TY - JOUR AU - Abubakar, Shamsuddeen M. AU - Sufyanu, Zahraddeen AU - Abubakar, Miyim M. PY - 2020/04/13 Y2 - 2024/03/29 TI - A SURVEY OF FEATURE SELECTION METHODS FOR SOFTWARE DEFECT PREDICTION MODELS JF - FUDMA JOURNAL OF SCIENCES JA - FJS VL - 4 IS - 1 SE - Research Articles DO - UR - https://fjs.fudutsinma.edu.ng/index.php/fjs/article/view/18 SP - 62 - 68 AB - Feature selection is a technique used to select an optimal feature subset from the original input features according to a specific criterion. The criterion is often formulated as an objective function that finds which features are most appropriate for some tasks at hand. The reason why it is interesting to find a subset of features is because that it always easier to solve a problem in a lower dimension. This helps in understanding the nonlinear mapping between input and output variables. This paper reviewed the basic Feature Selection Techniques for Software Defect Prediction Model and their domain applications. The Subsets selection are categorized into three distinct models and are discussed in a concise form to provide young researchers with the general methods of Subset Selection. Support Vector Machine with Recursive Feature Elimination for both Logistic Regression and Random Forest was introduced to evaluate the performance between filter, wrapper, and embedded feature selection techniques. Hence, the research proposes an Embedded Feature Selection techniques for consistency of a subset of software metrics analysis. ER -