A bottom-up approach to MEDLINE indexing recommendations

Antonio Jimeno-Yepes, Bartlomiej Wilkowski, James G. Mork, Elizabeth Van Lenten, Dina Demner Fushman, Alan R. Aronson

AbstractMEDLINE indexing performed by the US National Library of Medicine staff describes the essence of a biomedical
publication in about 14 Medical Subject Headings (MeSH). Since 2002, this task is assisted by the Medical Text
Indexer (MTI) program. We present a bottom-up approach to MEDLINE indexing in which the abstract is searched for
indicators for a specific MeSH recommendation in a two-step process. In the first step, a rule-based triage significantly
reduces the number of candidate citations to which the MeSH heading is recommended. In the second step, the
candidate citation list is further reduced using supervised machine learning. Supervised machine learning combined
with triage rules improves sensitivity of recommendations while keeping the number of recommended terms relatively
small. Improvement in recommendations observed in this work warrants further exploration of this approach to MTI
recommendations on a larger set of MeSH headings.
KeywordsMeSH, indexing, MTI, machine learning
TypeConference paper [Submitted]
ConferenceAMIA 2011
Year2011    Month March
BibTeX data [bibtex]
IMM Group(s)Intelligent Signal Processing