Software effort estimation using machine learning techniques

Estimation of desired effort is one of the most important activities in software project management. This paper presents an approach for estimation based upon machine learning techniques for non-quantitative data and is carried out in two phases. The first phase concentrates on selection of optimal feature set in high dimensional data, related to projects undertaken in past. A quantitative analysis using Rough Set Theory is performed for feature reduction. The second phase estimates the effort based on the optimal feature set obtained from first phase. The estimation is carried out differently by applying Naive Bayes Classifier and Artificial Neural Network techniques respectively. The feature reduction process in first phase considers public domain data (USP05). The performance of the proposed methods is evaluated and compared based on the parameters such as Mean Magnitude of Relative Error (MMRE), Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Correlation Coefficient. It is observed that Naive Bayes classifier achieved better results for estimation when compared with that by using Neural Network technique.