An enhanced author name dataset for PubMed/MEDLINE

The incompleteness of author names is a well-known issue in the MEDLINE database. It was since 2002, the full author name has been systematically indexed in MEDLINE. Although the full author name has been systematically indexed since 2002, we still found a significant number of initialized names in papers published after 2002. In response to this issue, we built an enhanced author name dataset called EAN for MEDLINE, which is achieved by linking the whole PubMed to other large literature databases and performing name comparison and restoration based on multi-sources author names. Our evaluation shows that 91% of author names in EAN are complete as compared to the completeness ratio of 56% of the original author names in MEDLINE, demonstrating that the issue has been largely resolved.