抄録
Recently, public interest in health has grown, and increasing numbers of people are relying on a regular medical checkup for the early detection of disease. At the same time, the rapid development of information technology gives us much medical information, and the demands for an effective medical support system will continue to grow. The aim of this study is to develop a Hospital Search Support System (HSSS) as a part of integrated medical support system which includes electronic medical chart sharing system, specialist introduction system, medicine information system and so on. It employs Hierarchical Keyword Graph (HK Graph) for the Disease Search Support System (DSSS). DSSS can list diseases according to the strength of the relation to the symptoms inputted by a user. First, this paper shows the extraction method of symptom words for DSSS. And then this paper also shows the method to present the candidates of diseases with their rank based on the symptoms on HK Graph and the symptom severity inputted by user. This paper investigates the effectiveness of the proposed method based on the accuracy and coverage rate of symptoms and the transition of ranking of diseases with the change of symptom severity.
