Academic literature on the topic 'Shapley Additive Explanations'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the lists of relevant articles, books, theses, conference reports, and other scholarly sources on the topic 'Shapley Additive Explanations.'
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.
Journal articles on the topic "Shapley Additive Explanations"
1
Vega García, María, and José L. Aznarte. "Shapley additive explanations for NO2 forecasting." Ecological Informatics 56 (March 2020): 101039. http://dx.doi.org/10.1016/j.ecoinf.2019.101039.
Full text
2
Antwarg, Liat, Ronnie Mindlin Miller, Bracha Shapira, and Lior Rokach. "Explaining anomalies detected by autoencoders using Shapley Additive Explanations." Expert Systems with Applications 186 (December 2021): 115736. http://dx.doi.org/10.1016/j.eswa.2021.115736.
Full text
3
Ogami, Chika, Yasuhiro Tsuji, Hiroto Seki, Hideaki Kawano, Hideto To, Yoshiaki Matsumoto, and Hiroyuki Hosono. "An artificial neural network−pharmacokinetic model and its interpretation using Shapley additive explanations." CPT: Pharmacometrics & Systems Pharmacology 10, no. 7 (May 27, 2021): 760–68. http://dx.doi.org/10.1002/psp4.12643.
Full text
4
Tideman, Leonoor E. M., Lukasz G. Migas, Katerina V. Djambazova, Nathan Heath Patterson, Richard M. Caprioli, Jeffrey M. Spraggins, and Raf Van de Plas. "Automated biomarker candidate discovery in imaging mass spectrometry data through spatially localized Shapley additive explanations." Analytica Chimica Acta 1177 (September 2021): 338522. http://dx.doi.org/10.1016/j.aca.2021.338522.
Full text
5
Wieland, Ralf, Tobia Lakes, and Claas Nendel. "Using Shapley additive explanations to interpret extreme gradient boosting predictions of grassland degradation in Xilingol, China." Geoscientific Model Development 14, no. 3 (March 16, 2021): 1493–510. http://dx.doi.org/10.5194/gmd-14-1493-2021.
Full textAbstract:
Abstract. Machine learning (ML) and data-driven approaches are increasingly used in many research areas. Extreme gradient boosting (XGBoost) is a tree boosting method that has evolved into a state-of-the-art approach for many ML challenges. However, it has rarely been used in simulations of land use change so far. Xilingol, a typical region for research on serious grassland degradation and its drivers, was selected as a case study to test whether XGBoost can provide alternative insights that conventional land-use models are unable to generate. A set of 20 drivers was analysed using XGBoost, involving four alternative sampling strategies, and SHAP (Shapley additive explanations) to interpret the results of the purely data-driven approach. The results indicated that, with three of the sampling strategies (over-balanced, balanced, and imbalanced), XGBoost achieved similar and robust simulation results. SHAP values were useful for analysing the complex relationship between the different drivers of grassland degradation. Four drivers accounted for 99 % of the grassland degradation dynamics in Xilingol. These four drivers were spatially allocated, and a risk map of further degradation was produced. The limitations of using XGBoost to predict future land-use change are discussed.
6
Knapič, Samanta, Avleen Malhi, Rohit Saluja, and Kary Främling. "Explainable Artificial Intelligence for Human Decision Support System in the Medical Domain." Machine Learning and Knowledge Extraction 3, no. 3 (September 19, 2021): 740–70. http://dx.doi.org/10.3390/make3030037.
Full textAbstract:
In this paper, we present the potential of Explainable Artificial Intelligence methods for decision support in medical image analysis scenarios. Using three types of explainable methods applied to the same medical image data set, we aimed to improve the comprehensibility of the decisions provided by the Convolutional Neural Network (CNN). In vivo gastral images obtained by a video capsule endoscopy (VCE) were the subject of visual explanations, with the goal of increasing health professionals’ trust in black-box predictions. We implemented two post hoc interpretable machine learning methods, called Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive exPlanations (SHAP), and an alternative explanation approach, the Contextual Importance and Utility (CIU) method. The produced explanations were assessed by human evaluation. We conducted three user studies based on explanations provided by LIME, SHAP and CIU. Users from different non-medical backgrounds carried out a series of tests in a web-based survey setting and stated their experience and understanding of the given explanations. Three user groups (n = 20, 20, 20) with three distinct forms of explanations were quantitatively analyzed. We found that, as hypothesized, the CIU-explainable method performed better than both LIME and SHAP methods in terms of improving support for human decision-making and being more transparent and thus understandable to users. Additionally, CIU outperformed LIME and SHAP by generating explanations more rapidly. Our findings suggest that there are notable differences in human decision-making between various explanation support settings. In line with that, we present three potential explainable methods that, with future improvements in implementation, can be generalized to different medical data sets and can provide effective decision support to medical experts.
7
Mangalathu, Sujith, Seong-Hoon Hwang, and Jong-Su Jeon. "Failure mode and effects analysis of RC members based on machine-learning-based SHapley Additive exPlanations (SHAP) approach." Engineering Structures 219 (September 2020): 110927. http://dx.doi.org/10.1016/j.engstruct.2020.110927.
Full text
8
Pokharel, Sugam, Pradip Sah, and Deepak Ganta. "Improved Prediction of Total Energy Consumption and Feature Analysis in Electric Vehicles Using Machine Learning and Shapley Additive Explanations Method." World Electric Vehicle Journal 12, no. 3 (June 29, 2021): 94. http://dx.doi.org/10.3390/wevj12030094.
