Adaption of RWI DT D algorithm medical text documents clustering to enhance the performance

Medical domain specific words occur less in documents as rare words. As per survey, the role of rare words is important to enhance the quality of classification performance. Rare word identification algorithm using dynamic threshold determination and dictionary based matching RWI DT D is used for medical document classification to enhance the accuracy of classifications. Same way, the text clustering process also required proper pre-processes medical text data for effective clustering. Traditional approaches manually define a threshold value to remove all rare words even though this domain specific rare words are important. Therefore, in this paper, authors adapted an RWI DT D dynamic threshold determination algorithm with medical domain word dictionary-based selection for medical documents clustering. Results indicate that this RWI DT D method is very effective compared to traditional approaches for all sized medical text documents.