Scientific paper resulting from the work of Thaís Rodrigues Neubauer. Published at the 1st International Workshop on Computational Intelligence for Process Mining (WCCI 2022), titled “Interactive Trace Clustering to Enhance Incident Completion Time Prediction in Process Mining” and presented by Thais in the conference. The paper features collaboration from researchers MSc. Alexandre Gastaldi L. Fernandes, Prof. Dr. Marcelo Fantinato and Prof. Dr. Sarajane Marques Peres (Thais’ advisor).
Abstract: When it comes to process mining, the more singularities there are in a business process, the more human-in-the-loop strategies are needed. In order to apply discovery, compliance, profile identification, prediction or recommendation algorithms, domain experts are often involved in various tasks, such as pre-processing event logs, parameterizing algorithms and post-processing the results. However, experts’ knowledge is rarely used to directly influence the internal decision making of process mining algorithms. The knowledge-oriented adjustment of the behavior of the algorithm facilitates data analysis, corrects distortions, and produces results more adherent to the expectations of a domain analyst. In this paper, we introduce the approach named interactive trace clustering, in which the human-in-the-loop strategy is implemented through experts’ knowledge based constraint rules, modeled and merged with the decisions
of the k-Means algorithm by means of the Cop-k-Means algorithm. We discuss our proposal through a proof of concept built on an incident completion time prediction problem. A real-world event log was used and the results when applying interactive trace clustering showed the usefulness of our approach.
Complete paper here.
Reference: Neubauer, T. R.; Fernandes, A. G. L.; Fantinato, M.; Peres, S. M. Interactive Trace Clustering to Enhance Incident Completion Time Prediction in Process Mining. In: WCCI 2022 Workshops: CI4PM-22 – 1st International Workshop on Computational Intelligence for Process Mining, 2022, Padua. CEUR Workshop Proceedings, 2022. p. 1-12.
Artigo científico resultante do trabalho de Thaís Rodrigues Neubauer. Publicado no 1st International Workshop on Computational Intelligence for Process Mining (WCCI 2022), intitulado “Interactive Trace Clustering to Enhance Incident Completion Time Prediction in Process Mining” e apresentado por Thais na conferência. O paper conta com a colaboração de MSc. Alexandre Gastaldi L. Fernandes, Prof. Dr. Marcelo Fantinato e Profa. Dra. Sarajane Marques Peres (orientadora).
Artigo completo aqui – em inglês.
Referência: Neubauer, T. R.; Fernandes, A. G. L.; Fantinato, M.; Peres, S. M. Interactive Trace Clustering to Enhance Incident Completion Time Prediction in Process Mining. In: WCCI 2022 Workshops: CI4PM-22 – 1st International Workshop on Computational Intelligence for Process Mining, 2022, Padua. CEUR Workshop Proceedings, 2022. p. 1-12.
