An efficient programming rule extraction and detection of violations in software source code using neural networks

The larger size and complexity of software source code builds many challenges in bug detection. Data mining based bug detection methods eliminate the bugs present in software source code effectively. Rule violation and copy paste related defects are the most concerns for bug detection system. Traditional data mining approaches such as frequent Itemset mining and frequent sequence mining are relatively good but they are lacking in accuracy and pattern recognition. Neural networks have emerged as advanced data mining tools in cases where other techniques may not produce satisfactory predictive models. The neural network is trained for possible set of errors that could be present in software source code. From the training data the neural network learns how to predict the correct output. The processing elements of neural networks are associated with weights which are adjusted during the training period.