Extracting a Linguistic Summary from a Medical Database

In general, medical clustering concerns a big database. The present paper aims at extracting a fuzzy linguistic summary from a large medical database. A linguistic summary is used to reduce large volumes of data to simple sentences. It is worth noting that with the increase of the amount of medical data, different techniques of machine learning have been developed recently. In this article, an attempt is made to build a medical linguistic summary template. Our linguistic summary model is based on the calculated fuzzy cardinality. It deals with semantic queries in natural language. Our proposal is to develop a classification system based on the linguistic summary of two medical databases in which the calculation of similarity between different sets of linguistic summaries is used; the patient’s class is then identified by calculating the Sugeno integral. The present study was successful in developing a classification system that is based on the linguistic summary of two datasets from the UCI Machine Learning Repository, i.e. Pima Indians Diabetes dataset and Wisconsin Diagnostic Breast Cancer (WDBC) dataset. The results obtained were then employed for a benchmark test.