Background:

For more than 150 years, the National Library of Medicine has provided access to biomedical journal literature through the analytical efforts of human indexers. Since 1966, access has been provided in the form of electronically searchable document surrogates consisting of bibliographic citations, descriptors assigned by indexers from the MeSH controlled vocabulary and, since 1974, author abstracts of many, but not all, items. The objective of the Indexing Initiative project work is to investigate methods whereby automated indexing methods partially or completely substitute for current indexing practices. The project will be considered a success if methods can be designed and implemented that result in retrieval performance that is equal to or better than retrieval of citations based on humanly assigned index terms.

A project of this scope necessarily involves the efforts of many people. Project team members are from several NLM divisions, including the LHNCBC, LO, and NCBI.

The project will assume the availability of free text in the form of titles and abstracts but will also consider the increasing availability of the full text of journal articles in electronic form. The project will investigate concept-based indexing methods that go well beyond automatic word-based indexing (such as the inverted word index already part of MEDLINE). As insights are gained throughout the project, current operational processes or systems may be iteratively modified and improved in keeping with those insights.

The following list of presentations and papers provide a good overiew of the Indexing Initiative. Latest publications are first in the list.

Indexing Initiative Related Publications
(1.1mb)	User-centered Evaluation of the MTI System, 2007
(39kb)	Fine-Grained Indexing of the Biomedical Literature: MeSH Subheading Attachment for a MEDLINE Indexing Tool, AMIA 2007
(103kb)	Multiple Approaches to Fine-Grained Indexing of the Biomedical Literature, Proc Pacific Symposium on Biocomputing 2007
(44kb)	Automatic Indexing of Specialized Documents: Using Generic vs. Domain-Specific Document Representations, BioNLP 2007
(158kb)	From Indexing the Biomedical Literature to Coding Clinical Text: Experience with MTI and Machine Learning Approaches, BioNLP 2007
(100kb)	Semi-Automatic Indexing of Full Text Biomedical Articles, AMIA 2005
(50kb)	Evaluation of French and English MeSH Indexing Systems with a Parallel Corpus, AMIA 2005
(54kb)	The NLM Indexing Initiative's Medical Text Indexer, MedInfo 2004
(319kb)	Application of a Medical Text Indexer to an Online Dermatology Atlas, MedInfo 2004
(510kb)	A MEDLINE Indexing Experiment Using Terms Suggested by MTI, June 2002
(2.1mb)	Automated and Semi-automated Indexing, Report to the Board of Regents 2002
(130kb)	Automatic MeSH Term Assignment and Quality Assessment, AMIA 2001
(139kb)	The NLM Indexing Initiative, 2000
(203kb)	1999 Report to the Board of Scientific Counselors
(203kb)	1999 AMIA Poster Presentation: Automated Assignment of Medical Subject Headings (HTML)
(503kb)	Medical Text Indexer (MTI) Processing Flow