A fuzzy information retrieval for respiratory diseases

Abstract

Respiratory diseases can be found in many countries, particular in a tropical climate country such as Thailand. Generally, the respiratory disease associated with breathing system which is several organs involved such as nose, throat, larynx, bronchial lung etc. The symptoms are consistent with organs; moreover, the nature of the symptoms is very similar. Therefore, inexperienced peoples are difficult to diagnose the symptoms correctly. In this paper, we present an information retrieval system based on the concept of fuzzy logic to be considered vagueness information and uncertainly on text document. The fuzzy logic model applied in a simple way to comprehensive query form to help input uncertainly of users. The diseased content represented as a vector space model; furthermore, cosine similarity technique is used to rank the retrieved results. To experiment, we provide an IR system that work on mentioned model. Moreover, we get query from three type's users that are general peoples, patients and doctors. The results show the average precision around 67 percent, recall 70 percent and F-measure about 70 percent.