Case Study: Streamlining Content Indexing for the American College of Physicians (ACP) with Access Innovations

Client Information:

  • Organization: American College of Physicians (ACP)
  • Address: 190 N Independence Mall W, Philadelphia, Pennsylvania, 19106
  • Contact Person: Christina Rudyj, Director of Electronic Publishing and Product Development
  • Email:
  • Direct Phone: +1.215.351.2827
  • Engagement Period: December 27, 2019, to present

Project Overview:

Every physician has a copy of the ACP Annuals in their office.  Access Innovations collaborated with the American College of Physicians (ACP) to develop a robust modification of medical topics as a taxonomy and knowledge domain (rules base) aimed at efficiently indexing their widely respected repository of content, comprising approximately 56,000 articles in their backfile.

Phase 1: Building Taxonomy and Indexing Back Files:

During the initial phase, Access Innovations worked on indexing the back files of the Annals site, covering journals, videos, and podcasts. The process involved several steps:

  • ACP provided the entire 56,000 record backfile in an XML spreadsheet.
  • Access Innovations merged ACP headings and applied word terms, analyzing ¬†terms to add to the base vocabulary.
  • Additional terms were harvested by processing the ACP back file against knowledge domains developed in-house, including those based on terminology lists from MeSH, PLOS, and other sources.
  • An n-grams program was utilized to identify relevant word terms within the data.
  • Hierarchy was established, determining Top Terms, Broader Terms, Narrower Terms, Preferred and Non-Preferred Terms (synonyms).
  • Preferred terms were identified based on their usage within the ACP content.
  • Consultation with ACP Subject Matter Experts (SMEs) was conducted regarding the appropriateness of the ACP taxonomy.
  • ACP staff and SME feedback were incorporated into the ACP taxonomy.
  • A knowledge base with simple rules was automatically created.
  • The ACP backfile was processed through the created knowledge base.
  • The quality of indexing was analyzed, and the AI layer of the Data Harmony MAIstro‚ĄĘ app was fine-tuned accordingly.
  • The full backfile of 56,000 records was then reprocessed and loaded to the ACP Journals platform.

Phase 2: Ongoing Indexing:

Currently in Phase 2, Access Innovations is indexing approximately 29 articles from the journal of the Annals of Internal Medicine and approximately 37 articles from the Annals: Case Studies section of the Annals site monthly, totaling around 66 content objects per month.


  • ACP benefited from a streamlined content indexing process, enabling efficient retrieval and management of their extensive content repository.
  • ACP uses the taxonomy in their manuscript submission system to tag the incoming data for faster identification and distribution to the peer reviewers who are also tagged with the taxonomy to create a semantic profile of each person and article in the publication workflow.
  • Enhanced searchability and accessibility of ACP content were achieved, facilitating improved user experience and knowledge dissemination.
  • Continuous collaboration with ACP SMEs ensured the relevance and accuracy of the indexing system.
  • Access Innovations’ expertise in taxonomy development and knowledge domain creation continues to support the success of the project, empowering ACP to stay at the forefront of information management in their field.


Through collaborative efforts and meticulous implementation, ACP successfully enhanced their ¬†content indexing capabilities, enabling efficient retrieval and management of their extensive content repository using Access Innovations and its tools. The project not only addressed immediate indexing needs but also laid the groundwork for sustained efficiency and productivity in content management. Access Innovations’ ongoing support ensures that ACP remains well-equipped to meet the evolving demands of their audience and industry.