Rapid bibliometric analysis in deep learning domain

Abstract

The paper systematizes the Deep Learning domain and calculates the dynamics of changes in the number of scientific articles according to Google Scholar. The method of data acquisition and calculation of dynamic indicators of changes in publication activity is described: speed (DI) and acceleration of growth (D2) of scientific publications. Analysis of publication activity, in particular, showed a high interest in modern transformer models, the development of datasets for some industries, and a sharp increase in interest in methods of explicable machine learning. Relatively small research domains are receiving increasing attention, as evidenced by the negative correlation between the number of articles and Dl and D2 scores. The results show that, despite the limitations of the method, it is possible to identify fast-growing areas of research regardless of the number of articles. The paper presents result for more than 400 search queries related to classified research areas. Calculation results and software can be downloaded https://www.dropbox.com/sh/fkfw3a1hkf0suvc/AACRZ7v9qympen_ht00jeiF6a?dl=0.