Scientific paper resulting from the work of Alexandre Gastaldi L. Fernandes. Published at the 1st International Workshop on Computational Intelligence for Process Mining (WCCI 2022), titled “Impact of non-fitting cases for remaining time prediction in a multi-attribute process-aware method” and presented by Thais in the conference. The paper features collaboration from researchers MSc. Thais Rodrigues Neubauer, Prof. Dr. Marcelo Fantinato (Alexandre’s advisor) and Prof. Dr. Sarajane Marques Peres.
Abstract: Several studies have shown valuable results in remaining time prediction. However, the analysis of nonfitting cases and their impact on the prediction accuracy have been carried out superficially. Non-fitting cases are those for which there is no full match for a new case presented to the predictor. We analyzed the impact of non-fitting cases on a remaining time prediction process-aware method based on an annotated transition system (ATS) using multiple descriptive attributes. The results showed that, as the number of attributes added to the ATS-based predictor increases, the number of non-fitting cases increases rapidly. Increasing the maximum horizon and the state representation also influences the number of non-fitting cases, which reach over 90% in complex scenarios. To reduce the impact of non-fitting cases, the effectiveness of similarity techniques was analyzed. About 60% error reduction can be achieved with high model specialization, for all state representations, mainly for multi-sets and sequences.
Complete paper here
Reference: Fernandes, A. G. L.; Neubauer, T. R.; Fantinato, M. ; Peres, S. M. Impact of non-fitting cases for remaining time prediction in a multi-attribute process-aware method. In: WCCI 2022 Workshops: CI4PM-22 – 1st international workshop on computational intelligence for process mining, 2022, Padua. CEUR Workshop Proceedings, 2022. p. 1-23.
Artigo científico resultante do trabalho de Alexandre Gastaldi L. Fernandes. 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. Thais Rodrigues Neubauer, Prof. Dr. Marcelo Fantinato e Profa. Dra. Sarajane Marques Peres (orientadora).
Artigo completo aqui – em inglês
Referência: Fernandes, A. G. L.; Neubauer, T. R.; Fantinato, M. ; Peres, S. M. Impact of non-fitting cases for remaining time prediction in a multi-attribute process-aware method. In: WCCI 2022 Workshops: CI4PM-22 – 1st international workshop on computational intelligence for process mining, 2022, Padua. CEUR Workshop Proceedings, 2022. p. 1-23.
