, Lei, Haifeng Song, Yingying Xu, and Songsong Dai. "An Architecture as an Alternative to Gradient Boosted Decision Trees for Multiple Machine Learning Tasks." Electronics 13, no. 12 (2024): 2291. http://dx.doi.org/10.3390/electronics13122291.
Full textAbstract:
Deep networks-based models have achieved excellent performances in various applications for extracting discriminative feature representations by convolutional neural networks (CNN) or recurrent neural networks (RNN). However, CNN or RNN may not work when handling data without temporal/spatial structures. Therefore, finding a new technique to extract features instead of CNN or RNN is a necessity. Gradient Boosted Decision Trees (GBDT) can select the features with the largest information gain when building trees. In this paper, we propose an architecture based on the ensemble of decision trees and neural network (NN) for multiple machine learning tasks, e.g., classification, regression, and ranking. It can be regarded as an extension of the widely used deep-networks-based model, in which we use GBDT instead of CNN or RNN. This architecture consists of two main parts: (1) the decision forest layers, which focus on learning features from the input data, (2) the fully connected layers, which focus on distilling knowledge from the decision forest layers. Powered by these two parts, the proposed model could handle data without temporal/spatial structures. This model can be efficiently trained by stochastic gradient descent via back-propagation. The empirical evaluation results of different machine learning tasks demonstrate the the effectiveness of the proposed method.
10
Alrayes, Fatma S., Mohammed Zakariah, Maha Driss, and Wadii Boulila. "Deep Neural Decision Forest (DNDF): A Novel Approach for Enhancing Intrusion Detection Systems in Network Traffic Analysis." Sensors 23, no. 20 (2023): 8362. http://dx.doi.org/10.3390/s23208362.
Full textAbstract:
Intrusion detection systems, also known as IDSs, are widely regarded as one of the most essential components of an organization’s network security. This is because IDSs serve as the organization’s first line of defense against several cyberattacks and are accountable for accurately detecting any possible network intrusions. Several implementations of IDSs accomplish the detection of potential threats throughout flow-based network traffic analysis. Traditional IDSs frequently struggle to provide accurate real-time intrusion detection while keeping up with the changing landscape of threat. Innovative methods used to improve IDSs’ performance in network traffic analysis are urgently needed to overcome these drawbacks. In this study, we introduced a model called a deep neural decision forest (DNDF), which allows the enhancement of classification trees with the power of deep networks to learn data representations. We essentially utilized the CICIDS 2017 dataset for network traffic analysis and extended our experiments to evaluate the DNDF model’s performance on two additional datasets: CICIDS 2018 and a custom network traffic dataset. Our findings showed that DNDF, a combination of deep neural networks and decision forests, outperformed reference approaches with a remarkable precision of 99.96% by using the CICIDS 2017 dataset while creating latent representations in deep layers. This success can be attributed to improved feature representation, model optimization, and resilience to noisy and unbalanced input data, emphasizing DNDF’s capabilities in intrusion detection and network security solutions.
11
Kode, Hepseeba, and Buket D. Barkana. "Deep Learning- and Expert Knowledge-Based Feature Extraction and Performance Evaluation in Breast Histopathology Images." Cancers 15, no. 12 (2023): 3075. http://dx.doi.org/10.3390/cancers15123075.
Full textAbstract:
Cancer develops when a single or a group of cells grows and spreads uncontrollably. Histopathology images are used in cancer diagnosis since they show tissue and cell structures under a microscope. Knowledge-based and deep learning-based computer-aided detection is an ongoing research field in cancer diagnosis using histopathology images. Feature extraction is vital in both approaches since the feature set is fed to a classifier and determines the performance. This paper evaluates three feature extraction methods and their performance in breast cancer diagnosis. Features are extracted by (1) a Convolutional Neural Network, (2) a transfer learning architecture VGG16, and (3) a knowledge-based system. The feature sets are tested by seven classifiers, including Neural Network (64 units), Random Forest, Multilayer Perceptron, Decision Tree, Support Vector Machines, K-Nearest Neighbors, and Narrow Neural Network (10 units) on the BreakHis 400× image dataset. The CNN achieved up to 85% for the Neural Network and Random Forest, the VGG16 method achieved up to 86% for the Neural Network, and the knowledge-based features achieved up to 98% for Neural Network, Random Forest, Multilayer Perceptron classifiers.
12
Rahman, Md Motiur, Eaftekhar Ahmed Rana, Nafisa Nawar Tamzi, Indrajit Saha, and Fazlul Hasan Siddiqui. "A Deep Learning Approach-FDNN: Forest Deep Neural Network to Predict Cow’s Parturition Date." Journal of Applied Artificial Intelligence 3, no. 1 (2022): 61–74. http://dx.doi.org/10.48185/jaai.v3i1.522.
Full textAbstract:
In this prospective study, we integrate neural network architecture with a supervised random forest feature detector to develop a new model named Forest Deep Neural Network (FDNN) to predict daily and hourly calving time of cattle. To overcome challenges of prediction problems like data sparsity along with unknown correlation structures, we incorporate the benefits of random forest (RF) with a deep neural network (DNN) to predict the daily and hourly calving time of cattle, which is nobody done yet. For this study, we take a total of 45 Holstein-Friesian cows (27 primiparous and 18 multiparous) for collecting physical activities. Using IceQube and HR Tag technologies, we record daily and hourly lying time, the number of stand-ups, ruminating time, the number of steps, and the number of head moves of cattle from 15 days before the actual calving time. Different statistical analysis has been carried out over the daily and hourly-captured data, and we have found that these monitored physical activities change very significantly over time. We have applied five classifiers such as FDNN, DNN, RF, decision tree (DT), and support vector machine (SVM) over the daily and hourly datasets. Hyperparameter optimization has been conducted over the classifiers using Grid Search approach to filter out the optimal parameter configurations. With optimal parameters, our developed model overpowered the other four classifiers in terms of accuracy, sensitivity, specificity, and ROC score (ACC= 98.38, SN=88.19, SP=98.41, and ROC=99 of predicting daily calving time; ACC=97.93, SN=97.40, SP=89.42, and ROC=98 of predicting hourly calving time).
13
Bhaskara Rao B. "Renaissance for Alzheimer’s Disease Detection using ML DL Techniques." Power System Technology 49, no. 1 (2025): 982–98. https://doi.org/10.52783/pst.1646.
Full textAbstract:
The early diagnosis of Alzheimer’s disease (AD) is very important in ensuring proper intervention in a timely manner so that patients can be managed better and have better outcomes. Earlier, cognitivedata and numerical patient information were used with conventional machine learning (ML) methods for early detection. One of the past ensemble-based approaches trained up to seven ML classifiers that included Decision Tree, Random Forest, SVM, ANN, and AdaBoost, gaining 93.92% accuracy. However, deep learning integration and optimal feature extraction was not present in that study. This study proposes a novel deep learning (DL) approach using advanced feature selection by Decision Tree and Random Forest, and Synthetic Minority Over-Sampling Technique (SMOTE) for class balancing. A hybrid deep learning approach based on Convolutional Neural Networks (CNN) and Artificial Neural Networks (ANN) is presented. The model provides efficient, scalable, and adaptive, and a robust Alzheimer’s disease (AD) detection for the real-life patients. The models are trained successfully with patients' data from Kaggle.The model provides efficient early AD detection with better accuracy and stability.
