Application of table based similarity to classification of bio-medical documents

This research is concerned with applying the table based approach to text categorization in the bio-medical domain. The popularity and interests in the bio-medical domain are increasing recently with respect to the text processing. In this research, we encode texts in the domain into tables, define the similarity measure between tables representing the texts, and apply it to categorizations of the bio-medical texts. We may expect better performance and stability of the proposed approach as well as in other domains. Thus, we will use the collection of medical documents, called Oshumed, as the test data for evaluating the proposed approach.