关键词: 软件缺陷预测, 神经网络, 粒子群优化

Abstract: Software defect detection aims to automatically identify defective software modules that accelerates efficiently software test and improves the quality of a software system. Due to the application of traditional software prediction model being limited for its low accuracy and applicability, this paper puts forward a software prediction model based on PSO-BP, which employs Particle Swarm Optimization（PSO） to optimize weight and threshold value of BP. It uses cross-validation method as experiment method, and compares the results with other machine learning methods-BP and J48. The results show that PSO-BP has higher prediction accuracy.

Key words: software defect prediction, Artificial Neural Network（ANN）, Particle Swarm Optimization（PSO）