Abstract Process discovery is a method for attaining process scheme relies on traces exists in the event log. Today, information systems generate the streaming event logs to store their enormous processes. The truncated event log streaming is a challenging problem in process detection since it accuses an incomplete traces, which makes the inaccurate process in a process model. Several conventional techniques have been introduced for retrieving the truncated streaming of event log. This research proposes a method, namely Fractional Improved Invasive Lion Algorithm (FrIILA) for performing the concept drift handling on event log. For that, the event log data is processed under process dimension trimming using bounding model. Moreover, the process mining is carried out using developed FrIILA, which is deliberated by the integration of Fractional calculus (FC) and Improved Invasive Lion Algorithm (IILA). For the incremental data, the same processing is carried out for determining the process discovery. Here, the concept drift detection is carried out using two conditions, such as new event label and max min position of trace. The experimental outcome demonstrates that the devised method achieved better performance based on the replayability and precision of 98.01% and 80.39%.
Alan : Mühendislik
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|