Short - Text Mining Approach for Medical Domain

The growing amount of published articles in medicine represents a massive source of knowledge, which can only efficiently be accessed by a new creation of automated information extraction tools. By discovering predictive relationships between different pieces of extracted data from the medicine field, data-mining algorithms can be used to improve the precision of information extraction. Medical text retrieval refers to text retrieval techniques applied to biomedical resources and articles available in the biomedical domain. In this paper we propose an approach to extract the important medical text from the published medical articles. The medical text here means the disease and the treatment keywords. Here we extract the relationship between the disease and the treatment. The three relations that we extract are cure, prevention and side-effect. The amount of published medical articles, and therefore the underlying medical knowledge base, is expanding at an increasing rate. So retrieval of the reliable information is a difficult task. To overcome that we propose this method and also we develop an expert system to develop medical diagnosis applications and the medication management which serves a good purpose for the laypeople as well as for the people who do research in the medical field. This paper presents a technique for using data mining algorithm to increase the accuracy of medical text extraction.