An important goal during the cycle of software development is to find and fix existing defects as early as possible. Although, big software development companies have their own development repository, which typically includes a version control system and a bug tracking system. There is no doubt that these systems proved useful for software defects prediction. Still, lot of work has to be done for software defect prediction and management. In this paper, two models have been proposed for defect prediction: Support Vector Machine (SVM) is combined with Principle Component Analysis (PCA) and Adaptive Boost of Support Vector Machine (SVM) - Radial Basis Function (RBF) is combined with Principle Component Analysis (PCA).