[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125447-en":3,"doc-seo-125447-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},125447,687197207919,"Theodora","https://ap-avatar.wpscdn.com/avatar/a000253d6f5f7c60be?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779446848396160552",8,"Research & Report","Informative Machine Learning Model Explanation Techniques","Explainable AI (XAI) focuses on providing human-interpretable insights into complex, often black-box machine learning (ML) models. This thesis studies Shapley value attribution (SVA), which assigns contributions of features to model behavior via local (prediction-level) and global (metric-level) explanations. Prior work shows SVA limitations that can produce biased or incorrect explanations and can be manipulated adversarially. Global SVAs also become unreliable on imbalanced datasets used in fraud detection and disease prediction.","Informative Machine Learning Model Explanation Techniques  \nNingsheng Zhao  \nA Thesis  \nin  \nThe Department  \nof  \nConcordia Institute for Information Systems Engineering (CIISE)  \nPresented in Partial Fulfillment of the Requirements for the Degree of  \nDoctor of Philosophy (Information and Systems Engineering) at  \nConcordia University  \nMontrÂeal, QuÂebec, Canada  \nJanuary 2025  \n© Ningsheng Zhao, 2025  \nCONCORDIA UNIVERSITY  \nSchool of Graduate Studies  \nThis is to certify that the thesis prepared  \nBy: Mr. Ningsheng Zhao  \nEntitled: Informative Machine Learning Model Explanation Techniques  \nand submitted in partial fulfillment of the requirements for the degree of  \nDoctor of Philosophy (Information and Systems Engineering)  \ncomplies with the regulations of this University and meets the accepted standards with respect to originality and quality.  \nSigned by the Final Examining Committee:  \n  Chair  \nDr. Lan Lin  \n  External Examiner Dr. Kim Khoa Nguyen  \n  Examiner  \nDr. Mazdak Nik-Bakht  \n  Examiner  \nDr. Nizar Bouguila  \n  Examiner  \nDr. Chun Wang  \nDr. Jia Yuan Yu  Supervisor  \nDr. Yong Zeng  Co-supervisor  \nApproved by Dr. Farnoosh Naderkhani, Graduate Program Director  2025.01.14  \nDr. Mourad Debbabi, Dean of Faculty  \nAbstract  \nInformative Machine Learning Model Explanation Techniques  \nNingsheng Zhao, Ph.D.  \nConcordia University, 2025  \nExplainable AI (XAI) is an emerging field focused on providing human-interpretable insights into complex and often black-box machine learning (ML) models. Shapley value attribution (SVA) is an increasingly popular XAI method that quantifies the contribution of each feature to a model’s behavior, which can be either an individual prediction (local SVAs) or a performance metric (global SVAs) . However, recent research has highlighted several limitations in existing SVA methods, leading to biased or incorrect explanations that fail to capture the true relationships between features and model behaviors. What’s worse, these explanations are vulnerable to adversarial manipulation.  \nAdditionally, global SVAs, while widely used in applied studies to gain insights into underlying information systems, face challenges when applied to ML models trained on imbalanced datasets, such as those used in fraud detection or disease prediction. In these scenarios, global SVAs can yield misleading or unstable explanations.  \nThis thesis aims to address these challenges and improve the reliability and informativeness of SVA explanations. It makes three key contributions: 1) Proposing a novel error analysis framework that comprehensively examines the underlying sources of bias in existing SVA methods; 2) Introducing a series of refinement methods that significantly enhance the informativeness of SVA explanations, as well as their robustness against adversarial attacks; 3) Developing a standardization method for evaluating global model behaviors on imbalanced datasets, advancing the development of an explainable model monitoring system. Our experiments demonstrate that these methods substantially improve the ability of SVAs to uncover informative patterns in model behaviors, making them valuable tools for knowledge discovery, model debugging, and performance monitoring.  \nAcknowledgments  \nThis journey has been long and challenging, and I have many people to thank for their support. First and foremost, I want to thank my supervisor, Dr. Jia Yuan Yu. His invaluable guidance, support, and advice have been with me every step of the way, and I wouldn’t have reached this point without his encouragement and mentorship. His thoughtful feedback not only improved this thesis but also helped me grow as a researcher, and his continued belief in my work gave me the confidence to keep pushing forward. I feel truly fortunate to have had him as my mentor and hope to learn more from him in the years to come.  \nI also want to express my sincere gratitude to my co-supervisor, Dr. Yong Zeng, for always being there with su","cbCair47o9tQxOtv","https://ap.wps.com/l/cbCair47o9tQxOtv","pdf",16966226,1,153,"English","en",105,"# Abstract\n## Explainable AI and SVA\n## Limitations and challenges\n## Thesis contributions\n## Experimental outcomes","[{\"question\":\"What is the main research focus of this thesis?\",\"answer\":\"The thesis focuses on improving the reliability and informativeness of Shapley value attribution (SVA) explanations for machine learning models within Explainable AI (XAI).\"},{\"question\":\"What kinds of issues do existing SVA methods face?\",\"answer\":\"Existing SVA methods may generate biased or incorrect explanations that fail to reflect true feature–behavior relationships, and they are vulnerable to adversarial manipulation. Global SVAs can also become misleading or unstable on imbalanced datasets.\"},{\"question\":\"What contributions does the thesis make to address these challenges?\",\"answer\":\"It proposes an error analysis framework to identify bias sources, introduces refinement methods to enhance informativeness and robustness against adversarial attacks, and develops a standardization approach for evaluating global behaviors on imbalanced datasets.\"}]","Informative Machine Learning Model Explanation Techniques | PDF",1785899054,386,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"informative-machine-learning-model-explanation-techniques","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/informative-machine-learning-model-explanation-techniques/125447/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What is the main research focus of this thesis?","Question",{"text":75,"@type":76},"The thesis focuses on improving the reliability and informativeness of Shapley value attribution (SVA) explanations for machine learning models within Explainable AI (XAI).","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What kinds of issues do existing SVA methods face?",{"text":80,"@type":76},"Existing SVA methods may generate biased or incorrect explanations that fail to reflect true feature–behavior relationships, and they are vulnerable to adversarial manipulation. Global SVAs can also become misleading or unstable on imbalanced datasets.",{"name":82,"@type":73,"acceptedAnswer":83},"What contributions does the thesis make to address these challenges?",{"text":84,"@type":76},"It proposes an error analysis framework to identify bias sources, introduces refinement methods to enhance informativeness and robustness against adversarial attacks, and develops a standardization approach for evaluating global behaviors on imbalanced datasets.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]