Özet:

Abstract In the period of computerized media, the quickly expanding volume and intricacy of sight and sound information cause numerous issues in putting away, handling, and questioning data in a sensible time. Huge storehouses of source code set out new difficulties and open doors for factual machine learning. Here we initially foster Sourcerer, a foundation for the mechanized slithering, parsing, and data set capacity of open source programming. This research propose novel technique in software Repositories for data mining and data analytics in feature extraction and classification using deep learning. here the software Repositories for data mining is carried out based on Markov Chain Monte Carlo model. Then the data analytics has been carried out for feature extraction and classification using heuristic Gaussian bayes neural network with principal component analysis. The experimental analysis has been carried out for various dataset in terms of accuracy, precision, recall, MSE, MAP.The proposed technique attainedaccuracy of 96%, precision of 85%, recall of 79%, MSE of 66%, MAP of 63%.

Anahtar Kelimeler: