ABSTRACT
MOTIVATION: Document similarity metrics such as PubMed's 'Find related articles' feature, which have been primarily used to identify studies with similar topics, can now also be used to detect duplicated or potentially plagiarized papers within literature reference databases. However, the CPU-intensive nature of document comparison has limited MEDLINE text similarity studies to the comparison of abstracts, which constitute only a small fraction of a publication's total text. Extending searches to include text archived by online search engines would drastically increase comparison ability. For large-scale studies, submitting short phrases encased in direct quotes to search engines for exact matches would be optimal for both individual queries and programmatic interfaces. We have derived a method of analyzing statistically improbable phrases (SIPs) for assistance in identifying duplicate content. RESULTS: When applied to MEDLINE citations, this method substantially improves upon previous algorithms in the detection of duplication citations, yielding a precision and recall of 78.9% (versus 50.3% for eTBLAST) and 99.6% (versus 99.8% for eTBLAST), respectively. AVAILABILITY: Similar citations identified by this work are freely accessible in the Déjà vu database, under the SIP discovery method category at http://dejavu.vbi.vt.edu/dejavu/.
Subject(s)
Abstracting and Indexing/methods , Duplicate Publications as Topic , Databases, Factual , MEDLINE , Plagiarism , PubMed , Search Engine , United StatesABSTRACT
In the scientific research community, plagiarism and covert multiple publications of the same data are considered unacceptable because they undermine the public confidence in the scientific integrity. Yet, little has been done to help authors and editors to identify highly similar citations, which sometimes may represent cases of unethical duplication. For this reason, we have made available Déjà vu, a publicly available database of highly similar Medline citations identified by the text similarity search engine eTBLAST. Following manual verification, highly similar citation pairs are classified into various categories ranging from duplicates with different authors to sanctioned duplicates. Déjà vu records also contain user-provided commentary and supporting information to substantiate each document's categorization. Déjà vu and eTBLAST are available to authors, editors, reviewers, ethicists and sociologists to study, intercept, annotate and deter questionable publication practices. These tools are part of a sustained effort to enhance the quality of Medline as 'the' biomedical corpus. The Déjà vu database is freely accessible at http://spore.swmed.edu/dejavu. The tool eTBLAST is also freely available at http://etblast.org.