Full textAbstract:
Electric vehicles (EVs) have emerged as the green energy alternative for conventional vehicles. While various governments promote EVs, people feel “range anxiety” because of their limited driving range or charge capacity. A limited number of charging stations are available, which results in a strong demand for predicting energy consumed by EVs. In this paper, machine learning (ML) models such as multiple linear regression (MLR), extreme gradient boosting (XGBoost), and support vector regression (SVR) were used to investigate the total energy consumption (TEC) by the EVs. The independent variables used for the study include changing real-life situations or external parameters, such as trip distance, tire type, driving style, power, odometer reading, EV model, city, motorway, country roads, air conditioning, and park heating. We compared the ML models’ performance along with the error analysis. A pairwise correlation study showed that trip distance has a high correlation coefficient (0.87) with TEC. XGBoost had better prediction accuracy (~92%) or R2 (0.92). Trip distance, power, heating, and odometer reading were the most important features influencing the TEC, identified using the shapley additive explanations method.
9
Machado Poletti Valle, Luis Fernando, Camille Avestruz, David J. Barnes, Arya Farahi, Erwin T. Lau, and Daisuke Nagai. "shaping the gas: understanding gas shapes in dark matter haloes with interpretable machine learning." Monthly Notices of the Royal Astronomical Society 507, no. 1 (August 6, 2021): 1468–84. http://dx.doi.org/10.1093/mnras/stab2252.
Full textAbstract:
ABSTRACT The non-spherical shapes of dark matter and gas distributions introduce systematic uncertainties that affect observable–mass relations and selection functions of galaxy groups and clusters. However, the triaxial gas distributions depend on the non-linear physical processes of halo formation histories and baryonic physics, which are challenging to model accurately. In this study, we explore a machine learning approach for modelling the dependence of gas shapes on dark matter and baryonic properties. With data from the IllustrisTNG hydrodynamical cosmological simulations, we develop a machine learning pipeline that applies XGBoost, an implementation of gradient-boosted decision trees, to predict radial profiles of gas shapes from halo properties. We show that XGBoost models can accurately predict gas shape profiles in dark matter haloes. We also explore model interpretability with the SHapley Additive exPlanations (shap), a method that identifies the most predictive properties at different halo radii. We find that baryonic properties best predict gas shapes in halo cores, whereas dark matter shapes are the main predictors in the halo outskirts. This work demonstrates the power of interpretable machine learning in modelling observable properties of dark matter haloes in the era of multiwavelength cosmological surveys.
10
Manikis, Georgios C., Georgios S. Ioannidis, Loizos Siakallis, Katerina Nikiforaki, Michael Iv, Diana Vozlic, Katarina Surlan-Popovic, Max Wintermark, Sotirios Bisdas, and Kostas Marias. "Multicenter DSC–MRI-Based Radiomics Predict IDH Mutation in Gliomas." Cancers 13, no. 16 (August 5, 2021): 3965. http://dx.doi.org/10.3390/cancers13163965.
Full textAbstract:
To address the current lack of dynamic susceptibility contrast magnetic resonance imaging (DSC–MRI)-based radiomics to predict isocitrate dehydrogenase (IDH) mutations in gliomas, we present a multicenter study that featured an independent exploratory set for radiomics model development and external validation using two independent cohorts. The maximum performance of the IDH mutation status prediction on the validation set had an accuracy of 0.544 (Cohen’s kappa: 0.145, F1-score: 0.415, area under the curve-AUC: 0.639, sensitivity: 0.733, specificity: 0.491), which significantly improved to an accuracy of 0.706 (Cohen’s kappa: 0.282, F1-score: 0.474, AUC: 0.667, sensitivity: 0.6, specificity: 0.736) when dynamic-based standardization of the images was performed prior to the radiomics. Model explainability using local interpretable model-agnostic explanations (LIME) and Shapley additive explanations (SHAP) revealed potential intuitive correlations between the IDH–wildtype increased heterogeneity and the texture complexity. These results strengthened our hypothesis that DSC–MRI radiogenomics in gliomas hold the potential to provide increased predictive performance from models that generalize well and provide understandable patterns between IDH mutation status and the extracted features toward enabling the clinical translation of radiogenomics in neuro-oncology.
Dissertations / Theses on the topic "Shapley Additive Explanations"
1
Lattouf, Mouzeina. "Assessment of Predictive Models for Improving Default Settings in Streaming Services." Thesis, KTH, Skolan för kemi, bioteknologi och hälsa (CBH), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-284482.
