WikiCSSH: Extracting and Evaluating Computer Science Subject Headings from Wikipedia

TL;DR

This paper presents WikiCSSH, a large-scale hierarchical computer science vocabulary derived from Wikipedia, using a human-in-the-loop process that combines data extraction, community detection, machine learning, and heuristics, outperforming existing vocabularies.

Contribution

The work introduces a novel workflow for building domain-specific hierarchical vocabularies from Wikipedia, demonstrated through the creation of WikiCSSH for computer science.

Findings

WikiCSSH outperforms existing CS vocabularies in size and key-phrase extraction.

The vocabulary distinguishes between coarse and fine-grained CS concepts.

The workflow can be adapted to other domains in Wikipedia.

Abstract

Hierarchical domain-specific classification schemas (or subject heading vocabularies) are often used to identify, classify, and disambiguate concepts that occur in scholarly articles. In this work, we develop, apply, and evaluate a human-in-the-loop workflow that first extracts an initial category tree from crowd-sourced Wikipedia data, and then combines community detection, machine learning, and hand-crafted heuristics or rules to prune the initial tree. This work resulted in WikiCSSH; a large-scale, hierarchically organized vocabulary for the domain of computer science (CS). Our evaluation suggests that WikiCSSH outperforms alternative CS vocabularies in terms of vocabulary size as well as the performance of lexicon-based key-phrase extraction from scholarly data. WikiCSSH can further distinguish between coarse-grained versus fine-grained CS concepts. The outlined workflow can serve as a template for building hierarchically-organized subject heading vocabularies for other domains that are covered in Wikipedia.