Automatic Workload Estimation for Software House

Abstract

Normally, organizations have to estimate the workload relying on limited resources. An appropriate estimation method can improve workforce optimization. In the software house, workload categorization and estimation can be acquired from the information technology management. Nevertheless, there are several factors such as work priority and specific goals that affect the workload level. General workload management spends a long time and decreases task management quality. Therefore, this study applies machine learning, Naïve Bayes, to estimate the workload automatically. This classification method increases the accuracy of workload estimation, along with reducing the time consumption for the whole system.