The existing literature on software development effort estimation is extensive and focuses particularly on traditional software projects, while few studies have been devoted to agile projects, especially estimating the effort needed for completing the whole software projects. Story points and team velocity are among the commonly effort drivers used for planning and predicting when a software will be completed. In this paper, we propose an improved model for estimating the software effort based on support vector regression (SVR) optimized by grid search method (GS). The story point and velocity were used as inputs of the prediction model. The empirical evaluation is carried out using 21 historical agile software projects through leave-one-out cross validation method. The obtained results demonstrate that our approach was able to improve the performance of SVR technique. Moreover, it outperforms, in terms of Pred(0.25), MMRE and MdMRE, some recent methods reported in the recent literature.

Accurate and reliable software development effort estimation (SDEE) is one of the main concerns for project managers. Planning and scheduling a software projects using and inaccurate estimate may cause severe risks to software project under development such as delayed delivery, poor quality software, missing features. Therefore, accurate prediction of software effort plays an important role in the minimization of these risks that can lead to projects failure. Nowadays, application of artificial intelligence techniques has grown dramatically for predicting software effort. The researchers found that these techniques are suitable tools for accurate prediction. In this study, an improved model is designed for estimating software effort using support vector regression (SVR) and two feature selection (FS) methods. Prior to building model step, a preprocessing stage is performed by random forest or Boruta feature selection methods to remove unimportant features. Next, the SVR model is tuned by a grid search approach. The performance of model is then evaluated over eight well-known datasets through 30%holdout validation method. To show the impact of feature selection on the accuracy of SVR model, the proposed model was compared with SVR model without feature selection. The results indicated that SVR with feature selection outperforms SVR without FS in terms of the three accuracy measures used in this empirical study.

The decision tree (DT) represents a nonparametric estimation method that has been mostly used for both classification and regression problems. DTs were adopted for software development effort estimation (SDEE) generally for their simplicity of use and interpretation contrary to other learning methods. Nevertheless, to our self-knowledge, no systematic mapping has been devoted especially to decision trees. The aim of this study is to elaborate a systematic mapping study that classifies DTs papers in conformity with the succeeding criteria: research approach, contribution type, techniques employed in combination with DT methods besides identifying publication channels and trends. An automated search of five digital libraries was made to carry out a systematic mapping of DT studies mainly devoted to SDEE that were published in the period 1985–2017. We identify 46 relevant studies. Basically, the results revealed that most researchers focus on technique contribution type. In addition, the majority of papers deal with improving the existing DT models while few studies have proposed novel models to improve the reliability of SDEE. Furthermore, solution proposal and case study are the most frequently used approaches.

Computing advances in data storage are leading to rapid growth in large-scale datasets. Using all features increases temporal/spatial complexity and negatively influences performance. Feature selection is a fundamental stage in data preprocessing, removing redundant and irrelevant features to minimize the number of features and enhance the performance of classification accuracy. Numerous optimization algorithms were employed to handle feature selection (FS) problems, and they outperform conventional FS techniques. However, there is no metaheuristic FS method that outperforms other optimization algorithms in many datasets. This motivated our study to incorporate the advantages of various optimization techniques to obtain a powerful technique that outperforms other methods in many datasets from different domains. In this article, a novel combined method GASI is developed using swarm intelligence (SI) based feature selection techniques and genetic algorithms (GA) that uses a multi-objective fitness function to seek the optimal subset of features. To assess the performance of the proposed approach, seven datasets have been collected from the UCI repository and exploited to test the newly established feature selection technique. The experimental results demonstrate that the suggested method GASI outperforms many powerful SI-based feature selection techniques studied. GASI obtains a better average fitness value and improves classification performance.