[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-125604-en":3,"doc-seo-125604-105":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":4,"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},125604,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",6,"Technology","Fighting the disagreement in Explainable Machine Learning with consensus - Preprint","Machine learning models are commonly valued for prediction accuracy, yet in critical scientific domains the internal decision process matters as much as performance. Interpretability algorithms help reveal model workings, but multiple methods often produce conflicting explanations, resulting in contradictory, unclear insights. Consensus functions can reconcile these outputs, though results depend on the chosen consensus rule and influencing factors. This work evaluates six consensus functions on explanations from five models trained on four synthetic datasets, using model-agnostic local and global interpretability. The proposed function yields fairer, more consistent and accurate explanations across datasets.","Graphical Abstract  \nFighting the disagreement in Explainable Machine Learning with consensus  \nAntonio Jess Banegas-Luna , Carlos Mart´ınez-Corts, Horacio Prez-Snchez  \nHighlights  \nFighting the disagreement in Explainable Machine Learning with consensus  \nAntonio Jess Banegas-Luna, Carlos Mart´ınez-Corts, Horacio Prez-Snchez  \n• The interpretability of machine learning models is essential in critical contexts, such as medicine.  \n• There are several interpretability approaches that often disagree.  \n• The disagreement problem can be combated with consensus.  \n• A novel consensus function taking into account model accuracy, class probability and feature attribution has been developed.  \n• The proposed function leads to more consistent explanations than the others in four synthetic datasets.  \nFighting the disagreement in Explainable Machine Learning with consensus  \nAntonio Jess Banegas-Luna a, Carlos Mart´ınez-Cortsa, Horacio  \nPrez-Sncheza  \naStructural Bioinformatics and High-Performance Computing (BIO-HPC), Universidad Catlica de Murcia (UCAM), Campus de los  \nJernimos, Guadalupe, 30107, Murcia, Spain  \nAbstract  \nMachine learning (ML) models are often valued by the accuracy of their predictions. However, in some areas of science, the inner workings of models areas relevant as their accuracy. To understand how ML models work internally, the use of interpretability algorithms is the preferred option. Unfortunately, despite the diversity of algorithms available, they often disagree in explaining a model, leading to contradictory explanations. To cope with this issue, consensus functions can be applied once the models have been explained. Nevertheless, the problem is not completely solved because the final result will depend on the selected consensus function and other factors. In this paper, six consensus functions have been evaluated for the explanation of five ML models. The models were previously trained on four synthetic datasets whose internal rules were known in advance. The models were then explained with model-agnostic local and global interpretability algorithms. Finally, consensus was calculated with six different functions, including one developed by the authors. The results demonstrated that the proposed function is fairer than the others and provides more consistent and accurate explanations.  \nKeywords: explainable machine learning, interpretability, consensus, ensemble models  \n1. Introduction  \nMachine learning (ML) is a branch of artificial intelligence that aims to develop techniques to make computers learn. Thanks to ML, computers can perform certain tasks without being programmed for it. Its strong statistical  \nPreprint submitted to Knowledge-Based Systems July 3, 2023  \nbasis makes ML a reliable approach to address complex tasks (e.g. pattern recognition, computer vision) . Because of its accuracy in identifying patterns in data, ML models have been widely adopted in many fields of science, such as biology [1, 2], drug discovery [3, 4, 5], meteorology [6, 7, 8] and medicine [9, 10, 11] .  \nML models are frequently evaluated on the basis of their accuracy. Metrics such as AUC (Area Under the Curve), precision and recall are often used to measure the accuracy of classification models, while R 2, MSE (Mean Squared Error) and MAE (Mean Absolute Error) are common regression metrics [12] . However, some models still tend to be seen as black-boxes [13] . This opacity is frequently a problem in critical areas such as medicine or finance, where users need a deep understanding of the inner workings of models [14] . Such users are not only interested in accuracy but they also need to know how the model made the decision to gain knowledge. This need to interpret models has given rise to eXplainable Machine Learning (XML) [15] . The interpretation of models is often referred to in literature as explainability or interpretability, but some authors make a distinction between both terms [16] . This manuscript does not","cbCaijwym6ihTPG4","https://ap.wps.com/l/cbCaijwym6ihTPG4","pdf",756463,1,26,"English","en",105,"# Highlights\n# Abstract\n# Keywords\n# 1. Introduction\n## Interpretability vs. explainability\n## Taxonomies of interpretability algorithms\n## Limitations and the disagreement problem\n## Consensus functions as a solution","[{\"question\":\"Why is interpretability important even when models are accurate?\",\"answer\":\"In critical areas, users need insight into how decisions are made, not only prediction accuracy. This motivates eXplainable Machine Learning to reveal internal workings.\"},{\"question\":\"What problem occurs when using different interpretability algorithms?\",\"answer\":\"Different algorithms can disagree about which features drive predictions, producing contradictory and fuzzy explanations that undermine interpretability.\"},{\"question\":\"How does the paper address disagreement in explanations?\",\"answer\":\"It applies consensus functions to combine explanations after models have been explained. It evaluates six consensus functions, including one developed by the authors.\"}]","Fighting the disagreement in Explainable Machine Learning with consensus - Preprint | PDF",1785900175,66,{"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},"fighting-the-disagreement-in-explainable-machine-learning-with-consensus-preprint","",{"@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/technology/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/fighting-the-disagreement-in-explainable-machine-learning-with-consensus-preprint/125604/",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":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is interpretability important even when models are accurate?","Question",{"text":75,"@type":76},"In critical areas, users need insight into how decisions are made, not only prediction accuracy. This motivates eXplainable Machine Learning to reveal internal workings.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What problem occurs when using different interpretability algorithms?",{"text":80,"@type":76},"Different algorithms can disagree about which features drive predictions, producing contradictory and fuzzy explanations that undermine interpretability.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the paper address disagreement in explanations?",{"text":84,"@type":76},"It applies consensus functions to combine explanations after models have been explained. It evaluates six consensus functions, including one developed by the authors.","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},[93,97,101,105,110,113,118,123,128,131,135],{"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":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":111,"slug":112},50,"technology",{"id":114,"doc_module":4,"doc_module_name":46,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":119,"doc_module":4,"doc_module_name":46,"category_name":120,"show_sort_weight":121,"slug":122},8,"Research & Report",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"]