event-icon
Description

Current information retrieval systems return search results as ranked lists. This presentation is not effective or efficient given the vast digital collections of today. A potential approach toward improved concept and document retrieval is to develop a cluster-based model that leverages unsupervised machine learning and visual analytics. As the first step toward this direction, this poster investigates the fundamental problems regarding the quality of document clusters and develops a systematic evaluation methodology.

Learning Objective: To learn about novel information systems for biomedical information retrieval that leverage unsupervised machine learning and evaluation of such systems.

Authors:

Michael Ortiz (Presenter)
University of North Carolina at Chapel Hill

Kazuhiro Seki, Konan University
Javed Mostafa, University of North Carolina at Chapel Hill

Presentation Materials:

Tags