Comparative Analysis of Machine Learning Techniques in Effort Estimation

In Software engineering effort estimation provides an important role for software development and managing project cost, quality, and time. Since last decades, software estimation has been receiving significant attention from researchers and substantial research has been performed using various techniques and algorithms of machine learning. This paper suggests different machine learning techniques such as Naïve Bayes, Random Forests Logistic Regression, stochastic gradient boosting, decision tree, and story point for estimation to assess prediction more efficiently. Nowadays uses of these approaches by software development industries for software estimation aim to tackle deficiencies of parametric and traditional estimation techniques, rise project. A comparative study of the mentioned techniques is presented and examined in this paper to critically evaluate the performance of these techniques.