Machine Learning for Text Categorization to Enhance Emergency Response Efficiency

Abstract

This research paper attempts to improve the emergency response efficiency of an organization by employing text categorization methods to group related requirements from different departments of the organization. The paper evaluates multiple machine learning-based text categorization methods to find which method can best group requirements from various departments that work towards the same goal allowing departments to synchronize operations and work more efficiently. The two methods that will be reviewed in this paper are text-embedding-based clustering and LDA models. Each data sorting method is trained, if necessary, and tested on a dataset of sample requirements from multiple departments of an organization.