Biomedical Literature Mining and Knowledge Discovery

Mathew J. Palakal

Department of Computer and Information Science

Indiana University Purdue University Indianapolis, Indiana, USA

Summary

The biological literature databases continue to grow rapidly with
vital information that is important for conducting sound biomedical
research. BioMap is an attempt to create a scalable knowledgebase of
biological relationships from vast amount of literature data. BioMap
will be a new type of "secondary" knowledge resource derived from
primary resources such as MEDLINE. It will be the "window" to biomedical
researchers who will be seeking knowledge from the literature databases,
however, without being overwhelmed by its large volume. The complete
prototype development of BioMap addresses several innovative research
issues related to knowledge discovery from literature documents and
real-time, interactive access of this knowledge. Specific problems
that are being investigated are: development of a multi-level approach
comprising of statistical means, stochastic models and dictionaries
for the identification of multiple biological objects and discovering
their relationships; creation of a hypergraph to represent relationship
knowledge for a given user; generating pathways of new hypothesis using
the hypergraph based on graph algorithms; and, an online validation
mechanism to validate automatically discovered relationships. Another
significant contribution of this work is the ability of the system
to efficiently discover associations not explicitly reported in any
one article, but based on associations implicitly hidden in multiple
documents. The overall system organization, current status and preliminary
results from various components of the system will be presented during
this talk.

Dr. Mathew Palakal is a Professor and Chair of the Department of Computer
and Information Science and Director of Informatics Research Institute at
Indiana University Purdue University Indianapolis. His primary research
interests are in pattern analysis and machine intelligence. He is working
on problems related to information management using information filtering
and text mining approaches and structural health monitoring and smart
diagnostics based on intelligent computational methods. One of his
current primary areas of interest is in intelligent systems applied to
bioinformatics area.