Automatic Categorization of Google Search Results for Medical Queries using JDI

The web has become the primary source of medical information for consumers and health professionals. It is quite common for people to “Google” for information related to a medical topic. But the problem remains that as the number of documents increases on the web, the difficulty in quickly locating the best documents increases. Classifying results into meaningful categories, helps guide users to the most relevant set of results. Journal Descriptor Indexing (JDI) is a novel approach to fully automatic indexing. In this paper we explore the feasibility of using JDI to organize Google search results for medical queries into meaningful categories. For our experiments, we used JDI in combination with a set of heuristics to automatically categorize the search results for 5 query terms. Three independent reviewers reviewed and evaluated the automatic categorization for 3 documents for each query term. The results clearly suggest that this method offers promise. Additional work for improving the categorization as well as to determining whether a term is medical or not is also discussed.