14
Chinmoy Modak, Sandip Kumar Ghosh, Md Ariful Islam Sarkar, et al. "Machine Learning Model in Digital Marketing Strategies for Customer Behavior: Harnessing CNNs for Enhanced Customer Satisfaction and Strategic Decision-Making." Journal of Economics, Finance and Accounting Studies 6, no. 3 (2024): 178–86. http://dx.doi.org/10.32996/jefas.2024.6.3.14.
Full textAbstract:
In the realm of digital marketing for the banking industry, the integration of deep learning methodologies, particularly Convolutional Neural Networks (CNNs) such as VGG16, Resnet50, and InceptionV3, has revolutionized strategic decision-making and customer satisfaction. This study explores how deep learning models leverage neural networks with multiple layers to analyze vast and complex datasets, uncovering intricate patterns in customer behavior and preferences. By enhancing customer segmentation, optimizing campaign performance, and refining personalized experiences, CNNs empower banks to make precise, data-driven decisions that elevate customer satisfaction and loyalty. Comparative analyses demonstrate CNNs' superior performance over traditional models like Random Forest and Logistic Regression, achieving accuracies up to 89% and F1 scores of 88%, thereby highlighting their transformative potential in reshaping digital marketing strategies within the banking sector. This research underscores the critical implications of adopting advanced deep learning techniques to meet the evolving demands of customers in today's dynamic digital landscape.
15
Sekhar Reddy, Peram Chandra. "Hybrid Worm Detection Based on Signature & Anomaly." INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT 09, no. 04 (2025): 1–9. https://doi.org/10.55041/ijsrem46521.
Full textAbstract:
Abstract - Internet worms pose a significant threat by propagating through network traffic, compromising system security, and exfiltrating sensitive information. To enhance detection accuracy, a hybrid two-factor worm detection system is proposed, integrating signature-based and anomaly-based methodologies. Signature-based detection employs packet capture (PCAP) analysis and NetFlow inspection to identify malicious signatures using predefined rule sets. Honeypot logs are leveraged to detect and mitigate attack attempts by monitoring unauthorized access attempts. Anomaly-based detection utilizes machine learning models, including Random Forest, Decision Tree, and Bayesian Networks, to classify network traffic as normal or abnormal based on behavioral patterns. Experimental results demonstrate that Random Forest and Decision Tree achieve the highest accuracy of 98%, outperforming Bayesian Networks. Additionally, deep learning models such as Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and Gated Recurrent Units (GRU) are employed for anomaly detection, with GRU achieving superior performance. The proposed framework effectively enhances worm detection capabilities, reducing false positives and improving cybersecurity resilience. Future enhancements include integrating evolutionary feature selection techniques such as Genetic Algorithms and Particle Swarm Optimization to optimize detection accuracy. Key Words: Anomaly Detection, Worm Detection, Machine Learning, Deep Learning, Random Forest, Decision Tree, Convolutional Neural Networks, Long Short-Term Memory, Gated Recurrent Units.
16
SRIVALLI, PARUCHURI, PODAMEKALA LAHARI, PULETIPALLI SAFEENA, MRS VIDHYA ,, MR J. JAYAPRAKASH ,, and MRS CHINCHU NAIR. "Fake Account Detection on Social Media Using Machine Learning and Deep Learning." INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT 09, no. 03 (2025): 1–9. https://doi.org/10.55041/ijsrem42987.
Full textAbstract:
Social networking websites have become an essential part of life, making it easy for individuals to stay connected and exchange information. They provide numerous features, including the ability to chat with others, share news, plan events, and many more. But with the increasing number of users and the volume of personal information, the bad guys have also seen opportunities to exploit these networks. They exploit security loopholes to pilfer personal data, propagate false information, and partake in other malicious activities. As a result, researchers have been focusing on developing effective methods to detect suspicious activities and identify fake accounts. While some features of social media accounts can help with these efforts, they may sometimes have little to no effect, or even negatively impact the results. Additionally, relying on standalone classification algorithms doesn’t always yield optimal outcomes. This paper suggests using the Decision Tree algorithm to effectively detect fake Instagram accounts by employing four feature selection and dimensionality reduction techniques. In previous research, algorithms such as Decision Trees, Random Forest, Logistic Regression, and Convolutional Neural Networks (CNN) were explored for classification. Among these, CNNs performed exceptionally well, accurately identifying fake accounts and producing satisfying results. Given their high performance, we applied CNNs for Instagram account classification in our study. With deep learning offering various types of neural networks, CNNs have proven to be the most effective for this type of task. KEYWORDS: Decision Tree, Random Forest, Logistic Regression, and CNN (convolution neural network)
17
Bouzakraoui, Moulay Smail, Abdelalim Sadiq, and Alaoui Abdessamad Youssfi. "Deep Learning Model to Analyze Customer's Satisfaction." International Journal of Engineering and Advanced Technology (IJEAT) 9, no. 4 (2020): 1709–14. https://doi.org/10.35940/ijeat.C6610.049420.
Full textAbstract:
Nowadays, measuring customer satisfaction is an important strategic tool for companies; many manual methods exist to measure customer’s satisfaction. However, the results have not effective and efficient. In this paper, we propose a new method for facial emotion detection to recognize customer’s satisfaction using a deep learning model. We used a convolutional neural network to detect facial key points. These key points help us to extract geometric features from customer’s emotional faces. Indeed, we computed distances between neutral face and negative or positive feedback. After that, we classified these distances by using Support Vector Machine (SVM), KNN, Random Forest, and Decision Tree. To evaluate the performance of our approach, we tested our algorithm by using FACEDB and JAFFE datasets. We found that SVM is the most performant classifier. We obtained 96% as accuracy by using FACEDB dataset and 95% by using JAFFE dataset.
18
Mate Landry, Gilgen, Rodolphe Nsimba Malumba, Fiston Chrisnovi Balanganayi Kabutakapua, and Bopatriciat Boluma Mangata. "PERFORMANCE COMPARISON OF CLASSICAL ALGORITHMS AND DEEP NEURAL NETWORKS FOR TUBERCULOSIS PREDICTION." Jurnal Techno Nusa Mandiri 21, no. 2 (2024): 126–33. http://dx.doi.org/10.33480/techno.v21i2.5609.
Full textAbstract:
This study compares the performance of several classical machine learning algorithms and deep neural networks for the prediction of tuberculosis in the Democratic Republic of Congo (DRC), using a sample of 1000 cases including clinical and demographic data. The sample is divided into two sets: 80% for training and 20% for testing. The algorithms evaluated include Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree, Random Forest and Convolutional Neural Networks (CNN). The results show that the CNN has the best overall performance with an accuracy of 94%, an AUC of the ROC curve of 93%, an accuracy of 90%, an accuracy of 95%, a sensitivity of 88%, an F1-score of 91.3% and a Log Loss of 0.0386. The Random Forest follows closely behind with an accuracy of 92% and an AUC of 86%. The SVM and KNN models also performed strongly, but slightly less well. The Decision Tree obtained acceptable results, but inferior to the other algorithms evaluated. These results indicate that deep neural networks, and in particular the CNN, are superior for predicting tuberculosis compared with conventional machine learning algorithms. This superiority is particularly marked in terms of accuracy, sensitivity and reliability of predictions, as shown by the performance metrics obtained.