Full textAbstract:
Streaming services provide different settings where customers can choose a sound and video quality based on personal preference. The majority of users never make an active choice; instead, they get a default quality setting which is chosen automatically for them based on some parameters, like internet connection quality. This thesis explores personalising the default audio setting, intending to improve the user experience. It achieves this by leveraging machine learning trained on the fraction of users that have made active choices in changing the quality setting. The assumption that user similarity in users who make an active choice can be leveraged to impact user experience was the idea behind this thesis work. It was issued to study which type of data from different categories: demographic, product and consumption is most predictive of a user's taste in sound quality. A case study was conducted to achieve the goals for this thesis. Five predictive model prototypes were trained, evaluated, compared and analysed using two different algorithms: XGBoost and Logistic Regression, and targeting two regions: Sweden and Brazil. Feature importance analysis was conducted using SHapley Additive exPlanations(SHAP), a unified framework for interpreting predictions with a game theoretic approach, and by measuring coefficient weights to determine the most predictive features. Besides exploring the feature impact, the thesis also answers how reasonable it is to generalise these models to non-selecting users by performing hypothesis testing. The project also covered bias analysis between users with and without active quality settings and how that affects the models. The models with XGBoost had higher performance. The results showed that demographic and product data had a higher impact on model predictions in both regions. Although, different regions did not have the same data features as most predictive, so there were differences observed in feature importance between regions and also between platforms. The results of hypothesis testing did not indicate a valid reason to consider the models to work for non-selective users. However, the method is negatively affected by other factors such as small changes in big datasets that impact the statistical significance. Data bias in some data features was found, which indicated a correlation but not the causation behind the patterns. The results of this thesis additionally show how machine learning can improve user experience in regards to default sound quality settings, by leveraging models on user similarity in users who have changed the sound quality to the most suitable for them.Streamingtjänster erbjuder olika inställningar där kunderna kan välja ljud- och videokvalitet baserat på personliga preferenser. Majoriteten av användarna gör aldrig ett aktivt val; de tilldelas istället en standardkvalitetsinställning som väljs automatiskt baserat på vissa parametrar, som internetanslutningskvalitet. Denna avhandling undersöker anpassning av standardljudinställningen, med avsikt att förbättra användarupplevelsen. Detta uppnås genom att tillämpa maskininlärning på den andel användare som har aktivt ändrat kvalitetsinställningen. Antagandet att användarlikhet hos användare som gör ett aktivt val kan utnyttjas för att påverka användarupplevelsen var tanken bakom detta examensarbete. Det utfärdades för att studera vilken typ av data från olika kategorier: demografi, produkt och konsumtion är mest förutsägande för användarens smak i ljudkvalitet. En fallstudie genomfördes för att uppnå målen för denna avhandling. Fem prediktiva modellprototyper tränades, utvärderades, jämfördes och analyserades med två olika algoritmer: XGBoost och Logistisk Regression, och inriktade på två regioner: Sverige och Brasilien. Analys av funktionsvikt genomfördes med SHapley Additive exPlanations (SHAP), en enhetlig ram för att tolka förutsägelser med en spelteoretisk metod, och genom att mäta koefficientvikter för att bestämma de mest prediktiva funktionerna. Förutom att utforska funktionens påverkan, svarar avhandlingen också på hur rimligt det är att generalisera dessa modeller för icke-selektiva användare genom att utföra hypotesprövning. Projektet omfattade också biasanalys mellan användare med och utan aktiva kvalitetsinställningar och hur det påverkar modellerna. Modellerna med XGBoost hade högre prestanda. Resultaten visade att demografisk data och produktdata hade en högre inverkan på modellförutsägelser i båda regionerna. Däremot hade olika regioner inte samma datafunktioner som mest prediktiva, skillnader observerades i funktionsvikt mellan regioner och även mellan plattformar. Resultaten av hypotesprövningen indikerade inte på vägande anledning för att anse att modellerna skulle fungera för icke-selektiva användare. Däremot har metoden påverkats negativt av andra faktorer som små förändringar i stora datamängder som påverkar den statistiska signifikansen. Data bias hittades i vissa datafunktioner, vilket indikerade en korrelation men inte orsaken bakom mönstren. Resultaten av denna avhandling visar dessutom hur maskininlärning kan förbättra användarupplevelsen när det gäller standardinställningar för ljudkvalitet, genom att utnyttja modeller för användarlikhet hos användare som har ändrat ljudkvaliteten till det mest lämpliga för dem.
2
Norrie, Christian. "Explainable AI techniques for sepsis diagnosis : Evaluating LIME and SHAP through a user study." Thesis, Högskolan i Skövde, Institutionen för informationsteknologi, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:his:diva-19845.
Full textAbstract:
Articial intelligence has had a large impact on many industries and transformed some domains quite radically. There is tremendous potential in applying AI to the eld of medical diagnostics. A major issue with applying these techniques to some domains is an inability for AI models to provide an explanation or justication for their predictions. This creates a problem wherein a user may not trust an AI prediction, or there are legal requirements for justifying decisions that are not met. This thesis overviews how two explainable AI techniques (Shapley Additive Explanations and Local Interpretable Model-Agnostic Explanations) can establish a degree of trust for the user in the medical diagnostics eld. These techniques are evaluated through a user study. User study results suggest that supplementing classications or predictions with a post-hoc visualization increases interpretability by a small margin. Further investigation and research utilizing a user study surveyor interview is suggested to increase interpretability and explainability of machine learning results.
3
Saluja, Rohit. "Interpreting Multivariate Time Series for an Organization Health Platform." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-289465.
Full textAbstract:
