Temporal stability in predictive process monitoring

• Published in: Data mining and knowledge discovery Vol. 32; no. 5; pp. 1306 - 1338
• Main Authors: Teinemaa, Irene, Dumas, Marlon, Leontjeva, Anna, Maggi, Fabrizio Maria
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.09.2018; Springer Nature B.V
• Subjects: Accuracy; Artificial Intelligence; Chemistry and Earth Sciences; Computer Science; Data Mining and Knowledge Discovery; Decision trees; Information Storage and Retrieval; Journal Track of ECML PKDD 2018; Monitoring; Neural networks; Optimization; Physics; Predictions; Stability analysis; Statistics for Engineering; Predictive process monitoring; Early sequence classification; Stability
• ISSN: 1384-5810; 1573-756X
• DOI: 10.1007/s10618-018-0575-9

Predictive process monitoring is concerned with the analysis of events produced during the execution of a business process in order to predict as early as possible the final outcome of an ongoing case. Traditionally, predictive process monitoring methods are optimized with respect to accuracy. However, in environments where users make decisions and take actions in response to the predictions they receive, it is equally important to optimize the stability of the successive predictions made for each case. To this end, this paper defines a notion of temporal stability for binary classification tasks in predictive process monitoring and evaluates existing methods with respect to both temporal stability and accuracy. We find that methods based on XGBoost and LSTM neural networks exhibit the highest temporal stability. We then show that temporal stability can be enhanced by hyperparameter-optimizing random forests and XGBoost classifiers with respect to inter-run stability. Finally, we show that time series smoothing techniques can further enhance temporal stability at the expense of slightly lower accuracy.