19
Wang, Shaoyang. "A comparative study of the deep learning based model and the conventional machine learning based models in human activity recognition." Applied and Computational Engineering 54, no. 1 (2024): 117–23. http://dx.doi.org/10.54254/2755-2721/54/20241421.
Full textAbstract:
Human activity recognition (HAR) has been widely studied as a research field in human behavior analysis due to its huge potential in various application domains such as health care and behavioral science. Recently, deep learning (DL) based methods have also been successfully applied to predict various human activities. This research aims at building different Python-based models to perform HAR using smartphones and calculating and comparing the accuracy of the models to select the optimal one. Four models were built to classify and predict human activities: Deep Convolutional Neural Network (DCNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF). The results of the experiments in this paper show that the Deep Convolutional Neural Network achieves an average recognition accuracy rate of 95.49%, exceeding the other three models. The underlying reason may be that Deep Convolutional Neural Network is based on a more advanced algorithm deep learning technique.
20
Sreekanth, S., Ch Sriram, and P. Sujan. "Forest Fire Detection using Deep Learning." International Journal for Research in Applied Science and Engineering Technology 11, no. 8 (2023): 1118–21. http://dx.doi.org/10.22214/ijraset.2023.55296.
Full textAbstract:
Abstract: The threat posed by forest fires to humanity has increased. In addition to giving a lot of living things a place to live and protection, they have also been a significant source of food, wood, and a lot of other goods. Throughout the dawn of time, woods have been integral to social, economic, and religious endeavors and have improved human life in numerous tangible and intangible ways. We must exercise sufficient caution when making decisions that could ultimately result in a tragic outcome if we want to safeguard our environment from these forest fires that are spreading quickly. Thus, we suggest an image identification technique based on convolutional neural networks for the early detection of forest fires (CNN).
21
Quaderi, Shah Jafor Sadeek, Sadia Afrin Labonno, Sadia Mostafa, and Shamim Akhter. "Identify the Beehive Sound using Deep Learning." International Journal of Computer Science and Information Technology 14, no. 4 (2022): 13–29. http://dx.doi.org/10.5121/ijcsit.2022.14402.
Full textAbstract:
Flowers play an essential role in removing the duller from the environment. The life cycle of the flowering plants involves pollination, fertilization, flowering, seed- formation, dispersion, and germination. Honeybees pollinate approximately 75% of all flowering plants. Environmental pollution, climate change, natural landscape demolition, and so on, threaten the natural habitats, thus continuously reducing the number of honeybees. As a result, several researchers are attempting to resolve this issue. Applying acoustic classification to recordings of beehive sounds may be a way of detecting changes within them. In this research, we use deep learning techniques, namely Sequential Neural Network, Convolutional Neural Network, and Recurrent Neural Network, on the recorded sounds to classify bee sounds from the nonbeehive noises. In addition, we perform a comparative study among some popular non-deep learning techniques, namely Support Vector Machine, Decision Tree, Random Forest, and Naïve Bayes, with the deep learning techniques. The techniques are also verified on the combined recorded sounds (25-75% noises).
22
Kislov, Dmitry E., and Kirill A. Korznikov. "Automatic Windthrow Detection Using Very-High-Resolution Satellite Imagery and Deep Learning." Remote Sensing 12, no. 7 (2020): 1145. http://dx.doi.org/10.3390/rs12071145.
Full textAbstract:
Wind disturbances are significant phenomena in forest spatial structure and succession dynamics. They cause changes in biodiversity, impact on forest ecosystems at different spatial scales, and have a strong influence on economics and human beings. The reliable recognition and mapping of windthrow areas are of high importance from the perspective of forest management and nature conservation. Recent research in artificial intelligence and computer vision has demonstrated the incredible potential of neural networks in addressing image classification problems. The most efficient algorithms are based on artificial neural networks of nested and complex architecture (e.g., convolutional neural networks (CNNs)), which are usually referred to by a common term—deep learning. Deep learning provides powerful algorithms for the precise segmentation of remote sensing data. We developed an algorithm based on a U-Net-like CNN, which was trained to recognize windthrow areas in Kunashir Island, Russia. We used satellite imagery of very-high spatial resolution (0.5 m/pixel) as source data. We performed a grid search among 216 parameter combinations defining different U-Net-like architectures. The best parameter combination allowed us to achieve an overall accuracy for recognition of windthrow sites of up to 94% for forested landscapes by coniferous and mixed coniferous forests. We found that the false-positive decisions of our algorithm correspond to either seashore logs, which may look similar to fallen tree trunks, or leafless forest stands. While the former can be rectified by applying a forest mask, the latter requires the usage of additional information, which is not always provided by satellite imagery.
23
Prakash, M., S. Neelakandan, M. Tamilselvi, S. Velmurugan, S. Baghavathi Priya, and Eric Ofori Martinson. "Deep Learning-Based Wildfire Image Detection and Classification Systems for Controlling Biomass." International Journal of Intelligent Systems 2023 (September 16, 2023): 1–18. http://dx.doi.org/10.1155/2023/7939516.
Full textAbstract:
Forests are essential natural resources that directly impact the ecosystem. However, the rising frequency of forest fires due to natural and artificial climate change has become a critical issue. A revolutionary municipal application proposes deploying an artificial intelligence-based forest fire warning system to prevent major disasters. This work aims to present an overview of vision-based methods for detecting and categorizing forest fires. The study employs a forest fire detection dataset to address the classification difficulty of discriminating between photos with and without fire. This method is based on convolutional neural network transfer learning with Inception-v3. Thus, automatic identification of current forest fires (including burning biomass) is a critical field of research for reducing negative repercussions. Early fire detection can also assist decision-makers in developing mitigation and extinguishment strategies. Radial basis function Networks (RBFNs) with rapid and accurate image super resolution (RAISR) is a deep learning framework trained on an input dataset to detect active fires and burning biomass. The proposed RBFN-RAISR model’s performance in recognizing fires and nonfires was compared to earlier CNN models using several performance criteria. The water wave optimization technique is used for image feature selection, noise and blurring reduction, image improvement and restoration, and image enhancement and restoration. When classifying fire and no-fire photos, the proposed RBFN-RAISR fire detection approach achieves 97.55% accuracy, 93.33% F-Score, 96.44% recall, 94.19% precision, and an error rate of 24.89. Given the one-of-a-kind forest fire detection dataset, the suggested method achieves promising results for the forest fire categorization problem.
24
Laith Al-Ali. "Prediction of Corona-Virus Using Deep Learning." Tikrit Journal of Pure Science 27, no. 1 (2022): 122–32. http://dx.doi.org/10.25130/tjps.v27i1.89.
Full textAbstract:
With the rapid spread of the Corona virus in most parts of the worldwide, it has become necessary to find solutions to contain and treat this epidemic. This research presents a method to predict the occurrence of COVID-19 based on different symptoms of the disease, using non-clinical methods such as artificial intelligence, to help medical staff, save the cost of testing (PCR), and get results in a short time. Artificial intelligence provides many tools for data analysis, statistical analysis, and intelligent research. In this paper, we focus on predicting COVID-19 infection, using Artificial Neural Networks (ANN), random forests and decision trees, to effectively analyze medical datasets, based on the most common and acute symptoms, such as cough, fever, headache, diarrhea, living in infected areas Pain and shortness of breath. Breathing, chills, nasal congestion and some other symptoms of the disease. A data set consisting of (1495) patients is used to determine whether or not a person has this disease, after determining the symptoms that appear on it. The data set is divided into 75% of the training data and 25% of the test data after applying deep learning algorithms. Python libraries such as pandas, NumPy, and matplotlib are also used in addition to sklearn and Keras. The search results show very high accuracy indicated by 91% of Random Forest with estimators = 200 and 91% of the decision tree. the accuracy of an artificial neural network is 85%. Thus, this research provides an important indicator for the possible prediction of COVID-19 infection.
