Philippine E-Journals

Business Process Discovery from Emails: Text Classification and Process Mining - A Case Study of Procurement Process

Yaghoub Rashnavadi | Sina Behzadifard | Sina Zamani | Reza Farzadnia

 

Abstract:

Messages and emails are traveling with the speed of light and making communication more accessible than ever, which has transformed organizations to the degree that they generate billions of emails daily to facilitate their operations and processes. This vast corpus of human-generated content is a rich dataset that can benefit businesses. To address the potential application of such data, we propose a framework to mine and extract the implicit information behind the email loops. This article examines the opportunity that email logs can bring to organizations and proposes a framework to discover process models based on a supervised machine learning technique to classify emails to the activities and Fuzzy Miner to extract the process model from the labeled emails. We also examined the framework with a real-life dataset from the procurement department of the case study company in Iran. The findings demonstrated discrepancies between the discovered process model and the designed business process, highlighting the needed improvements.