Resumen
The article discusses issues related to the use of data science and data mining methods for software repositories. The paper attempts to provide an overview of the technologies that are used in the analysis of programs and are based on static data that can be extracted directly from the code or the code repositories. The paper reviews papers using deep learning methods (recurrent neural networks), classification methods based on other machine learning models, and the use of clustering in software engineering. Practical applications of the methods under consideration include, for example, classification and prediction of errors, determining the characteristics of code change over time, searching for duplicate fragments, automatically detecting design errors, recommending code refactoring.