25
Liu, Pengshuai, Xiaojun Yin, Mingrui Ding, and Shaoliang Pan. "Research on Protective Forest Change Detection in Aral City Based on Deep Learning." Forests 16, no. 5 (2025): 775. https://doi.org/10.3390/f16050775.
Full textAbstract:
Protective forests play a crucial role in ecosystems, particularly in arid and semi-arid regions, where they provide irreplaceable ecological functions such as windbreaks, sand fixation, soil and water conservation, and climate regulation. This study selects Aral City in Xinjiang as the research area and proposes a method that integrates high-resolution remote sensing data (GF-2) with a Spatiotemporal Attention Neural Network (STANet) model to improve the accuracy of protective forest change detection. The study utilizes GF-2 remote sensing imagery and employs a spatiotemporal attention mechanism to incorporate spatial and temporal information, overcoming the limitations of traditional methods in processing long-term time-series remote sensing data. The results demonstrate that the combination of GF-2 imagery and the STANet model effectively detects protective forest changes in Aral City, achieving an F1-score of 83.64% and an accuracy of 78.52%, indicating significant detection capability. Spatial analysis based on the change detection results reveals notable changes in the protective forest area within the study region, with a decline in vegetation coverage in certain areas. This study suggests that the STANet method has strong application potential in protective forest change detection in arid regions, providing precise spatiotemporal change information for protective forest restoration and management. The findings offer a scientific basis for ecological restoration and sustainable development in Aral City, Xinjiang, and are of great significance for improving protective forest management and land use decision-making.
26
Shanmugarajeshwari, V., and M. Ilayaraja. "Intelligent Decision Support for Identifying Chronic Kidney Disease Stages." International Journal of Intelligent Information Technologies 20, no. 1 (2023): 1–22. http://dx.doi.org/10.4018/ijiit.334557.
Full textAbstract:
The decision tree classification algorithm is becoming increasingly important in machine learning (ML) technology. It is being used in a variety of fields to solve extremely complicated issues. DTCA is also utilised in medical health data to identify chronic kidney disorders such as cancer and diabetes utilising computer-aided diagnosis. Deep learning is an intelligent area of machine learning in which neural networks are used to learn unsupervised from unstructured or unlabeled data. For CKD, the DL employed the deep stacked auto-encoder and soft-max classifier techniques. Kidney illness is another condition that can lead to a variety of health problems. Random forest, SVM, C5.0, decision tree classification algorithm, C4.5, ANN, neuro-fuzzy systems, classification and clustering, DSAE, DNN, FNC, MLP are used in this study to predict and identify an early diagnosis of CKD patients using various machine and deep learning algorithms using R Studio and Python Colab software. The many stages of chronic kidney disease are identified in this paper.
27
Bella, Kamal, Azidine Guezzaz, Said Benkirane, et al. "An efficient intrusion detection system for IoT security using CNN decision forest." PeerJ Computer Science 10 (September 9, 2024): e2290. http://dx.doi.org/10.7717/peerj-cs.2290.
Full textAbstract:
The adoption and integration of the Internet of Things (IoT) have become essential for the advancement of many industries, unlocking purposeful connections between objects. However, the surge in IoT adoption and integration has also made it a prime target for malicious attacks. Consequently, ensuring the security of IoT systems and ecosystems has emerged as a crucial research area. Notably, advancements in addressing these security threats include the implementation of intrusion detection systems (IDS), garnering considerable attention within the research community. In this study, and in aim to enhance network anomaly detection, we present a novel intrusion detection approach: the Deep Neural Decision Forest-based IDS (DNDF-IDS). The DNDF-IDS incorporates an improved decision forest model coupled with neural networks to achieve heightened accuracy (ACC). Employing four distinct feature selection methods separately, namely principal component analysis (PCA), LASSO regression (LR), SelectKBest, and Random Forest Feature Importance (RFFI), our objective is to streamline training and prediction processes, enhance overall performance, and identify the most correlated features. Evaluation of our model on three diverse datasets (NSL-KDD, CICIDS2017, and UNSW-NB15) reveals impressive ACC values ranging from 94.09% to 98.84%, depending on the dataset and the feature selection method. Notably, our model achieves a remarkable prediction time of 0.1 ms per record. Comparative analyses with other recent random forest and Convolutional Neural Networks (CNN) based models indicate that our DNDF-IDS performs similarly or even outperforms them in certain instances, particularly when utilizing the top 10 features. One key advantage of our novel model lies in its ability to make accurate predictions with only a few features, showcasing an efficient utilization of computational resources.
28
Aksoy, Ceren, Ayhan Küçükmanisa, and Zeynep Hilal Kilimci. "Forecasting Customer Churn using Machine Learning and Deep Learning Approaches." Kocaeli Journal of Science and Engineering 8, no. 1 (2025): 60–70. https://doi.org/10.34088/kojose.1526621.
Full textAbstract:
Customer churn forecasting is a challenging task recommended for churn prevention for companies operating in various industries such as banking, telecommunications, and insurance. Forecasting customer churn is very important for many companies because gaining potential customers usually costs more than retaining present ones. That is why companies, analysts, and researchers are center on analyzing the dynamics behind customer churn behaviors. In this study, we present a comparative study for the purpose of forecasting customer churn employing publicly available datasets, namely, IBM Watson and Call-Detailed Record (CDR). For this purpose, logistic regression, random forest, decision tree, k-nearest neighbor, extreme gradient boosting, and naive Bayes techniques are evaluated as machine learning approaches while artificial neural networks and convolutional neural networks are assessed as deep learning models. Experiment results indicate that the random forest method exhibits superior performance with 79.94% accuracy for the IBM Watson dataset and 96.34% accuracy for the Call Detailed Report (CDR) dataset. To demonstrate the effectiveness of the suggested framework, a comparison with the state-of-the-art studies is performed.
29
Zhang, Jianming, Kebin Shi, Hadelibieke Majiti, et al. "Study on the Classification and Identification Methods of Surrounding Rock Excavatability Based on the Rock-Breaking Performance of Tunnel Boring Machines." Applied Sciences 13, no. 12 (2023): 7060. http://dx.doi.org/10.3390/app13127060.
