What Time Is It? Finding Which Temporal Features is More Useful for Next Activity Prediction
Process Mining merges data science and process science that allows for the analysis of recorded process data by capturing activities within event-logs. It finds more and more applications for the optimization of the production and administrative processes of private companies and public administrati...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Open Journal of the Computer Society |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10810467/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832590323518275584 |
|---|---|
| author | Lerina Aversano Martina Iammarino Antonella Madau Giuseppe Pirlo Gianfranco Semeraro |
| author_facet | Lerina Aversano Martina Iammarino Antonella Madau Giuseppe Pirlo Gianfranco Semeraro |
| author_sort | Lerina Aversano |
| collection | DOAJ |
| description | Process Mining merges data science and process science that allows for the analysis of recorded process data by capturing activities within event-logs. It finds more and more applications for the optimization of the production and administrative processes of private companies and public administrations. This field consists of several areas: process discovery, compliance monitoring, process improvement, and predictive process monitoring. Considering predictive process monitoring, the subarea of next activity prediction helps to obtain a prediction about the next activity performed using control flow data, event data with no attributes other than the timestamp, activity label, and case identifier. A popular approach in this subarea is to use sub-sequences of events, called prefixes and extracted with a sliding window, to predict the next activity. In the literature, several features are added to increase performance. Specifically, this article addresses the problem of predicting the next activity in predictive process monitoring, focusing on the usefulness of temporal features. While past research has explored a variety of features to improve prediction accuracy, the contribution of temporal information remains unclear. In this article it is proposed a comparative analysis of temporal features, such as differences in timestamp, time of day, and day of week, extracted for each event in a prefix. Using both k-fold cross-validation for robust benchmarking and a 75/25 split to simulate real scenarios in which new process events are predicted based on past data, it is shown that timestamp differences within the same prefix consistently outperform other temporal features. Our results are further validated by Shapley's value analysis, highlighting the importance of timestamp differences in improving the accuracy of next activity prediction. |
| format | Article |
| id | doaj-art-4c2a832aaf48432d9288ca7bc122c8c3 |
| institution | Kabale University |
| issn | 2644-1268 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Open Journal of the Computer Society |
| spelling | doaj-art-4c2a832aaf48432d9288ca7bc122c8c32025-01-24T00:02:13ZengIEEEIEEE Open Journal of the Computer Society2644-12682025-01-01626127110.1109/OJCS.2024.351981510810467What Time Is It? Finding Which Temporal Features is More Useful for Next Activity PredictionLerina Aversano0Martina Iammarino1https://orcid.org/0000-0001-8025-733XAntonella Madau2https://orcid.org/0009-0003-2227-9778Giuseppe Pirlo3https://orcid.org/0000-0002-7305-2210Gianfranco Semeraro4https://orcid.org/0000-0003-1666-8323Department of Agricultural Sciences, Food, Natural Resources and Engineering, University of Foggia, Foggia, ItalyDepartment of Information Science and Technology, Pegaso University, Naples, ItalyDepartment of Engineering, University of Sannio, Benevento, ItalyDepartment of Computer Science, University of Bari Aldo Moro, Bari, ItalyEuropean Digital Innovation Hub for Digital Transformation (EDIH4DT), Bari, ItalyProcess Mining merges data science and process science that allows for the analysis of recorded process data by capturing activities within event-logs. It finds more and more applications for the optimization of the production and administrative processes of private companies and public administrations. This field consists of several areas: process discovery, compliance monitoring, process improvement, and predictive process monitoring. Considering predictive process monitoring, the subarea of next activity prediction helps to obtain a prediction about the next activity performed using control flow data, event data with no attributes other than the timestamp, activity label, and case identifier. A popular approach in this subarea is to use sub-sequences of events, called prefixes and extracted with a sliding window, to predict the next activity. In the literature, several features are added to increase performance. Specifically, this article addresses the problem of predicting the next activity in predictive process monitoring, focusing on the usefulness of temporal features. While past research has explored a variety of features to improve prediction accuracy, the contribution of temporal information remains unclear. In this article it is proposed a comparative analysis of temporal features, such as differences in timestamp, time of day, and day of week, extracted for each event in a prefix. Using both k-fold cross-validation for robust benchmarking and a 75/25 split to simulate real scenarios in which new process events are predicted based on past data, it is shown that timestamp differences within the same prefix consistently outperform other temporal features. Our results are further validated by Shapley's value analysis, highlighting the importance of timestamp differences in improving the accuracy of next activity prediction.https://ieeexplore.ieee.org/document/10810467/Process miningnext activity predictiontemporal informationclassificationpredictive process monitoring |
| spellingShingle | Lerina Aversano Martina Iammarino Antonella Madau Giuseppe Pirlo Gianfranco Semeraro What Time Is It? Finding Which Temporal Features is More Useful for Next Activity Prediction IEEE Open Journal of the Computer Society Process mining next activity prediction temporal information classification predictive process monitoring |
| title | What Time Is It? Finding Which Temporal Features is More Useful for Next Activity Prediction |
| title_full | What Time Is It? Finding Which Temporal Features is More Useful for Next Activity Prediction |
| title_fullStr | What Time Is It? Finding Which Temporal Features is More Useful for Next Activity Prediction |
| title_full_unstemmed | What Time Is It? Finding Which Temporal Features is More Useful for Next Activity Prediction |
| title_short | What Time Is It? Finding Which Temporal Features is More Useful for Next Activity Prediction |
| title_sort | what time is it x003f finding which temporal features is more useful for next activity prediction |
| topic | Process mining next activity prediction temporal information classification predictive process monitoring |
| url | https://ieeexplore.ieee.org/document/10810467/ |
| work_keys_str_mv | AT lerinaaversano whattimeisitx003ffindingwhichtemporalfeaturesismoreusefulfornextactivityprediction AT martinaiammarino whattimeisitx003ffindingwhichtemporalfeaturesismoreusefulfornextactivityprediction AT antonellamadau whattimeisitx003ffindingwhichtemporalfeaturesismoreusefulfornextactivityprediction AT giuseppepirlo whattimeisitx003ffindingwhichtemporalfeaturesismoreusefulfornextactivityprediction AT gianfrancosemeraro whattimeisitx003ffindingwhichtemporalfeaturesismoreusefulfornextactivityprediction |