Trace Clustering Techniques for Process Mining

Abstract

Process mining connects the fields of process science and data science. Process mining is a set of techniques used for obtaining knowledge by extracting insights from processes by the means of analysing the event logs, generated during the execution of the process. The process models that are generated using the process mining techniques may be structured or semi structured, it may be complex and difficult to comprehend. The complexity can be reduced using trace clustering. In this work, two clustering algorithms the K-means and Hierarchical clustering and applied to event logs before building the process models. The efficiency of these two algorithms on log data is analysed and it is shown that the complexity of the process models is reduced and it is easy to comprehend the models.