Full textAbstract:
Rock mass conditions are extremely sensitive to tunnel boring machine (TBM) tunneling. Therefore, establishing a surrounding rock excavatability (SRE) classification system applicable to TBM tunnels. Accurately and intelligently identifying excavatability grades can also facilitate efficient TBM tunneling and intelligent construction. Specific excavation and penetration rates were used to evaluate SRE. Their correlations with geological and tunneling parameters were explored using the field data from two water conveyance tunnels in China with different lithologies. A high-precision empirical SRE classification system was constructed using TOPSIS for multi-objective decision-making, and it was verified using engineering cases. An intelligent identification model for SRE grades in the stable phase of a TBM excavation cycle was established using 12,382 TBM rock-breaking datasets and deep forest models. Ten characteristic parameters, e.g., total thrust, were selected as model input features. Hyperparameter optimization was achieved using the grid search method. Deep forest was compared with decision tree, random forest, support vector classifier, and deep neural network. The contribution of the model’s features was measured using random forest. The main conclusions are as follows: the proposed SRE classification method is feasible and matches well with the actual excavation. In the intelligent identification of SRE classification, the accuracy and F1 scores when using deep forest were 96.33% and 0.9581, respectively. Deep forest exhibited better grade identification performance than the four models. Among the ten input features, penetration is the most important feature for the model’s input, while the top shield cylinder rod’s chamber pressure is the least important. The findings can provide some references for SRE classification and prediction and intelligent TBM control.
30
Shchetinin, E. Yu. "EMOTIONS RECOGNITION IN HUMAN SPEECH USING DEEP NEURAL NETWORKS." Vestnik komp'iuternykh i informatsionnykh tekhnologii, no. 199 (January 2021): 44–51. http://dx.doi.org/10.14489/vkit.2021.01.pp.044-051.
Full textAbstract:
The recognition of human emotions is one of the most relevant and dynamically developing areas of modern speech technologies, and the recognition of emotions in speech (RER) is the most demanded part of them. In this paper, we propose a computer model of emotion recognition based on an ensemble of bidirectional recurrent neural network with LSTM memory cell and deep convolutional neural network ResNet18. In this paper, computer studies of the RAVDESS database containing emotional speech of a person are carried out. RAVDESS-a data set containing 7356 files. Entries contain the following emotions: 0 – neutral, 1 – calm, 2 – happiness, 3 – sadness, 4 – anger, 5 – fear, 6 – disgust, 7 – surprise. In total, the database contains 16 classes (8 emotions divided into male and female) for a total of 1440 samples (speech only). To train machine learning algorithms and deep neural networks to recognize emotions, existing audio recordings must be pre-processed in such a way as to extract the main characteristic features of certain emotions. This was done using Mel-frequency cepstral coefficients, chroma coefficients, as well as the characteristics of the frequency spectrum of audio recordings. In this paper, computer studies of various models of neural networks for emotion recognition are carried out on the example of the data described above. In addition, machine learning algorithms were used for comparative analysis. Thus, the following models were trained during the experiments: logistic regression (LR), classifier based on the support vector machine (SVM), decision tree (DT), random forest (RF), gradient boosting over trees – XGBoost, convolutional neural network CNN, recurrent neural network RNN (ResNet18), as well as an ensemble of convolutional and recurrent networks Stacked CNN-RNN. The results show that neural networks showed much higher accuracy in recognizing and classifying emotions than the machine learning algorithms used. Of the three neural network models presented, the CNN + BLSTM ensemble showed higher accuracy.
31
Duan, Ziheng, Yizhou He, and Zhongyu Wang. "Exploring the influence of lifestyle on sleep health based on deep learning." Applied and Computational Engineering 48, no. 1 (2024): 24–30. http://dx.doi.org/10.54254/2755-2721/48/20241087.
Full textAbstract:
Sleep plays a crucial role in maintaining overall health. However, various lifestyle factors significantly influence sleep quality and duration. Understanding the relationship between lifestyle choices and sleep health is crucial for individuals seeking to improve their sleep patterns. The purpose of this study is to provide valuable insights into the causes and effects of sleep disorders in order to help individuals make informed decisions to optimize their sleep health. This article implements the CatBoost gradient algorithm for predictive modeling. Among various models including K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Xtreme Gradient Boosting (XGBoost), Gradient Boosting Decision Tree (GBDT), Random Forest (RF), Decision Tree (DT), Logistic Regression (LR), Deep Neural Network (DNN), CatBoost shows better overall performance with an accuracy of 0.93, an Fl-score of 0.925, and a recall of 0.95. Through data analysis, Blood-pressure-Systolic, Blood-Pressure-Diastolic, and Stress Level are found to have the greatest impact on the model's output.
32
Kim, Sangwon, Byoung-Chul Ko, and Jaeyeal Nam. "Model Simplification of Deep Random Forest for Real-Time Applications of Various Sensor Data." Sensors 21, no. 9 (2021): 3004. http://dx.doi.org/10.3390/s21093004.
Full textAbstract:
The deep random forest (DRF) has recently gained new attention in deep learning because it has a high performance similar to that of a deep neural network (DNN) and does not rely on a backpropagation. However, it connects a large number of decision trees to multiple layers, thereby making analysis difficult. This paper proposes a new method for simplifying a black-box model of a DRF using a proposed rule elimination. For this, we consider quantifying the feature contributions and frequency of the fully trained DRF in the form of a decision rule set. The feature contributions provide a basis for determining how features affect the decision process in a rule set. Model simplification is achieved by eliminating unnecessary rules by measuring the feature contributions. Consequently, the simplified and transparent DRF has fewer parameters and rules than before. The proposed method was successfully applied to various DRF models and benchmark sensor datasets while maintaining a robust performance despite the elimination of a large number of rules. A comparison with state-of-the-art compressed DNNs also showed the proposed model simplification’s higher parameter compression and memory efficiency with a similar classification accuracy.
33
Li, Han. "Predicting Tencent's Stock Price: A Comparative Analysis of Machine Learning Algorithms." Advances in Economics, Management and Political Sciences 45, no. 1 (2023): 183–92. http://dx.doi.org/10.54254/2754-1169/45/20230281.
Full textAbstract:
Chinese internet companies, such as Tencent, are bringing a new energy to the global market. As one of the largest internet companies in China, Tencent's stock price fluctuations hold significant importance for investors and the market. Accurate forecasting of stock prices is of great importance to make high-yield decisions and manage risk. This paper aims to provide investors with effective and scientific references for decision-making by analyzing various forecasting methods. The dataset China-Techgiant-Stock-Data-in-HK-Market-2022 from Kaggle is used in this paper. This study utilizes various machine learning and time series analysis methods, including linear regression, SVM, random forest, LSTM neural network, and ARIMA, to predict stock prices using historical data. The models' accuracy and performance are compared, with traditional machine learning methods (linear regression, SVM, and random forest) contrasted against deep learning and time series analysis methods (LSTM neural network and ARIMA). It turns out that LSTM has the best prediction results. This can better guide investors in their stock decisions.
34
Pradnya, Samit Mehta, Wankhade Renuka, Pankaj Chhabada Dev, et al. "Elevating sentiment analysis with deep convolutional neural network model facial expression insights." IAES International Journal of Artificial Intelligence (IJ-AI) 13, no. 3 (2024): 3395–403. https://doi.org/10.11591/ijai.v13.i3.pp3395-3403.
Full textAbstract:
In today's data-driven world, the ability to analyze emotional responses is essential. The pressing necessity that drives this study is to revolutionize the field of sentiment analysis by extracting the hidden information from people's facial expressions. It examines people's preferences, worries, and pleasure, revealing their views on many topics. Beyond text-based sentiment analysis, this research adds facial expression-based sentiment analysis into existing systems for tailored recommendations and mental health monitoring. The system emphasizes visual stimuli's emotional influence to improve decision making, content adaptability, and user experiences. The implementation involves transfer learning with the pre-trained VGG-16 model, which enhances ability to discern intricate emotional cues from facial expressions. Convolutional neural network (CNN) and contextual analysis allow the model to understand users' emotions and provide insights into their thoughts, feelings, and behaviors. To improve emotion recognition reliability and reactivity, this study examines random forest, support vector machine (SVM), and CNN methodologies. The visual geometry group (VGG-16) CNN model outperforms over SVM and random forest classifiers with accuracy of 95%. This study highlights facial expression-based sentiment analysis.
