Passage-Based Bibliographic Coupling: An Inter-Article Similarity Measure for Biomedical Articles

Area of Science:

• Biomedical Informatics
• Information Retrieval
• Bibliometrics

Background:

• Biomedical literature is crucial for evidence-based research.
• Identifying highly related articles is challenging for current search engines.
• Related articles should share core content like goals, methods, and findings.

Purpose of the Study:

• To develop a technique for improving the retrieval of highly related biomedical articles.
• To address the limitations of existing search engines in finding relevant literature.
• To enhance the process of evidence translation in biomedical studies.

Main Methods:

• A novel technique called Passage-based Bibliographic Coupling (PBC) was developed.
• PBC integrates bibliographic coupling with information from context passages around citations.
• The method estimates inter-article similarity based on shared references and their surrounding text.

Main Results:

• PBC significantly improves the retrieval of articles deemed highly related by biomedical experts.
• The technique demonstrated effectiveness for gene-disease association research.
• Improved retrieval was observed even for articles with minimal or no citations.

Conclusions:

• Passage-based Bibliographic Coupling enhances the ability to find relevant biomedical literature.
• PBC is valuable for researchers and text mining systems needing to cross-validate evidence.
• The method offers a robust solution for discovering related articles, overcoming citation limitations.