CHMM for Discovering Intentional Process Model From Event Logs by Considering Sequence of Activities

Kelly R. Sungkono, Riyanarto Sarno

Abstract


An intentional process model is known to analyze processes deeply and provide recommendations for the upcoming processes. Nevertheless, the discovery of intentions is a difficult task because the intentions are not recorded in the event log, but they encourage the executable activities in the event log. Map Miner is the latest algorithm to depict the intentional process model. A disadvantage of this algorithm is the inability to determine   strategies   that   contain   same   activities   with   the different sequence with other strategies. This disadvantage leads failure on the intentional process model. This research proposes an  algorithm for  discovering  an intentional  process  model  by considering the sequence of activities and CHMM (Coupled Hidden Markov Model). The probabilities and states of CHMM are utilized for the formation of the intentional process model. The experiment shows that the proposed algorithm with considering the sequence of activities gets an appropriate intentional process model. It also demonstrates that an obtained intentional  process  model  using  proposed  algorithm  gets  the better  validity  than  an  intentional  process  model  using  Map Miner Method.

Keywords


Coupled Hidden Markov Model; Event Log; Intention Mining; Process Model; Validity;

Full Text: PDF

Refbacks

  • There are currently no refbacks.