35
Shafi, Numan, Faisal Bukhari, Waheed Iqbal, Khaled Mohamad Almustafa, Muhammad Asif, and Zubair Nawaz. "Cleft prediction before birth using deep neural network." Health Informatics Journal 26, no. 4 (2020): 2568–85. http://dx.doi.org/10.1177/1460458220911789.
Full textAbstract:
In developing countries like Pakistan, cleft surgery is expensive for families, and the child also experiences much pain. In this article, we propose a machine learning–based solution to avoid cleft in the mother’s womb. The possibility of cleft lip and palate in embryos can be predicted before birth by using the proposed solution. We collected 1000 pregnant female samples from three different hospitals in Lahore, Punjab. A questionnaire has been designed to obtain a variety of data, such as gender, parenting, family history of cleft, the order of birth, the number of children, midwives counseling, miscarriage history, parent smoking, and physician visits. Different cleaning, scaling, and feature selection methods have been applied to the data collected. After selecting the best features from the cleft data, various machine learning algorithms were used, including random forest, k-nearest neighbor, decision tree, support vector machine, and multilayer perceptron. In our implementation, multilayer perceptron is a deep neural network, which yields excellent results for the cleft dataset compared to the other methods. We achieved 92.6% accuracy on test data based on the multilayer perceptron model. Our promising results of predictions would help to fight future clefts for children who would have cleft.
36
Nalatissifa, Hiya, and Hilman Ferdinandus Pardede. "Customer Decision Prediction Using Deep Neural Network on Telco Customer Churn Data." Jurnal Elektronika dan Telekomunikasi 21, no. 2 (2021): 122. http://dx.doi.org/10.14203/jet.v21.122-127.
Full textAbstract:
Customer churn is the most important problem in the business world, especially in the telecommunications industry, because it greatly influences company profits. Getting new customers for a company is much more difficult and expensive than retaining existing customers. Machine learning, part of data mining, is a sub-field of artificial intelligence widely used to make predictions, including predicting customer churn. Deep neural network (DNN) has been used for churn prediction, but selecting hyperparameters in modeling requires more time and effort, making the process more challenging for the researcher. Therefore, the purpose of this study is to propose a better architecture for the DNN algorithm by using a hard tuner to obtain more optimal hyperparameters. The tuning hyperparameter used is random search in determining the number of nodes in each hidden layer, dropout, and learning rate. In addition, this study also uses three variations of the number of hidden layers, two variations of the activation function, namely rectified linear unit (ReLu) and Sigmoid, then uses five variations of the optimizer (stochastic gradient descent (SGD), adaptive moment estimation (Adam), adaptive gradient algorithm (Adagrad), Adadelta, and root mean square propagation (RMSprop)). Experiments show that the DNN algorithm using hyperparameter tuning random search produces a performance value of 83.09 % accuracy using three hidden layers, the number of nodes in each hidden layer is [20, 35, 15], using the RMSprop optimizer, dropout 0.1, the learning rate is 0.01, with the fastest tuning time of 21 seconds. Better than modeling using k-nearest neighbor (K-NN), random forest (RF), and decision tree (DT) as comparison algorithms.
37
Sikdar, Ayanika. "Comparative study on Early esophageal cancer detection." INTERNATIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT 09, no. 05 (2025): 1–9. https://doi.org/10.55041/ijsrem48031.
Full textAbstract:
Abstract - Esophageal cancer (EC) is a rare but serious health condition due to its late diagnosis and high fatality rates. Its early detection greatly enhances patient outcomes, and recent advances in artificial intelligence (AI) introduce new possibilities in diagnosis. This research explores deep-learning models for early EC detection based on high-resolution endoscopic images. We compared logistic regression, decision-tree and random forest methods as conventional machine learning methods with deep learning models like convolutional neural networks (CNNs) and U-Net architectures to separate cancerous from non-cancerous esophageal tissues. Annotated endoscopic images were used as the dataset, with models assessed in terms of accuracy, sensitivity, and specificity. Pilot results indicate that deep learning, particularly CNNs, performs better than conventional machine learning. These results indicate AI-based detection combined with routine procedures can enhance early diagnosis, lower errors, and enhance EC patient survival. Research in the future should improve model robustness and establish efficacy across a wide population. Key Words: Logistic Regression, Decision Tree, Random Forest, U-Net architectures, CNN, Esophageal Cancer, Machine Learning, Early Detection
38
Note, Johan, and Maaruf Ali. "Comparative Analysis of Intrusion Detection System Using Machine Learning and Deep Learning Algorithms." Annals of Emerging Technologies in Computing 6, no. 3 (2022): 19–36. http://dx.doi.org/10.33166/aetic.2022.03.003.
Full textAbstract:
Attacks against computer networks, “cyber-attacks”, are now common place affecting almost every Internet connected device on a daily basis. Organisations are now using machine learning and deep learning to thwart these types of attacks for their effectiveness without the need for human intervention. Machine learning offers the biggest advantage in their ability to detect, curtail, prevent, recover and even deal with untrained types of attacks without being explicitly programmed. This research will show the many different types of algorithms that are employed to fight against the different types of cyber-attacks, which are also explained. The classification algorithms, their implementation, accuracy and testing time are presented. The algorithms employed for this experiment were the Gaussian Naïve-Bayes algorithm, Logistic Regression Algorithm, SVM (Support Vector Machine) Algorithm, Stochastic Gradient Descent Algorithm, Decision Tree Algorithm, Random Forest Algorithm, Gradient Boosting Algorithm, K-Nearest Neighbour Algorithm, ANN (Artificial Neural Network) (here we also employed the Multilevel Perceptron Algorithm), Convolutional Neural Network (CNN) Algorithm and the Recurrent Neural Network (RNN) Algorithm. The study concluded that amongst the various machine learning algorithms, the Logistic Regression and Decision tree classifiers all took a very short time to be implemented giving an accuracy of over 90% for malware detection inside various test datasets. The Gaussian Naïve-Bayes classifier, though fast to implement, only gave an accuracy between 51-88%. The Multilevel Perceptron, non-linear SVM and Gradient Boosting algorithms all took a very long time to be implemented. The algorithm that performed with the greatest accuracy was the Random Forest Classification algorithm.
39
Almendli, Muhammed, and Jamshid Bagherzadeh Mohasefi. "Anomaly detection system based on deep learning for cyber physical systems on sensory and network datasets." International Journal of Electrical and Computer Engineering (IJECE) 14, no. 6 (2024): 6827. http://dx.doi.org/10.11591/ijece.v14i6.pp6827-6837.