Machine learning-based systems are rapidly becoming popular because it has been realized that machines are more efficient and effective than humans at performing certain tasks. Although machine learning algorithms are extremely popular, they are also very literal and undeviating. This has led to a huge research surge in the field of interpretability in machine learning to ensure that machine learning models are reliable, fair, and can be held liable for their decision-making process. Moreover, in most real-world problems just making predictions using machine learning algorithms only solves the problem partially. Time series is one of the most popular and important data types because of its dominant presence in the fields of business, economics, and engineering. Despite this, interpretability in time series is still relatively unexplored as compared to tabular, text, and image data. With the growing research in the field of interpretability in machine learning, there is also a pressing need to be able to quantify the quality of explanations produced after interpreting machine learning models. Due to this reason, evaluation of interpretability is extremely important. The evaluation of interpretability for models built on time series seems completely unexplored in research circles. This thesis work focused on achieving and evaluating model agnostic interpretability in a time series forecasting problem. The use case discussed in this thesis work focused on finding a solution to a problem faced by a digital consultancy company. The digital consultancy wants to take a data-driven approach to understand the effect of various sales related activities in the company on the sales deals closed by the company. The solution involved framing the problem as a time series forecasting problem to predict the sales deals and interpreting the underlying forecasting model. The interpretability was achieved using two novel model agnostic interpretability techniques, Local interpretable model- agnostic explanations (LIME) and Shapley additive explanations (SHAP). The explanations produced after achieving interpretability were evaluated using human evaluation of interpretability. The results of the human evaluation studies clearly indicate that the explanations produced by LIME and SHAP greatly helped lay humans in understanding the predictions made by the machine learning model. The human evaluation study results also indicated that LIME and SHAP explanations were almost equally understandable with LIME performing better but with a very small margin. The work done during this project can easily be extended to any time series forecasting or classification scenario for achieving and evaluating interpretability. Furthermore, this work can offer a very good framework for achieving and evaluating interpretability in any machine learning-based regression or classification problem.Maskininlärningsbaserade system blir snabbt populära eftersom man har insett att maskiner är effektivare än människor när det gäller att utföra vissa uppgifter. Även om maskininlärningsalgoritmer är extremt populära, är de också mycket bokstavliga. Detta har lett till en enorm forskningsökning inom området tolkbarhet i maskininlärning för att säkerställa att maskininlärningsmodeller är tillförlitliga, rättvisa och kan hållas ansvariga för deras beslutsprocess. Dessutom löser problemet i de flesta verkliga problem bara att göra förutsägelser med maskininlärningsalgoritmer bara delvis. Tidsserier är en av de mest populära och viktiga datatyperna på grund av dess dominerande närvaro inom affärsverksamhet, ekonomi och teknik. Trots detta är tolkningsförmågan i tidsserier fortfarande relativt outforskad jämfört med tabell-, text- och bilddata. Med den växande forskningen inom området tolkbarhet inom maskininlärning finns det också ett stort behov av att kunna kvantifiera kvaliteten på förklaringar som produceras efter tolkning av maskininlärningsmodeller. Av denna anledning är utvärdering av tolkbarhet extremt viktig. Utvärderingen av tolkbarhet för modeller som bygger på tidsserier verkar helt outforskad i forskarkretsar. Detta uppsatsarbete fokuserar på att uppnå och utvärdera agnostisk modelltolkbarhet i ett tidsserieprognosproblem. Fokus ligger i att hitta lösningen på ett problem som ett digitalt konsultföretag står inför som användningsfall. Det digitala konsultföretaget vill använda en datadriven metod för att förstå effekten av olika försäljningsrelaterade aktiviteter i företaget på de försäljningsavtal som företaget stänger. Lösningen innebar att inrama problemet som ett tidsserieprognosproblem för att förutsäga försäljningsavtalen och tolka den underliggande prognosmodellen. Tolkningsförmågan uppnåddes med hjälp av två nya tekniker för agnostisk tolkbarhet, lokala tolkbara modellagnostiska förklaringar (LIME) och Shapley additiva förklaringar (SHAP). Förklaringarna som producerats efter att ha uppnått tolkbarhet utvärderades med hjälp av mänsklig utvärdering av tolkbarhet. Resultaten av de mänskliga utvärderingsstudierna visar tydligt att de förklaringar som produceras av LIME och SHAP starkt hjälpte människor att förstå förutsägelserna från maskininlärningsmodellen. De mänskliga utvärderingsstudieresultaten visade också att LIME- och SHAP-förklaringar var nästan lika förståeliga med LIME som presterade bättre men med en mycket liten marginal. Arbetet som utförts under detta projekt kan enkelt utvidgas till alla tidsserieprognoser eller klassificeringsscenarier för att uppnå och utvärdera tolkbarhet. Dessutom kan detta arbete erbjuda en mycket bra ram för att uppnå och utvärdera tolkbarhet i alla maskininlärningsbaserade regressions- eller klassificeringsproblem.
4
"Predicting and Interpreting Students Performance using Supervised Learning and Shapley Additive Explanations." Master's thesis, 2019. http://hdl.handle.net/2286/R.I.53452.
Full textAbstract:
abstract: Due to large data resources generated by online educational applications, Educational Data Mining (EDM) has improved learning effects in different ways: Students Visualization, Recommendations for students, Students Modeling, Grouping Students, etc. A lot of programming assignments have the features like automating submissions, examining the test cases to verify the correctness, but limited studies compared different statistical techniques with latest frameworks, and interpreted models in a unified approach.
In this thesis, several data mining algorithms have been applied to analyze students’ code assignment submission data from a real classroom study. The goal of this work is to explore
and predict students’ performances. Multiple machine learning models and the model accuracy were evaluated based on the Shapley Additive Explanation.
The Cross-Validation shows the Gradient Boosting Decision Tree has the best precision 85.93% with average 82.90%. Features like Component grade, Due Date, Submission Times have higher impact than others. Baseline model received lower precision due to lack of non-linear fitting.Dissertation/Thesis
Masters Thesis Computer Science 2019
Books on the topic "Shapley Additive Explanations"
1
Kaplan, David M. Integrating Mind and Brain Science. Oxford University Press, 2018. http://dx.doi.org/10.1093/oso/9780199685509.003.0001.
Full textAbstract:
There is growing appreciation that understanding the complex relationship between neuroscience and psychological science is of fundamental importance to achieving progress across these scientific domains. One primary strategy for addressing this issue centers around understanding the nature of explanation in these different domains. This chapter provides a field guide to some of the core topics that have shaped and continue to influence the debate about explanation and integration across the mind and brain sciences. In addition to surveying the overall intellectual terrain, it also introduces the main proposals defended in the individual chapters included in the volume and highlights important similarities and differences between them.
2
Corbett, Jack, and Wouter Veenendaal. The Small State Challenge. Oxford University Press, 2018. http://dx.doi.org/10.1093/oso/9780198796718.003.0001.
Full textAbstract:
Chapter 1 introduces the main arguments of the book; outlines the approach, method, and data; defines key terms; and provides a chapter outline. Global theories of democratization have systematically excluded small states, which make up roughly 20 per cent of countries. These cases debunk mainstream theories of why democratization succeeds or fails. This book brings small states into the comparative politics fold for the first time. It is organized thematically, with each chapter tackling one of the main theories from the democratization literature. Different types of data are examined—case studies and other documentary evidence, interviews and observation. Following an abductive approach, in addition to examining the veracity of existing theory, each chapter is also used to build an explanation of how democracy is practiced in small states. Specifically, we highlight how small state politics is shaped by personalization and informal politics, rather than formal institutional design.
