A Multi-model Software Defect Recommendation System

The proposed recommendation system has integrated data collection framework, which can collect repository source data and thus generate base dataset for machine learning and artificial neural network models. So proposed framework provides an end-to end solution for software defect recommendation on python projects in GitHub repository. Recommendation system is comprised of multiple components like data collection system, hybrid feature selection strategy, machine learning models and deep learning models, commit monitor. Framework has been validated on several projects from GitHub repository and outcome of recommendation is favourable for usage in real projects.