[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-128426-en":3,"doc-seo-128426-105":31,"detail-sidebar-cat-0-en-105":92},{"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},128426,962084925290,"Ophelia","https://ap-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","Applying machine learning to process mining data - Master’s thesis","A global electronics manufacturing enterprise applies process mining data to predictive analytics for sales opportunity outcomes. Sales opportunity data from the enterprise process mining platform serves as the use case, with the goal of evaluating machine learning models’ accuracy in predicting whether opportunities in the sales funnel are won or lost. Using a Design Science Research approach, the study frames the problem, performs narrative literature review and exploratory analysis, iterates four development cycles with XGBoost selection, and builds time-aware snapshots to reduce data leakage.","Applying machine learning to process mining data  \nOskari Rannanniemi  \nMaster’s thesis January 2026  \nArtificial Intelligence and Data Analytics, Master of Engineering  \nRannanniemi, Oskari  \nApplying machine learning to process mining data  \nJyväskylä: Jamk University of Applied Sciences, January 2026, 63(+21) pages Degree Programme in Artificial Intelligence and Data Analytics, Master’s thesis.  \nPermission for open access publication: Yes  \nLanguage of publication: English  \nAbstract  \nThe study was conducted for a global electronics manufacturing enterprise aiming to leverage its process mining platform for predictive analytics. Sales opportunity data within the process mining platform was selected as the use case. The objective of the study was to evaluate the accuracy of machine learning models in predicting whether opportunities in the sales funnel would be won or lost.  \nThe study used Design Science Research approach, advancing through initial, intermediate, and final artifact development stages. The initial stage involved problem framing, a narrative literature review to identify optimal models, and exploratory data analysis on an initial data set. During the intermediate stage, four iterative cycles were completed, one of which had emphasis on model selection, with XGBoost ultimately being chosen as the final model. Data was transformed into time-aware snapshots to mitigate the risk of data leakage, and two distinct models were developed. At the final stage, the code repository for the model was finalized, concepts for the end-user interface were established, and the necessary data pipelines were concepted.  \nThe two final models achieved similar ROC-AUC scores of 77.53% . Both were considered suitable for win prediction, but they lacked reliability when predicting lost opportunities. The first model, with a decision threshold of 0 . 2, had an accuracy of 91 . 11% . This model only slightly outperformed the naive baseline, which involves guessing all opportunities as won. The second model had a strict decision threshold of 0.9 and performed well on precision (97 .93%), which means that when the model identifies a win, it is very likely correct. However, the overall accuracy of the second model decreased to 52.53%, indicating that many actual wins were missed. The main challenges encountered were class imbalance and a limited set of input features, suggesting that including more distinctive features is necessary for improved predictive performance.  \nThe study demonstrated effective use of temporal features from process mining, while noting frequent hidden data leakage issues in prior studies. It recommended enhancing process mining platforms with better tools for reconstructing historical states. Recommendations for future projects proposed development into richer feature sets, alternative encoding strategies, and segment-specific modeling to improve predictions.  \nKeywords/tags (subjects)  \nProcess mining, predictive analytics, sales opportunity prediction, machine learning, XGBoost  \nRannanniemi, Oskari  \nApplying machine learning to process mining data  \nJyväskylä: Jyväskylän ammattikorkeakoulu. Tammikuu 2026, 63(+21) sivua  \nDegree Programme in Artificial Intelligence and Data Analytics, Opinnäytetyö YAMK.  \nJulkaisun kieli: englanti  \nJulkaisulupa avoimessa verkossa: kyllä  \nTiivistelmä  \nTutkimus toteutettiin globaalille elektroniikkateollisuuden yritykselle, jonka tavoitteena oli hyödyntää prosessilouhintaa ennakoivassa analytiikassa. Käyttötapaukseksi valittiin prosessilouhinta-alustan myyntiprosessidataa. Tutkimuksen tavoitteena oli arvioida koneoppimismallien tarkkuuttamyyntimahdollisuuksien voittamisen tai häviämisen ennustamisessa.  \nTutkimuksessa hyödynnettiin Design Science Research-lähestymistapaa, joka eteni alku-, väli-ja loppuvaiheen artefaktien kehityksen kautta. Alkuvaiheessa rajattiin ongelma, laadittiin narratiivinenkirjallisuuskatsaus optimaalisten mallien löytämiseksi ja tehtiin eksploratiiv","cbCairsQAuaZKm7Z","https://ap.wps.com/l/cbCairsQAuaZKm7Z","pdf",3794900,2,1,87,"English","en",105,"# Abstract\n# Research objective and use case\n# Methodology and artifacts development\n# Model development and evaluation\n# Results, challenges, and recommendations","[{\"question\":\"What problem does the thesis address and what data is used?\",\"answer\":\"The thesis evaluates machine learning models for predicting whether sales opportunities in the sales funnel are won or lost. It uses sales opportunity data from the enterprise’s process mining platform.\"},{\"question\":\"How does the study reduce the risk of data leakage?\",\"answer\":\"The study transforms the data into time-aware snapshots so that information from future states is not leaked into model training.\"},{\"question\":\"What models are built and how do their performances differ?\",\"answer\":\"Two final models are developed, both reaching similar ROC-AUC scores of 77.53%. The first model performs with higher overall accuracy at a lower decision threshold, while the second model shows very high precision but lower overall accuracy due to missed wins.\"}]","Applying machine learning to process mining data - Master’s thesis | PDF",1785947552,219,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":29},"applying-machine-learning-to-process-mining-data-masters-thesis","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,48,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":20},"https://docshare.wps.com/document/","Document",{"item":49,"name":12,"@type":44,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/applying-machine-learning-to-process-mining-data-masters-thesis/128426/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-26","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does the thesis address and what data is used?","Question",{"text":76,"@type":77},"The thesis evaluates machine learning models for predicting whether sales opportunities in the sales funnel are won or lost. It uses sales opportunity data from the enterprise’s process mining platform.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the study reduce the risk of data leakage?",{"text":81,"@type":77},"The study transforms the data into time-aware snapshots so that information from future states is not leaked into model training.",{"name":83,"@type":74,"acceptedAnswer":84},"What models are built and how do their performances differ?",{"text":85,"@type":77},"Two final models are developed, both reaching similar ROC-AUC scores of 77.53%. The first model performs with higher overall accuracy at a lower decision threshold, while the second model shows very high precision but lower overall accuracy due to missed wins.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]