Book chapters on the topic "Shapley Additive Explanations"
1
Thanh-Hai, Nguyen, Toan Bao Tran, Nhi Yen Kim Phan, Tran Thanh Dien, and Nguyen Thai-Nghe. "Feature Selection Based on Shapley Additive Explanations on Metagenomic Data for Colorectal Cancer Diagnosis." In Soft Computing: Biomedical and Related Applications, 69–80. Cham: Springer International Publishing, 2021. http://dx.doi.org/10.1007/978-3-030-76620-7_6.
Full text
2
Jovanović, Gordana, Marijana Matek Sarić, Snježana Herceg Romanić, Svetlana Stanišić, Marija Mitrović Dankulov, Aleksandar Popović, and Mirjana Perišić. "Patterns of PCB-138 Occurrence in the Breast Milk of Primiparae and Multiparae Using SHapley Additive exPlanations Analysis." In Artificial Intelligence: Theory and Applications, 191–206. Cham: Springer International Publishing, 2021. http://dx.doi.org/10.1007/978-3-030-72711-6_11.
Full text
3
Biecek, Przemyslaw, and Tomasz Burzykowski. "Shapley Additive Explanations (SHAP) for Average Attributions." In Explanatory Model Analysis, 95–106. Chapman and Hall/CRC, 2021. http://dx.doi.org/10.1201/9780429027192-10.
Full text
4
Nesaragi, Naimahmed, and Shivnarayan Patidar. "An Explainable Machine Learning Model for Early Prediction of Sepsis Using ICU Data." In Infectious Diseases and Sepsis [Working Title]. IntechOpen, 2021. http://dx.doi.org/10.5772/intechopen.98957.
Full textAbstract:
Early identification of individuals with sepsis is very useful in assisting clinical triage and decision-making, resulting in early intervention and improved outcomes. This study aims to develop an explainable machine learning model with the clinical interpretability to predict sepsis onset before 6 hours and validate with improved prediction risk power for every time interval since admission to the ICU. The retrospective observational cohort study is carried out using PhysioNet Challenge 2019 ICU data from three distinct hospital systems, viz. A, B, and C. Data from A and B were shared publicly for training and validation while sequestered data from all three cohorts were used for scoring. However, this study is limited only to publicly available training data. Training data contains 15,52,210 patient records of 40,336 ICU patients with up to 40 clinical variables (sourced for each hour of their ICU stay) divided into two datasets, based on hospital systems A and B. The clinical feature exploration and interpretation for early prediction of sepsis is achieved using the proposed framework, viz. the explainable Machine Learning model for Early Prediction of Sepsis (xMLEPS). A total of 85 features comprising the given 40 clinical variables augmented with 10 derived physiological features and 35 time-lag difference features are fed to xMLEPS for the said prediction task of sepsis onset. A ten-fold cross-validation scheme is employed wherein an optimal prediction risk threshold is searched for each of the 10 LightGBM models. These optimum threshold values are later used by the corresponding models to refine the predictive power in terms of utility score for the prediction of labels in each fold. The entire framework is designed via Bayesian optimization and trained with the resultant feature set of 85 features, yielding an average normalized utility score of 0.4214 and area under receiver operating characteristic curve of 0.8591 on publicly available training data. This study establish a practical and explainable sepsis onset prediction model for ICU data using applied ML approach, mainly gradient boosting. The study highlights the clinical significance of physiological inter-relations among the given and proposed clinical signs via feature importance and SHapley Additive exPlanations (SHAP) plots for visualized interpretation.
5
Hadley, David P. "Introduction." In The Rising Clamor, 1–12. University Press of Kentucky, 2019. http://dx.doi.org/10.5810/kentucky/9780813177373.003.0001.
Full textAbstract:
The introduction examines the overall questions animating the core work. Using Director of Central Intelligence William E. Colby’s explanation of how opinion shaped CIA activity, it explores how the CIA both was influenced by the press and sought to influence the press to shape the environment in which it operated. The introduction also explores the previous understandings of how the press and the CIA interacted and disputes a persistent theory originated by Deborah Davis that there existed a program called Mockingbird designed by the CIA to manipulate the press. It argues also that, in addition to Cold War–related activities, the CIA was interested in the press as a way to promote its reputation and establish its security within the national security bureaucracy.
6
Ramsey, Jeffry L. "Realism, Essentialism, and Intrinsic Properties: The Case of Molecular Shape." In Of Minds and Molecules. Oxford University Press, 2000. http://dx.doi.org/10.1093/oso/9780195128345.003.0015.
Full textAbstract:
Figure, or shape, has long been ensconced in modern philosophy as a primary or essential quality of matter. Descartes, Malebranche, Hobbes, and Boyle all apparently endorsed the Lockean claim that shape is “in Bodies whether we perceive them or no” (Locke, [1700] 1975, p. 140). In addition, most seventeenth-century philosophers endorsed the inference that because shape is primary, it is one of the “ultimate, irreducible explanatory principles” (Dijksterhuis, 1961, p. 433; cf. Ihde, 1964, p. 28). Locke has often been read in this way, and in Origins of Forms and Qualities, Boyle claims the “sensible qualities . . . are but the effects or consequents of the . . . primary affections of matter,” one of which is figure (quoted in Harré, 1964, p. 80). Little appears to have changed. Most analytic philosophers and realist-minded philosophers of science “would endorse a distinction between primary and secondary qualities” (Smith, 1990, p. 221). Campbell (1972, p. 219) endorses the claim that “shape, size and solidity are generally held to be primary,” even though he argues that “the philosophy of primary and secondary qualities” is confused. Mackie (1976, p. 18) discounts solidity but endorses spatial properties and motion as “basic” physical features of matter. Most philosophers also endorse the inference to the explanatory character of the primary qualities. Mackie (1976, p. 25) asserts spatial properties are “starting points of explanation.” Boyd (1989, pp. 10-11) claims “realists agree” that “the factors which govern the behavior . . . of substances are the fundamental properties of the insensible corpuscles of which they are composed.” As befits our current situation, explanation purportedly flows from spatial microstructure. A body “possesses a certain potential only because it actually possesses a certain property (e.g., its molecular structure)” (Lange, 1994, pp. 109-110). Even Putnam, who argues all properties are Lockean secondaries, claims powers “have an explanation . . . in the particular microstructure” of matter (Putnam, 1981, p. 58).