Full textAbstract:
Cyber-physical systems (CPSs), a type of computing system integrated with physical devices, are widely used in many areas such as manufacturing, traffic control, and energy. The integration of CPS and networks has expanded the range of cyber threats. Intrusion detection systems (IDSs), use signature based and machine learning based techniques to protect networks, against threats in CPSs. Water purifying plants are among the important CPSs. In this context some research uses a dataset obtained from secure water treatment (SWaT) an operational water treatment testbed. These works usually focus solely on sensory dataset and omit the analysis of network dataset, or they focus on network information and omit sensory data. In this paper we work on both datasets. We have created IDSs using five traditional machine learning techniques, decision tree, support vector machine (SVM), random forest, naïve Bayes, and artificial neural network along with two deep methods, deep neural network, and convolutional neural network. We experimented with IDSs, on three different datasets obtained from SWaT, including network data, sensory data, and Modbus data. The accuracies of proposed methods show higher values on all datasets especially on sensory (99.9%) and Modbus data (95%) and superiority of random forest and deep learning methods compared to others.
40
Yao, Bo. "Walmart Sales Prediction Based on Decision Tree, Random Forest, and K Neighbors Regressor." Highlights in Business, Economics and Management 5 (February 16, 2023): 330–35. http://dx.doi.org/10.54097/hbem.v5i.5100.
Full textAbstract:
Sales forecasting is a very important research direction in the business and academic fields, and sales forecasting methods are also in full bloom, such as time series model, machine learning model and deep neural network model. This paper will use three machine learning models: Decision Tree Regressor, Random Forest Regressor, and K Neighbors Regressor to predict Walmart Recruiting - Store Sales data. Using correlation, mean absolute error, and mean square error to evaluate the prediction results of these three models, it is found that the prediction effect of Random Forest Registrar performs the best of these three models. The R2 value between the predicted sales volume of Random Forest Regressor and the sales volume of the test set is 0.937, the average absolute error is 1937.810, and the mean square error is 32993323.634. Therefore, Walmart can use Random Forest Regressor when forecasting the weekly sales of its own stores. At the same time, this paper provides a good model reference value (especially Random Forest Regressor) for other industries when researching the sales forecast, as well as methods for evaluating different model predictions. Overall, these results shed light on guiding further exploration of Sales forecasts for supermarkets.
41
Shri, Udayshankar B., R. Singh Veeraj, P. Sampras, and Dhage Aryan. "Fake Job Post Prediction Using Data Mining." Journal Of Scientific Research And Technology (JSRT) 1, no. 2 (2023): 39–47. https://doi.org/10.5281/zenodo.7954261.
Full textAbstract:
The proliferation of online job boards is a testament to the ease with which new positions may be publicized in today's connected society. As a result, the problem of predicting fraudulent job postings will be of paramount importance. Predicting the outcome of a bogus job posting is a challenging classification assignment, similar to many others. In order to determine if a job posting is genuine or not, this study proposes using a variety of data mining methods and classification algorithms, including KNN, decision tree, support vector machine, naive bayes classifier, random forest classifier, multilayer perceptron, and deep neural network. The Employment Scam Aegean Dataset (EMSCAD) was used for our experiments; it consists of 18000 data points. Using a deep neural network as a classifier yields excellent results in this setting. This classifier is a deep neural network with three thick layers. Classification accuracy (DNN) for identifying fake job postings is roughly 98% thanks to the trained classifier.
42
Tang, Chaofei, Nurbol Luktarhan, and Yuxin Zhao. "SAAE-DNN: Deep Learning Method on Intrusion Detection." Symmetry 12, no. 10 (2020): 1695. http://dx.doi.org/10.3390/sym12101695.
Full textAbstract:
Intrusion detection system (IDS) plays a significant role in preventing network attacks and plays a vital role in the field of national security. At present, the existing intrusion detection methods are generally based on traditional machine learning models, such as random forest and decision tree, but they rely heavily on artificial feature extraction and have relatively low accuracy. To solve the problems of feature extraction and low detection accuracy in intrusion detection, an intrusion detection model SAAE-DNN, based on stacked autoencoder (SAE), attention mechanism and deep neural network (DNN), is proposed. The SAE represents data with a latent layer, and the attention mechanism enables the network to obtain the key features of intrusion detection. The trained SAAE encoder can not only automatically extract features, but also initialize the weights of DNN potential layers to improve the detection accuracy of DNN. We evaluate the performance of SAAE-DNN in binary-classification and multi-classification on an NSL-KDD dataset. The SAAE-DNN model can detect normally and attack symmetrically, with an accuracy of 87.74% and 82.14% (binary-classification and multi-classification), which is higher than that of machine learning methods such as random forest and decision tree. The experimental results show that the model has a better performance than other comparison methods.
43
Bello, Rotimi-Williams, Zidiegha Seiyaboh, Daniel A. Olubummo, and Abdullah Zawawi Talib. "Classification of Dataset Using Deep Belief Networks Clustering Method." International Journal of Advanced Trends in Computer Science and Engineering 9, no. 3 (2020): 2856–60. https://doi.org/10.30534/ijatcse/2020/57932020.
Full textAbstract:
ABSTRACT Dataset in large collection involves considerable handling in its analysis especially when it is being employed in classification problems that involve big data. Due to the technology development, the manner and approach in which this dataset is being manipulated for classification purposes differ not only in one respect but in many respects with different uncorrelated results which sometimes make prediction inaccurate. By definition, classification is the act of arranging objects into classes or categories of the same type; these objects can be huge or otherwise, and to manually classify them will be a herculean task. The basic reason for classification is to punctiliously predict the class for each case in the dataset using class label. Notable classification, clustering and regression methods are support vector machines, neural networks, random forest, k-nearest neighbor and decision trees. The conventional clustering method that is widely employed to classification problems cannot handle the weight associated problem which characterized the transmission of neurons from layer to layer within the network. Employed in this work for the classification and clustering resolution is deep belief networks clustering method. The neural network architecture and loss function popularly employ in deep learning are considered for transforming the input data to clustering-friendly feature representation.
44
Bhavekar, Girish Shrikrushnarao, Pratiksha Vasantrao Chafle, Agam Das Goswami, et al. "Hybrid approach to medical decision-making: prediction of heart disease with artificial neural network." Bulletin of Electrical Engineering and Informatics 13, no. 6 (2024): 4124–33. http://dx.doi.org/10.11591/eei.v13i6.5583.
Full textAbstract:
Heart disease prediction is important in today’s world because it helps to reduce the unpredictable death rate of patients, and cardiac diseases are considered one of the most serious diseases affecting people. Hence, in this paper, a heart disease prediction model is designed for effective prediction of heart diseases by means of machine learning (ML) and deep learning (DL). This prediction uses the proposed method of an artificial neutral network and the Chi2 feature selection method applied to determine which features from the dataset were suitable for prediction. The proposed methodology uses classifiers like support vector machines (SVM), Naive Bayes (NB), logistic regression (LR), random forest (RF), and artificial neural networks (ANN). Python was used to conduct the study that assessed the ANN system proposal with the Cleveland heart disease dataset at the University of California (UCI). Compared to other algorithms, the model achieves an accuracy of 97.64% and takes 0.49 seconds to execute, making it superior in predicting heart disease.
45
Sulthan, M. Burhanis, Imam Wahyudi, and Luluk Suhartini. "Analisis Sentimen Pada Bencana Alam Menggunakan Deep Neural Network dan Information Gain." Jurnal Aplikasi Teknologi Informasi dan Manajemen (JATIM) 2, no. 2 (2021): 65–71. http://dx.doi.org/10.31102/jatim.v2i2.1273.
