ARTS: autonomous research topic selection system using word embeddings and network analysis

Abstract The materials science research process has become increasingly autonomous due to the remarkable progress in artificial intelligence. However, topic selection (ARTS) not yet been fully explored difficulty of estimating its promise and lack previous research. This paper introduces an ARTS system that autonomously selects potential topics are likely reveal new scientific facts have subject much by analyzing vast numbers articles. Potential selected difference between two concept networks constructed from information articles: one represents is word embeddings, known past activities statistical on appearance patterns concepts. also equipped with functions search visualize about assist final determination a scientist. We developed using approximately 100 00 articles published Computational Materials Science journal. results our evaluation demonstrated studied after 2016 could be generated analysis before 2015. suggests can effectively system.