7
Mitchell, Graham. "A Shape to Die for?" In How Giraffes Work, 481–510. Oxford University Press, 2021. http://dx.doi.org/10.1093/oso/9780197571194.003.0018.
Full textAbstract:
The product of natural selection over at least 15 million years is the elongated, slender shape of giraffes that fits the natural habitat giraffes now occupy. What selection pressures operated to produce their shape? Their shape is partly the product of gravity and could have been an accidental by-product of selection for a large body mass and the protection from predation that large size brings, but the prevailing explanation is that their shape confers a browsing advantage. Preferred browse is concentrated at a height easily reached by giraffes but not by other browsers and natural selection would have favored those giraffes that could reach it. An alternative hypothesis is that their shape confers thermoregulatory benefits in addition to improved vigilance. Another hypothesis is that a long neck evolved to counter long legs allowing giraffes to drink surface water. An attractive hypothesis is that their shape is a product of ‘runaway’ sexual selection by females for males with long heavy necks, but analysis of this hypothesis has shown that the morphology of male and female giraffe does not differ. Nevertheless, all these possibilities could have contributed. A consequence of selection for their shape is over-specialization: giraffes seem to be inextricably dependent on a narrow diet, a diet that is subject to the vagaries of climate and competition for resources. The greatest threat to their survival is, therefore, their shape.
8
Pugh, Martin. "Islamophobia." In Britain and Islam, 244–71. Yale University Press, 2019. http://dx.doi.org/10.12987/yale/9780300234947.003.0011.
Full textAbstract:
This chapter discusses how, misled by Islamophobic propaganda, Britain and America were unable to come to terms with what they called ‘Islamism’. The origins of what is variously known as Islamism, Islamic fundamentalism, and radical Islamism lie in the 1960s, in the ideas of a handful of Muslims in Pakistan, Egypt, and Iran who believed that Muslims had been led astray from their religion by nationalist movements. Although some Muslims were critical of Western morality and politics, Islamism was not primarily anti-Western: it was essentially a reaction against what were widely seen as the corrupt, authoritarian, and secular regimes that controlled much of the Muslim world. The aim was to evict them, return to a purer form of Islam and re-create an Islamic state. In view of the exaggerated reputation it enjoys in the West, it is worth remembering that this movement has largely been a failure. Yet while fundamentalism appeals to only a small minority, it is also the case that large numbers of Muslims have become aggrieved by the policies of the Western powers. The explanation for this can be found in long-term frustration with the consistently pro-Israeli policy of Britain and the United States over Palestine, in addition to the proximate causes in the shape of two Afghan wars, the genocide in Bosnia, the Rushdie affair, and the first Gulf War in 1990, which made many Muslims see themselves as the victims of Western aggression and interventionism.
9
Migon, Piotr. "Slope Development in Granite Terrains." In Granite Landscapes of the World. Oxford University Press, 2006. http://dx.doi.org/10.1093/oso/9780199273683.003.0013.
Full textAbstract:
Rock slopes developed in granite may take different forms, as reflected in their longitudinal profiles. Field observations and a literature survey (e.g. Dumanowski, 1964; Young, 1972) allow us to distinguish at least five major categories of slopes: straight, convex-upward, concave, stepped, and vertical rock walls. In addition, overhang slopes may occur, but their height is seldom more than 10 m high and their occurrence is very localized. These basic categories may combine to form compound slopes, for example convex-upward in the upper part and vertical towards the footslope. Somewhat different is Young’s (1972) attempt to identify most common morphologies of granite slopes. He lists six major categories: (1) bare rock domes, smoothly rounded or faceted; (2) steep and irregular bare rock slopes of castellated residual hills, tending towards rectangular forms; (3) concave slopes crowned by a free face; (4) downslope succession of free face, boulder-covered section and pediment; (5) roughly straight or concave slopes, but having irregular, stepped microrelief; (6) smooth convex-concave profile with a continuous regolith cover. The latter, lacking any outcrops of sound bedrock, are not considered as rock slopes for the purposes of this section. Young (1972) appears to seek explanation of this variety in climatic differences between regions, claiming that ‘Variations of slope form associated with climatic differences are as great as or greater, on both granite and limestone, than the similarity of form arising from lithology’ (Young, 1972: 219). This is a debatable statement and apparently contradicted by numerous field examples of co-existence of different forms in relatively small areas. Slope forms do not appear specifically subordinate to larger landforms but occur in different local and regional geomorphic settings. For example, the slopes of the Tenaya Creek valley in the Yosemite National Park include, in different sections of the valley, straight, vertical, convex-upward, and concave variants (Plate 5.1). Apparently, multiple glaciation was unable to give the valley a uniform cross-sectional shape.
10
Lowenstam, Heinz A., and Stephen Weiner. "Mollusca." In On Biomineralization. Oxford University Press, 1989. http://dx.doi.org/10.1093/oso/9780195049770.003.0008.
