1. Rey-Long Liu Dept. of Medical Informatics Tzu Chi University Taiwan
Identification of Biomedical Articles with Highly Related Core Contents
Rey-Long Liu
Dept. of Medical Informatics
Tzu Chi University
Taiwan

2. Outline Background Problem definition
The proposed technique: CCSE (Core Content Similarity Estimation)
Empirical evaluation
Conclusion

3. Background

4. Core Contents of Biomedical Articles
Core contents of a scholarly article a are the textual contents about
Research goal of a
Research background of a
Research conclusion of a

5. Similarity Estimation for the Core Contents
Goal: retrieval of highly related articles
Mining & analysis of highly related evidence
A typical goal of existing search engines
Challenge: recognition of the core contents
Core content of an article a may be briefly expressed in the title and scattered in the abstract

6. Selected by biomedical experts for <erythropoietin, anemia> 
They are highly related to each other
Recommended by PubMed, but NOT highly related to <erythropoietin, anemia> 6

7. Problem Definition

8. Goal Developing a technique CCSE (Core Content Similarity Estimation)
Given: titles and abstracts of two articles a1 and a2
Output: core content similarity between a1 and a2

9. Contributions
CCSE works on titles and abstracts only, which are publicly available
CCSE improves inter-article similarity estimation by considering the core contents of the articles

10. Related Work Inter-article similarity based on citation links
Example I: out-link citations (by bibliographic coupling, BC)
Example II: in-link citations (by co-citation, CC)
Weakness:
The citation links are often not available on the Internet (many articles even have no in-link citations)

11. Related Work (cont.)
Inter-article similarity based on textual contents
Working on publicly available parts
Titles and abstracts
Considering weights of terms (e.g., TFIDF weight)
Weakness:
Did not consider the core content similarity (due to the difficulty in recognizing the core contents)

12. The Proposed Technique: CCSE

13. Main Ideas How are goal terms of a1 related to the goal of a2?
Title of a1
Title of a2
Abstract of a1
Abstract of a2
How are background and conclusion terms of a1 related to the background and conclusion of a2, respectively?

14. Main Ideas (cont.)
Three kinds of relatedness of a term t the core content of an article a,
Rgoal: Relatedness to goal
Rback: Relatedness to background
Rconc: Relatedness to conclusion
Rgoal, Rback, and Rconc are estimated based on the positions of t in the title and the abstract of a

15. Step 1/2: Estimation of Rgoal, Rback, and Rconc based on positions of the term:

16. Step 2/2: Estimating inter-article similarity between two articles a1 and a2
 Any mismatch between the core contents will significantly reduce the inter-article similarity
 Similarity between a1 and a2 is based on goal match, background match, and conclusion match between a1 and a2

17. Min
 Matchgoal(a1, a2) is based on how terms in the title of a1 is related to the goals of a1 and a2

18. Min
 Matchback(a1, a2) is based on how terms in the abstract of a1 is related to the backgrounds of a1 and a2

19. Min
 Matchconc(a1, a2) is based on how terms in the abstract of a1 is related to the conclusions of a1 and a2

20. Interesting Features of CCSE
Inter-article similarity is composed of three parts
Goal similarity
Background similarity
Conclusion similarity
These similarities are estimated based on the positions of the terms appearing in the title and the abstract of the article
Any mismatch between the core contents will significantly reduce the inter-article similarity

21. Empirical Evaluation

22. The data Two sets of articles Highly related biomedical articles:
For each gene-disease pair <g,d>, collect the biomedical articles that biomedical experts selected to annotate the pair (noted by DisGeNET)
Near-miss biomedical articles (Non-highly related articles):
For each gene-disease pair <g,d>, collect articles using two queries: “g NOT d” and “d NOT g”

23. Data statistics 53 gene-disease pairs 9,875 articles, including
53 targets + 9,822 candidates
435,786 out-link references

24. The Baseline Systems (1) Link-based inter-article similarity
Bibliographic coupling (BC)
(2) Text-based inter-article similarity
BM25 (one of the best in the biomedical domain)
(3) Biomedical search engine
PubMed (popular and one of the best in the biomedical domain)

25. Evaluation Criteria MAP (Mean Average Precision)
If a system can rank higher those articles that are highly related to r, average precision (AvgP) for the gene-disease pair will be higher
MAP is simply the average of the AvgP values for all gene-disease pairs

26. Average
If those articles that are highly related to r, are ranked at top-X position, for the gene-disease pair will be higher
Average is simply the average of the values for all gene-disease pairs

27. Result
CCSE performs significantly better than BC and BM25

28. CCSE performs better than PubMed in all evaluation criteria:

29. Conclusion

30. Our Motivation: We develop CCSE that The idea of CCSE can be
Core contents of scholarly articles are essential for retrieval of highly related scientific evidence, BUT
The core contents are scattered in titles and abstracts of articles
We develop CCSE that
Estimates inter-article similarity based on the similarities in goals, backgrounds, and conclusions of two articles
The idea of CCSE can be
Incorporated into search engines to properly retrieve highly related scholarly articles