Interactive Curriculum Search Engine using Natural Language Processing

June 6, 2017 8:00 AM – 9:15 AM

Toufeeq Ahmed, Vanderbilt University
Tao Le, ScholarRx

Medical students and faculty are often seeking curriculum content at a granular level and are looking for smarter tools to locate a specific document or session. We have built a powerful interactive search engine that uses natural language processing (NLP) to efficiently find specific content in a few clicks using faceted search techniques. The curriculum search engine leverages our NLP engine (QuickMatch) to index medical concepts from course sessions and curriculum documents (eg, PPTs, PDFs). The system first extracts the full text from documents and then parses them using the NLP engine. The documents and associated sessions are searchable using an interactive UI that allows the user to easily browse through large results sets with adaptive guidance provided by dynamic filtering choices.

In this presentation, we will discuss users’ information seeking needs and how it can be used by faculty, deans and curriculum managers to track curriculum concepts efficiently across the entire curriculum.