Full textAbstract:
Mollusks have a well-deserved reputation for being expert mineralizers based only on their much-admired shell-making abilities. Table 6.1 shows that the reputation is deserved 10-fold as shell formation is just one of many different processes that these animals perform in which biogenic minerals are utilized. The table lists no less than 21 different minerals and about 17 different functions! The list contains both amorphous minerals (amorphous fluorite, calcium carbonate, calcium phosphate, calcium pyrophosphate, and silica) and many crystalline ones, including rather uncommon ones such as weddelite, calcium fluorite, barite, magnetite, lepidocrocite, and goethite. Weddelite, for example, is a calcium oxalate mineral frequently formed pathologically in vertebrates. Certain gastropods use the rather soft weddelite nonpathologically to cap pestlelike objects (gizzard plates) in their stomachs (Lowenstam 1968), which they use for crushing shelled prey. One mollusk, the chambered Nautilus, forms no less than five different minerals. An individual tooth of a chiton contains three different mature minerals that are products of two other transient minerals. In addition to the more familiar functions of mineralized tissues, mollusks use biogenic minerals as buoyancy devices, trap doors, egg shells, and love darts. The varieties of crystal shapes, sizes, organizational arrays, and tissue sites present a picture of overwhelming diversity all within one phylum. It is illustrative to compare the mollusks with the echinoderms. The echinoderms also use minerals for a wide variety of functions, but in contrast to the mollusks they use essentially the same “building material” for many different purposes. Thus, understanding how one echinoderm mineralized tissue forms provides insight into how most of the others form. This is not so with mollusks. It seems futile to expect that they too have adapted one basic process to form all their mineralized tissues. It seems just as futile to look for a different explanation for each type of mineralized product. The mollusks force us to seek a level of understanding of mineralization that identifies common approaches, strategies, and principles and, at the same time, appears to dispel any “dreams” about discovering the mechanism of mineralization. The mollusk phylum contains seven different taxonomic classes.
Conference papers on the topic "Shapley Additive Explanations"
1
Stojić, Andreja, Marijana Matek Sarić, and Snježana Herceg Romanić. "Shapley Additive Explanations of Indicator PCB-138 Distribution in Breast Milk." In Sinteza 2020. Beograd, Serbia: Singidunum University, 2020. http://dx.doi.org/10.15308/sinteza-2020-35-40.
Full text
2
Park, Sungwoo, Jihoon Moon, and Eenjun Hwang. "Explainable Anomaly Detection for District Heating Based on Shapley Additive Explanations." In 2020 International Conference on Data Mining Workshops (ICDMW). IEEE, 2020. http://dx.doi.org/10.1109/icdmw51313.2020.00111.
Full text
3
Han, Fuchang, Shenghui Liao, Siming Yuan, Renzhong Wu, Yuqian Zhao, and Yu Xie. "Explainable Prediction Of Renal Cell Carcinoma From Contrast-Enhanced CT Images Using Deep Convolutional Transfer Learning And The Shapley Additive Explanations Approach." In 2021 IEEE International Conference on Image Processing (ICIP). IEEE, 2021. http://dx.doi.org/10.1109/icip42928.2021.9506144.
Full text
4
M. Thimoteo, Lucas, Marley M. Vellasco, Jorge M. do Amaral, Karla Figueiredo, Cátia Lie Yokoyama, and Erito Marques. "Interpretable Machine Learning for COVID-19 Diagnosis Through Clinical Variables." In Congresso Brasileiro de Automática - 2020. sbabra, 2020. http://dx.doi.org/10.48011/asba.v2i1.1590.
Full textAbstract:
This work proposes an interpretable machine learning approach to diagnosesuspected COVID-19 cases based on clinical variables. Results obtained for the proposed models have F-2 measure superior to 0.80 and accuracy superior to 0.85. Interpretation of the linear model feature importance brought insights about the most relevant features. Shapley Additive Explanations were used in the non-linear models. They were able to show the difference between positive and negative patients as well as offer a global interpretability sense of the models.
5
Pereira, Filipe Dwan, Elaine Harada Teixeira de Oliveira, David Braga Fernandes de Oliveira, Leandro Silva Galvão de Carvalho, and Alexandra Ioana Cristea. "Interpretable AI to Understand Early Effective and Ineffective Programming Behaviours from CS1 Learners." In Anais Estendidos do Simpósio Brasileiro de Educação em Computação. Sociedade Brasileira de Computação, 2021. http://dx.doi.org/10.5753/educomp_estendido.2021.14853.
Full textAbstract:
Building predictive models to estimate the learner performance in the beginning of CS1 courses is essential in education to allow early interventions. However, the educational literature notes the lack of studies on early learner behaviours that can be effective or ineffective, that is, programming behaviours that potentially lead to success or failure, respectively. Hence, beyond the prediction, it is crucial to explain what leads the predictive model to make the decisions (e.g., why a given student s is classified as `passed'), which would allow a better understanding of which early programming behaviours are to be encouraged and triggered. In this work in progress, we use a state-of-the-art unified approach to interpret black-box model predictions, which uses SHapley Additive exPlanations (SHAP) method. SHAP method can be used to explain linearly a complex model (e.g. DL or XGboost) in instance level. In our context of CS1 performance prediction, this method gets the predictive model and the features values for a given student as input and the possibility of explanation of which feature values are increasing or decreasing the learner chances of passing as output. That is, using SHAP we can identify early effective and ineffective behaviours in student-level granularity. More than that, using this local explanation as building blocks, we can also extract global data insight and give a summarisation of the model. A video explaining this work can be found at the following link: https://youtu.be/pd6Ma6uInHo
6
Nohara, Yasunobu, Koutarou Matsumoto, Hidehisa Soejima, and Naoki Nakashima. "Explanation of Machine Learning Models Using Improved Shapley Additive Explanation." In BCB '19: 10th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics. New York, NY, USA: ACM, 2019. http://dx.doi.org/10.1145/3307339.3343255.
