Building a Core Biomedical Knowledge Base

Considering today's surge of information, the need for well organized knowledge bases is increasing rapidly for providing simplified access to knowledge and its further processing. In biomedical domain, heaps of information is buried in scientific publications and online forums. This calls for representing this information in a more expressive semantic way by determining and storing relational information into a machine readable form. So, the primary goal of this research endeavor has been to build a knowledge base on entities and relations containing amass of formalized background knowledge suitable for supporting reasoning in biomedical domain. We introduce a way for easily accessing the knowledge about body parts and symptoms of human diseases, along with environmental, social, nutritional and diagnostic factors that cause these diseases. The information for this knowledge base is extracted from the controlled vocabulary thesaurus "Medical Subject Headings" (MeSH), which is published by National Library of Medicine. In this paper, we will discuss the process that we adapted for extracting and assembling the information from MeSH according to our classification needs, along with a description of our knowledge base schema and a visual web interface for it, and then we will indicate the different problem areas in which our knowledge base (KB) can be useful.