Employment law expert system

Seeam, Preetila, Teckchandani, Nishant, Booneyad, Hansha, Torul, V. P. and Seeam, Amar ORCID logoORCID: https://orcid.org/0000-0001-8203-1545 (2018) Employment law expert system. 2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC). In: 2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC), 06-07 Dec 2018, Mon Tresor, Mauritius. e-ISBN 9781538664773, e-ISBN 9781538664766, pbk-ISBN 9781538664780. [Conference or Workshop Item] (doi:10.1109/ICONIC.2018.8601271)


This paper reviews the development and application architecture of an expert system to assist the Mauritian population with queries they may have about labor or employment law. The expert system makes use of Machine Learning, Speech Recognition/Synthesis and Natural Language Processing techniques to converse with users through a web interface. The expert system also takes advantage of a large knowledge base, that allows the system to teach itself employment law principles. The knowledge base is created from "Understanding Employment Law and Remuneration Orders in Mauritius", written by Ved Prakash Torul [1], which is a simplified version of the Employment Relations Act and the Employment Rights Act. The book explains employment law in common language, to help the public understand their constitutional rights. The expert system allows users to communicate and express their employment issues, so that they are aware of their next course of action, either they are an employer, employee, or a union. The paper also reviews the evaluation period, which consisted of a preliminary testing period. Through the evaluation, it was concluded that the expert system was able to respond to individual responses with a Precision of 66% and Recall of 85%. While the Expert System is able to converse with users on certain topics on Employment Law, further evaluation would need to be conducted. Additionally, the knowledge base will need to be updated over time.

Item Type: Conference or Workshop Item (Paper)
Research Areas: A. > School of Science and Technology > Computer Science
Item ID: 36109
Depositing User: Amar Kumar Seeam
Date Deposited: 05 Oct 2022 11:19
Last Modified: 16 Nov 2022 16:18
URI: https://eprints.mdx.ac.uk/id/eprint/36109

Actions (login required)

View Item View Item


Activity Overview
6 month trend
6 month trend

Additional statistics are available via IRStats2.