Co-clustering for enhancing interpretability in process mining: exploring frequency-based and semantic representations

• Grant number: 17/26487-4
• Support Opportunities: Scholarships abroad - Research
• Start date: August 01, 2018
• End date: July 31, 2019
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Sarajane Marques Peres
• Grantee: Sarajane Marques Peres
• Host Investigator: Hajo Alexander Reijers
• Host Institution: Escola de Artes, Ciências e Humanidades (EACH). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Institution abroad: University Amsterdam (VU), Netherlands

Abstract

Process models are essential tools for achieving success in business management in organizations. However, owing to cultural reasons or a lack of adequate human and material resources, it is common for organizations not to formalize these models and they are thus sometimes unaware of the actual process they are carrying out in day--to--day operations. In view of this, process mining plays a central and strategic role, since it provides the means for organizational processes to be automatically discovered, analyzed and enhanced. However, the inherent complexity of organizational processes -- especially those that are unstructured -- prevents the automated process mining from being undertaken with complete success, both in terms of producing useful results and improving the interpretability of the disclosed information. The purpose of this project is to ensure better results are obtained from process mining, as well as the right conditions required to refine them, and hence, increasing their interpretability. To this end, it is recommended that the partial similarities between process instances should be explored through the application of co-clustering methods to process representations based on simple counting and representations able to explore the semantic context of the descriptive attributes of traces. This strategy is expected to disclose refined information on trace profiles as well as handling the concept drift phenomenon, which is marginally exploited in the process mining field. The whole development of this project will be tested by means of synthetic event logs, as proof of concept, and real-life business processes event logs, so that the applicability and scalability of the solutions can be verified.

Scientific publications

(4)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

NEUBAUER, THAIS RODRIGUES; PERES, SARAJANE MARQUES; FANTINATO, MARCELO; LU, XIXI; REIJERS, HAJO ALEXANDER. Interactive clustering: a scoping review. ARTIFICIAL INTELLIGENCE REVIEW, v. 54, n. 4, p. 2765-2826, APR 2021. (17/26487-4, 17/26491-1)

CASTRO, CAMILA F.; FANTINATO, MARCELO; AKSU, UNAL; REIJERS, HAJO A.; THOM, LUCINEIA H.; FILIPE, J; SMIALEK, M; BRODSKY, A; HAMMOUDI, S. Towards a Conceptual Framework for Decomposing Non-functional Requirements of Business Process into Quality of Service Attributes. PROCEEDINGS OF THE 21ST INTERNATIONAL CONFERENCE ON ENTERPRISE INFORMATION SYSTEMS (ICEIS 2019), VOL 2, v. N/A, p. 12-pg., 2019-01-01. (17/26491-1, 17/26487-4)

SILVA, THANNER SOARES; AVILA, DIEGO TORALLES; FLESCH, JEAN AMPOS; PERES, SARAJANE MARQUES; MENDLING, JAN; THOM, LUCINEIA HELOISA; IEEE. A Service-Oriented Architecture for Generating Sound Process Descriptions. 2019 IEEE 23RD INTERNATIONAL ENTERPRISE DISTRIBUTED OBJECT COMPUTING CONFERENCE (EDOC), v. N/A, p. 10-pg., 2019-01-01. (17/26487-4)

BORGES, EVANDO S.; FANTINATO, MARCELO; AKSU, UNAL; REIJERS, HAJO A.; THOM, LUCINEIA H.; FILIPE, J; SMIALEK, M; BRODSKY, A; HAMMOUDI, S. Monitoring of Non-functional Requirements of Business Processes based on Quality of Service Attributes of Web Services. PROCEEDINGS OF THE 21ST INTERNATIONAL CONFERENCE ON ENTERPRISE INFORMATION SYSTEMS (ICEIS 2019), VOL 2, v. N/A, p. 8-pg., 2019-01-01. (17/26491-1, 17/26487-4)