PVAF: an environment for disambiguation of scientific publication venues

Abstract

A publication venue authority file stores variants of the names of journals and conferences that publish scientific articles. It is useful in the construction of search tools and data disambiguation, and it is of special interest to agencies funding research and evaluating graduate programs, which use the quality of publication venues as a basis for evaluating researchers’ and research groups’ publications. However, keeping an updated authority file is not a trivial task. Different names are used to refer to the same publication venue, these venues sometimes change their name, new venues emerge regularly, and journal bibliometrics are updated frequently. This paper presents the publication venue authority file (PVAF), an environment for the disambiguation of scientific publication venues. It consists of an authority file and a set of tools for updating and querying its data. We describe and experimentally evaluate each of these tools. We also propose a search algorithm based on an associative classifier, which allows for incremental updates of its learning model. The results show that the PVAF has coverage greater than 86% for publication venues in several fields of knowledge, and its tools attain a good accuracy in the classification of publication venues from curricula vitae formatted in various citation styles.