Intelligent artificial agent for information retrieval

Abstract

Throughout the day of the average employee at RS2, there will often be a need to search one of the company’s information repositories. Finding the information will often force employees to perform a context switch and search within the appropriate repository. We propose a system that will facilitate this process by creating a ChatBot that can perform the search within the company’s chat client by making use of the latest machine learning techniques, alongside several NLP techniques and established industry standard information retrieval technologies to allow for a single consolidated, optimised searching system. Results on benchmark datasets show that our system was able to achieve the best results when making use of a combination of traditional and modern techniques.