An Empirical Investigation of Different Classifiers, Encoding, and Ensemble Schemes for Next Event Prediction Using Business Process Event Logs

Abstract

There is a growing need for empirical benchmarks that support researchers and practitioners in selecting the best machine learning technique for given prediction tasks. In this article, we consider the next event prediction task in business process predictive monitoring, and we extend our previously published benchmark by studying the impact on the performance of different encoding windows and of using ensemble schemes. The choice of whether to use ensembles and which scheme to use often depends on the type of data and classification task. While there is a general understanding that ensembles perform well in predictive monitoring of business processes, next event prediction is a task for which no other benchmarks involving ensembles are available. The proposed benchmark helps researchers to select a high-performing individual classifier or ensemble scheme given the variability at the case level of the event log under consideration. Experimental results show that choosing an optimal number of events for feature encoding is challenging, resulting in the need to consider each event log individually when selecting an optimal value. Ensemble schemes improve the performance of low-performing classifiers in this task, such as SVM, whereas high-performing classifiers, such as tree-based classifiers, are not better off when ensemble schemes are considered.

A Survey on Event Prediction Methods from a Systems Perspective: Bringing Together Disparate Research Areas

“…The event prediction search pillar includes eleven surveys [20,109,120,130,135,205,209,230,250,281,283]. Each of these surveys has a certain scope, e.g., [205,209] focus on time-series data, [281,283] focus on event prediction of business processes, and [20] on unstructured text events.…”

“…The event prediction search pillar includes eleven surveys [20,109,120,130,135,205,209,230,250,281,283]. Each of these surveys has a certain scope, e.g., [205,209] focus on time-series data, [281,283] focus on event prediction of business processes, and [20] on unstructured text events. When compared to the survey at hand, on top of a narrow scope, the eleven surveys neither take a system perspective to generally place event prediction method in a context, nor elicit requirements for assessing the methods, nor comprehensively assess existing work.…”

A Road-map for Mining Business Process Models via Artificial Intelligence Techniques

“…Definition 1. (Business Process) A BP is simply defined as: A set of activities undertaken by one or more organizations in pursuit of some particular business goals [9].…”

Leveraging a Heterogeneous Ensemble Learning for Outcome-Based Predictive Monitoring Using Business Process Event Logs

“…Predictive process monitoring of business processes has emerged in the last ten years and aims at extracting insights about business processes by building predictive models using event log data [2]. There are several aspects of a process that can be predicted, such as the timestamp of next events [3], the future activities that will be executed in a process case [4][5][6], or the outcome of a process case [7]. Predictive monitoring enables proactive decision making, such as informing customers that their request may be processed later than expected or addressing the possible occurrence of an unfavorable exception by implementing protective measures.…”