Full textAbstract:
Kemajuan teknologi informasi dan komunikasi membuat informasi terkait bencana alam menjadi lebih cepat tersebar, salah satu soial media yang banyak digunakan yaitu twitter. Pada penelitian ini mengklasifikasikan teks terkait analisis sentimen terhadap bencana alam yang terjadi. Metode klasifikasi yang digunakan adalah Deep Neural Network (DNN),. Jadi untuk mempercepat proses klasifikasi digunakan teknik seleksi fitur yaitu Information Gain (IG) untuk memilih fitur-fitur yang terbaik dari hasil ekstrasi. Kemudian evaluasi dan validasi dilakukan untuk mengetahui hasil kinerja klasifikasi. Digunakan confusion matrix dan 10 fold validasi sebagai proses evaluasi dan validasi didalam penelitian ini. Pada penelitian ini menggunakan beberapa metode yaitu Naïve bayes, Random Forest, Decision Tree dan Support Vector Machine. Hasil akurasi dari metode Deep Neural Network dengan Information Gain lebih besar dari metode yang lain.
46
El-Amin, Mohamed F., Budoor Alwated, and Hussein A. Hoteit. "Machine Learning Prediction of Nanoparticle Transport with Two-Phase Flow in Porous Media." Energies 16, no. 2 (2023): 678. http://dx.doi.org/10.3390/en16020678.
Full textAbstract:
Reservoir simulation is a time-consuming procedure that requires a deep understanding of complex fluid flow processes as well as the numerical solution of nonlinear partial differential equations. Machine learning algorithms have made significant progress in modeling flow problems in reservoir engineering. This study employs machine learning methods such as random forest, decision trees, gradient boosting regression, and artificial neural networks to forecast nanoparticle transport with the two-phase flow in porous media. Due to the shortage of data on nanoparticle transport in porous media, this work creates artificial datasets using a mathematical model. It predicts nanoparticle transport behavior using machine learning techniques, including gradient boosting regression, decision trees, random forests, and artificial neural networks. Utilizing the scikit-learn toolkit, strategies for data preprocessing, correlation, and feature importance are addressed. Furthermore, the GridSearchCV algorithm is used to optimize hyperparameter tuning. The mean absolute error, R-squared correlation, mean squared error, and root means square error are used to assess the models. The ANN model has the best performance in forecasting the transport of nanoparticles in porous media, according to the results.
47
MacMichael, Duncan, and Dong Si. "Machine Learning Classification of Tree Cover Type and Application to Forest Management." International Journal of Multimedia Data Engineering and Management 9, no. 1 (2018): 1–21. http://dx.doi.org/10.4018/ijmdem.2018010101.
Full textAbstract:
This article is driven by three goals. The first is to use machine learning to predict tree cover types, helping to address current challenges faced by U.S. forest management agencies. The second is to bring previous research in the area up-to-date, owing to a lack of development over time. The third is to improve on previous results with new data analysis, higher accuracy, and higher reliability. A Deep Neural Network (DNN) was constructed and compared with three baseline traditional machine learning models: Naïve Bayes, Decision Tree, and K-Nearest Neighbor (KNN). The neural network model achieved 91.55% accuracy while the best performing traditional classifier, K-Nearest Neighbor, managed 74.61%. In addition, the neural network model performed 20.97% better than the past neural networks, which illustrates both advances in machine learning algorithms, as well as improved accuracy high enough to apply practically to forest management issues. Using the techniques outlined in this article, agencies can cost-efficiently and quickly predict tree cover type and expedite natural resource inventorying.
48
SivaKumar, Ramagowni, Hrushikesh Reddy K. V, Charan Reddy S, and Surya Prakash Reddy B. "A Comprehensive Multi-Model Approach to Kidney Stone Detection using Deep Learning." International Journal for Research in Applied Science and Engineering Technology 13, no. 4 (2025): 3699–702. https://doi.org/10.22214/ijraset.2025.69067.
Full textAbstract:
Abstract: This work presents an end-to-end deep learning-based kidney stone detection system based on a combination of Convolutional Neural Networks (CNNs) for ultrasound image processing and Random Forest for clinical data classification. The combination of the two models has an enormously improved diagnostic accuracy, reducing misclassification rates. With the use of a fusion model, the system effectively facilitates decision-making with accurate and autonomous detection. This study illustrates the advantages of multi-modal learning in medical diagnosis and aims to support healthcare professionals in early detection and treatment planning.
49
Davoulos, George, Iro Lalakou, and Ioannis Hatzilygeroudis. "From Single to Deep Learning and Hybrid Ensemble Models for Recognition of Dog Motion States." Electronics 14, no. 10 (2025): 1924. https://doi.org/10.3390/electronics14101924.
Full textAbstract:
Dog activities recognition, especially dog motion status recognition, is an active research area. Although several machine learning and deep learning approaches have been used for dog motion states recognition, the use of ensemble learning methods is rather missing, as well as a comparison with deep learning ones. This paper focuses on the use of deep learning neural networks and ensemble classifiers in recognizing dog motion states and their comparison. A dataset from the Kaggle database, which includes measures by accelerometer and gyroscope and concerns seven dog motion states (galloping, sitting, standing, trotting, walking, lying on chest, and sniffing), was used for our experiments. Gaussian Naive Bayes, Decision Tree, k-Nearest Neighbors (kNN), Random Forest, a Bagging Tree-Based Classifier, a Stacking Classifier, a Compound Stacking Model (CSM), a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN), and a Hybrid Cascading Model (HCM) were used in our experiments. Results showed a 1.78% superiority in accuracy (92.64% vs. 90.86%) of deep learning (RNN) vs. stacking (CSTAM) best classifier, but at the cost of larger complexity and training time for the deep learning classifier, which makes ensemble techniques still attractive. Finally, HCM gave the best result (96.82% accuracy).
50
Rahman, Senjuti, Mehedi Hasan, and Ajay Krishno Sarkar. "Prediction of Brain Stroke using Machine Learning Algorithms and Deep Neural Network Techniques." European Journal of Electrical Engineering and Computer Science 7, no. 1 (2023): 23–30. http://dx.doi.org/10.24018/ejece.2023.7.1.483.
Full textAbstract:
The brain is the human body's primary upper organ. Stroke is a medical disorder in which the blood arteries in the brain are ruptured, causing damage to the brain. When the supply of blood and other nutrients to the brain is interrupted, symptoms might develop. Stroke is considered as medical urgent situation and can cause long-term neurological damage, complications and often death. The World Health Organization (WHO) claims that stroke is the leading cause of death and disability worldwide. Early detection of the numerous stroke warning symptoms can lessen the stroke's severity. The main objective of this study is to forecast the possibility of a brain stroke occurring at an early stage using deep learning and machine learning techniques. To gauge the effectiveness of the algorithm, a reliable dataset for stroke prediction was taken from the Kaggle website. Several classification models, including Extreme Gradient Boosting (XGBoost), Ada Boost, Light Gradient Boosting Machine, Random Forest, Decision Tree, Logistic Regression, K Neighbors, SVM - Linear Kernel, Naive Bayes, and deep neural networks (3-layer and 4-layer ANN) were successfully used in this study for classification tasks. The Random Forest classifier has 99% classification accuracy, which was the highest (among the machine learning classifiers). The three layer deep neural network (4-Layer ANN) has produced a higher accuracy of 92.39% than the three-layer ANN method utilizing the selected features as input. The research's findings showed that machine learning techniques outperformed deep neural networks.