{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/temporal-stability-in-predictive-process","title":"Temporal Stability in Predictive Process Monitoring","arxiv_id":"1712.04165","date":"2017-12-12","proceeding":null,"authors":["Irene Teinemaa","Marlon Dumas","Anna Leontjeva","Fabrizio Maria Maggi"],"abstract":"Predictive process monitoring is concerned with the analysis of events\nproduced during the execution of a business process in order to predict as\nearly as possible the final outcome of an ongoing case. Traditionally,\npredictive process monitoring methods are optimized with respect to accuracy.\nHowever, in environments where users make decisions and take actions in\nresponse to the predictions they receive, it is equally important to optimize\nthe stability of the successive predictions made for each case. To this end,\nthis paper defines a notion of temporal stability for binary classification\ntasks in predictive process monitoring and evaluates existing methods with\nrespect to both temporal stability and accuracy. We find that methods based on\nXGBoost and LSTM neural networks exhibit the highest temporal stability. We\nthen show that temporal stability can be enhanced by hyperparameter-optimizing\nrandom forests and XGBoost classifiers with respect to inter-run stability.\nFinally, we show that time series smoothing techniques can further enhance\ntemporal stability at the expense of slightly lower accuracy.","url_abs":"http://arxiv.org/abs/1712.04165v3","url_pdf":"http://arxiv.org/pdf/1712.04165v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"temporal-stability-in-predictive-process","repo_url":"https://github.com/irhete/stability-predictive-monitoring","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"binary-classification","task_name":"Binary Classification"},{"task_slug":"predictive-process-monitoring","task_name":"Predictive Process Monitoring"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series","task_name":"Time Series Analysis"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}