Full text
7
Choudary, Movva Naga Sumanth, Vinay Babu Bommineni, Grandhi Tarun, Guvvala Prasanth Reddy, and G. Gopakumar. "Predicting Covid-19 Positive Cases and Analysis on the Relevance of Features using SHAP (SHapley Additive exPlanation)." In 2021 Second International Conference on Electronics and Sustainable Communication Systems (ICESC). IEEE, 2021. http://dx.doi.org/10.1109/icesc51422.2021.9532829.
Full text
8
Gallego, Juan A., and Just Herder. "Synthesis Methods in Compliant Mechanisms: An Overview." In ASME 2009 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. ASMEDC, 2009. http://dx.doi.org/10.1115/detc2009-86845.
Full textAbstract:
Compliant mechanisms are rapidly gaining importance, yet their design remains challenging. A great variety of methods are being developed as it is reported in a growing stream of publications. However, so far no review of this body of literature is available. This paper provides a comprehensive and conceptual overview of the main notions behind the most relevant design methods for compliant mechanisms. Rigid-Body-Replacement methods including the Pseudo-Rigid-Body model and the FACT approach are covered, as well as Building Block approaches. In addition an introduction and explanation on Topology Optimization and Shape Optimization is provided, including their most common parameterizations and formulations. This work aims to serve as an introduction into compliant mechanism design methods and as a reference work for more experienced scholars and professionals. It is intended to be a starting point for the exploration of the literature, as well as a guide to specific papers about a particular design problem one may have. For this reason, the paper presents the methods in a wide perspective, emphasizing the conceptual ideas behind every method and refers to literature for details and advanced features.
9
Zarea, Mures, Remi Batisse, Brian Leis, Philippe Cardin, and Geoff Vignal. "Full Scale Experimental Database of Dent and Gouge Defects to Improve Burst and Fatigue Strength Models of Pipelines." In 2012 9th International Pipeline Conference. American Society of Mechanical Engineers, 2012. http://dx.doi.org/10.1115/ipc2012-90620.
Full textAbstract:
External interference on gas and oil transmission pipelines is consistently reported as leading cause of leaks in Europe and USA as identified in the EGIG and PHMSA incident databases. External interference due to ground working machinery strikes, rock strikes during backfilling, etc. on buried pipelines result mainly in dent and gouge defects. The long-term integrity of a pipeline segment damaged by a dent and gouge defect is a complex function of a variety of parameters, including pipe material properties, pipe geometry, defect geometry linked to indenter shape, aggression conditions. The complexity and extreme variability of these dent and gouge defect shapes and pipe materials lead simple assessment models to scattered predictions, hinting towards an insufficient description of real structural and material behavior. To improve knowledge beyond the numerous studies led in the past, and to provide a sound foundation for developing and validating mechanistic models for predicting burst and fatigue strength of such defects, a large experimental program was funded by PRCI and US DoT and further coordinated with a complementary EPRG program. The experimental program part dealing with combined “Dent and Gouge” defects is covered for modern pipes (24″ OD, X52 and X70) by PRCI project MD-4-1: realistically created (with a Pipe Aggression Rig) defects submitted to full scale burst and fatigue tests, in addition to extensive characterization. This work interfaces with modeling to predict the immediate burst strength and fatigue resistance of such damage in two PRCI projects denoted MD-4-3 and MD-4-4 respectively. This paper gives an overview of some of these activities: PRCI project MD-4-1 results: material characterization, full scale burst and fatigue tests on Dents with Gouges, as well as detailed explanations concerning the initial approach to model burst and fatigue life of these defects, as developed byr PRCI project MD-4-4. The final outcome of the expected knowledge improvements about the mechanical strength of dent and gouge combinations will be applicable by pipeline operators, in order to enhance integrity management systems designed to manage the threat associated with mechanical damage.
10
Marie, S., C. Guerre, and E. Herms. "Analysis of the Truth Loading Conditions of a Austenitic CT Specimen During a SCC Experiment." In ASME 2011 Pressure Vessels and Piping Conference. ASMEDC, 2011. http://dx.doi.org/10.1115/pvp2011-57170.
Full textAbstract:
With the aim to investigate the influence of strain hardening on the stainless steels susceptibility to stress corrosion cracking, tests were conducted in PWR environment on CT specimens, taken from a 316L stainless steel sheet cold rolled to 40% in thickness reduction. The initial cracks obtained by the fatigue pre-cracking have an atypical ‘V’ shape with smaller propagation in the center of the CT thickness compared to nominal propagation observed at both sides. The initial explanation was to consider a stress intensity factor derived from classical reference solution on the basis of a straight crack front, and considering the local value of the crack depth in the equation. This assumption raised several problems analsyes in this paper. This particular shape of the initial defect may be related to several factors, and partly to the 40% cold rolling. It is likely that the hardening is not uniform, with a higher rate at the specimen sides than in the central area. In addition, significant residual stresses due to the gradient of mechanical properties are observed. Due to the high rate of work hardening by rolling of the sheet metal, a gradient of the mechanical properties through the thickness was determined, and the residual stresses profile induced by this process was measured. The variations obtained are consistent with each other: the material is more hardened in the vicinity of specimen surface and residual stresses are compressive in nature in the central part of the specimen and of tensile type on the flanks. All these data were firstly considered in order to assess their role regarding the particular form of the initial crack front obtained after fatigue: the 3D finite element calculations taking into account the true shape of the crack front demonstrate the relationship between the characteristics of the experimental crack front obtained after fatigue pre-cracking and the residual stresses. Moreover, from the residual stresses measured on the plate where samples have been machined/prepared, the residual stresses field in the specimen after its machining is calculated and then taken into account in the mechanical analysis. The characteristics of this field in addition to the mechanical loading applied during SCC testing can explain the crack propagation behavior observed experimentally.