Current research themes in software engineering: An application of text mining

Abstract

This paper presents the findings of a research project concerned with extracting the themes currently being researched within the Software Engineering field by automatically processing a collection of papers published thus far in 2015. Natural Language Processing techniques have been applied to extract the key features of each publication, with publications then being clustered with similar papers to identify their common themes and highlight the key themes currently being researched within the Software Engineering domain. The most dominant active theme was Education, with a cluster of papers concerning the teaching of software engineering, while other relevant topics included Components and Requirements.