Extracting Absenteeism at Work using Random 2 Satisfiability Modified Reverse Analysis

Abstract

Absenteeism in the workplace can be voluntary or involuntary which significantly impacts organization productivity, employee morale, and operational costs. An artificial neural network will be employed to extract insights on absenteeism helping the employers to understand and mitigate the issue. There are limited attempts that propose neural network model for knowledge extraction in the human resources domain, particularly research revolving around absenteeism of employees. In this paper, a modified evolutionary based algorithm named Modified Niche Genetic Algorithm has been proposed to enhance the training phase of the Discrete Hopfield Neural Network with Random 2 Satisfiability Modified Reverse Analysis method. The performance of the Modified Niche Genetic Algorithm with the Random 2 Satisfiability Modified Reverse Analysis is evaluated using various performance metrics. In light of the findings, the proposed model demonstrated superior performance compared to the traditional classification model with 73% more accuracy in doing knowledge extraction for the human resources field. The contributions in the proposed model provide a robust platform and enhance the capabilities of the classification model.