[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-139236-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-139236-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","human-centric-explanation-facilities-explainable-ai-for-the-pragmatic-understanding-of-non-expert-end-users","Human-centric Explanation Facilities - Explainable AI für das pragmatische Verständnis nicht-technischer Endnutzer","","AI-enhanced systems support or automate decision-making in sensitive domains such as medicine, education, criminal justice, loan approval, and recruiting, often relying on machine learning. The black-box nature of ML creates concerns about unintended consequences, triggering demands for interpretable and explainable approaches for stakeholders beyond system developers. This thesis focuses on non-expert users of Explainable AI (XAI), examining human factors and cognitive biases affecting how users interpret XAI explanations, and developing interaction concepts that promote pragmatic understanding.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/human-centric-explanation-facilities-explainable-ai-for-the-pragmatic-understanding-of-non-expert-end-users/139236/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/human-centric-explanation-facilities-explainable-ai-for-the-pragmatic-understanding-of-non-expert-end-users/139236.png","ImageObject",300,407,{"name":42,"@type":43},"Quinn","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-29","2026-08-23",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",16,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What is the main focus of this thesis?","Question",{"text":63,"@type":64},"The thesis focuses on a human-centric view of Explainable AI for non-expert end users and how they interpret XAI explanations to support their tasks.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"Which contributions does the thesis make?",{"text":68,"@type":64},"It empirically studies unknown human factors and cognitive biases, structures inconsistencies in evaluating explanations in empirical XAI studies, and presents case studies of explanation facility artifacts meeting naturalness, responsiveness, and sensitivity requirements.",{"name":70,"@type":61,"acceptedAnswer":71},"Why are explainable systems important in this context?",{"text":72,"@type":64},"Because ML systems can produce unintended consequences due to the black-box problem, social, ethical, and legal calls demand interpretability for a broader set of stakeholders.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},139236,1787514635,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":140,"language":141,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":12,"update_tm":80,"read_time":145},962075114765,"https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd","Human-centric Explanation Facilities: Explainable AI for the Pragmatic Understanding of Non-expert End Users  \nDissertation  \nan der Fakultät für Mathematik, Informatik und Statistik der Ludwig-Maximilians-Universität München  \neingereicht von  \nMICHAEL CHROMIK  \nMünchen, den 09.04.2021  \nErstgutachter: Prof. Dr. Andreas Butz Zweitgutachter: Prof. Dr.-Ing. Klaus Diepold Drittgutachter: Prof. Dr. Dieter Kranzlmüller  \nTag der mündlichen Prüfung: 28.06.2021  \nABSTRACT  \nAI-infused systems support or automate decision-making in many sensitive contexts of society, such as medicine, education, criminal justice, loan approval, or recruiting. To accomplish this, these systems often leverage machine learning (ML) methods. The risk of unintended consequences caused by the black-box problem of these systems drove social, ethical, and legal calls for more interpretable and explainable systems for stakeholders beyond the systems' developers. This field is often referred to as Explainable Artificial Intelligence (XAI) . XAI researchers face the accusation of putting their emphasis on generating explanatory models for like-minded ML experts instead of tailoring them to the actual users, who often lack the technical background about ML.  \nThis human-centric view on non-expert users of XAI is the focus of this thesis. By non-experts, we refer to users of XAI systems who have not been trained or educated in ML concepts, but who use the system's predictions to support or perform their tasks. The work presented in this thesis (i) examines human factors that may be encountered by end users when interpreting explanations from ML-based XAI systems, and (ii) explores the interaction space of XAI explanation facilities to foster a pragmatic understanding of the underlying ML model.  \nTo this end, this thesis makes three contributions: (i) it empirically explores unknown human factors and cognitive biases that influence the end user understanding gained through XAI explanations,(ii) it conceptually structures inconsistencies within the involved research communities on how explanations are evaluated with human subjects in empirical XAI studies and how user interaction with explanation interfaces in XAI can be described, and (iii) it presents case studies of constructive explanation facility artifacts that fulfill the requirements of naturalness (use natural language explanations), responsiveness (allow follow-up interactions), and sensitivity (elicit end user beliefs to calibrate their expectations) . In summary, this thesis raises XAI designers' awareness of the human aspects of interpretability in XAI.  \nZUSAMMENFASSUNG  \nMit Künstlicher Intelligenz (KI) angereicherte Systeme unterstützen oder automatisieren Entscheidungen in vielen sensiblen Bereichen der Gesellschaft, beispielsweise in der Medizin, im Bildungswesen, im Strafrecht, der Kreditvergabe oder der Personalrekrutierung. Dafür nutzen diese Systeme oftmals Methoden des Maschinellem Lernens (ML) . Das Risiko unbeabsichtigter Konsequenzen, das vom Blackbox-Problem des Maschinellen Lernens ausgehen kann, führte zu sozialen, ethischen undrechtlichen Forderungen nach besser interpretierbaren und erklärbaren Systemen für Nutzergruppen, die über die Entwickler dieser Systeme hinausgehen. Dieses Feld wird oft als Erklärbare Künstliche Intelligenz (engl. Explainable Artificial Intelligence, XAI) bezeichnet. XAI-Forscher sehen sich dem Vorwurf ausgesetzt, dass sie viel Wert darauf legen, Erklärungsmodelle für gleichgesinnte MLExperten zu generieren, anstatt sie auf die eigentlichen Endnutzer, die oftmals keine technischen Experten sind, auszurichten.  \nDiese menschenzentrierte Sichtweise auf nicht-technische Endnutzer von XAI ist der Fokus dieser Arbeit. Mit nicht-technischen Endnutzern bezeichnen wir Nutzer von XAI-Systemen, die wenig mit den Konzepten des maschinellen Lernens vertraut sind, aber dessen Vorhersagen zur Unterstützung oder Erfüllung ihrer Aufgaben nutzen. Die in dieser Thesis vorgestellten ","cbCaivTv4XYdQ4pZ","https://ap.wps.com/l/cbCaivTv4XYdQ4pZ","pdf",3880016,125,"English","# Abstract\n# Zusammenfassung\n# Acknowledgement","[{\"question\":\"What is the main focus of this thesis?\",\"answer\":\"The thesis focuses on a human-centric view of Explainable AI for non-expert end users and how they interpret XAI explanations to support their tasks.\"},{\"question\":\"Which contributions does the thesis make?\",\"answer\":\"It empirically studies unknown human factors and cognitive biases, structures inconsistencies in evaluating explanations in empirical XAI studies, and presents case studies of explanation facility artifacts meeting naturalness, responsiveness, and sensitivity requirements.\"},{\"question\":\"Why are explainable systems important in this context?\",\"answer\":\"Because ML systems can produce unintended consequences due to the black-box problem, social, ethical, and legal calls demand interpretability for a broader set of stakeholders.\"}]","Human-centric Explanation Facilities - Explainable AI für das pragmatische Verständnis nicht-technischer Endnutzer | PDF",315]