How can we spot a software citation in an academic paper?
The work reported in many scientific papers now relies on software, but the tools, libraries, and data analysis scripts used for producing results often remain hidden. There are a number of reasons for this, but one of the primary issues is that there are no definitive guidelines for citing software, or expectations that software should be described in full. Because of this, it is very difficult to understand whether software is being cited, and if it is, the form that this citation takes. This project aims to investigate methods for mining software citations, which can then be used by researchers in the field for understanding and improving the way scientists cite software.
The project will focus on mining references to software usage via academic library APIs, supported by an interview study. The objective is to produce one or more robust methods that can be used for locating software references within an acadmic paper, that can be used by researchers for determining the prevalence of software citation in different fields.