Shameer Pradhan - Which Tables are Mine(able)?

Process mining is a powerful set of techniques used to improve operational processes by leveraging event data typically stored in information systems like Enterprise Resource Planning (ERP) software. However, these algorithms cannot directly analyze data from such systems; the data must first be extracted and transformed into an event log during a crucial pre-analysis phase known as event log-building. This phase, which involves extensive data extraction and transformation, often adheres to the "80-20" principle: 80% of the effort is spent on data preparation, while only 20% is dedicated to actual analysis. Although various solutions have been proposed to automate the data extraction process, they often fail to address the complexity of navigating large databases that contain thousands of tables, only a fraction of which are relevant to a specific process mining project. As a result, practitioners must manually identify the relevant tables, a time-consuming and error-prone task, especially when they are unfamiliar with the database's structure. To overcome this challenge, we introduce a novel algorithm designed to semi-automatically identify database tables that are most likely linked to business documents and their associated transactions, thus streamlining the event log-building process in process mining projects.

Shameer is a pre-doctoral researcher in the Business Informatics Research Group at Hasselt University. He holds a Master’s degree in Software Engineering from Chalmers University of Technology, Sweden. His current research is centered on the pre-analysis phase of process mining, with a particular focus on data preprocessing. This includes leveraging natural language querying and large language models for data extraction and event log construction.

Beyond his academic work, Shameer is an active member of the UHasselt Ph.D. Council, where he serves as an executive member and vice president. His contributions to both academia and the professional community reflect his commitment to advancing the field of process automation.