Retrieval of highly related biomedical references by key passages of citations

Abstract

Biomedical researchers often need to carefully identify and read multiple articles to exclude unproven or controversial biomedical evidence about specific issues. These articles thus need to be highly related to each other. They should share similar core contents, including research goals, methods, and findings. However, given an article r, existing search engines and information retrieval techniques are difficult to retrieve highly related articles for r. We thus present a technique KPC (key passage of citations) that extracts key passages of the citations (out-link references) in each article, and based on the key passages, estimates the similarity between articles. Empirical evaluation on over ten thousand biomedical articles shows that KPC can significantly improve the retrieval of those articles that biomedical experts believe to be highly related to specific articles. The contribution is of practical significance to the writing, reviewing, reading, and analysis of biomedical articles.