This chapter presents a best classification of diabetes. The proposed approach work consists in two stages. In the first stage the Pima Indian diabetes dataset is obtained from the UCI repository of machine learning databases. In the second stage, the authors have performed the classification technique by using fuzzy decision tree on Pima Indian diabetes dataset. Then they applied PSO_SVM as a feature selection technique followed by the classification technique by using fuzzy decision tree on Pima Indian diabetes dataset. In this chapter, the optimization of SVM using PSO reduces the number of attributes, and hence, applying fuzzy decision tree improves the accuracy of detecting diabetes. The hybrid combinatorial method of feature selection and classification needs to be done so that the system applied is used for the classification of diabetes.
TopIntroduction
In the current scenario, there are certain factor such as an environmental factor, sedentary life style, genetically (hereditary) producing several diseases most popularly known as ‘diabetes’, which has become the most leading chronic diseases. So the primary need of this generation is to become the state of art healthcare. The main motto of this work is to provide a indigenous efficient diagnostic tool of detection of diabetes, even though there are already several established existing technique, which have been used for the diagnosis of diabetes.
Diabetes is a disease which increases the blood glucose known as hyperglycemia to a level which affects the body to a greater extent (http://www.diabetes.org/diabetes-basics).
The researchers provides a new and efficient technique for the medical diagnosis by integrating the feature selection using Support Vector Machine (SVM)(Brank et al., 2002) optimized with the Particle Swarm Optimization (PSO)(Mandal and N, 2012) and rule generation using Fuzzy based Decision Tree (FDT) hence the approach is the selection of features so that classification is done for the diabetes mellitus patients (Dash and Liu, 1997).
Feature Selection is a technique of identifying the selecting the most relevant features as possible (Liu and Sentino, 1998). By selecting the most relevant features from the dataset and removing the irrelevant features the high dimensionality data size can be reduces and allows the process to apply efficiently and quickly (Abraham et al., 2007). Particularly in the classification of diabetes dataset the number of features used increases the accuracy of predicting diabetes patients.
A Decision Tree (DT) is a recursive form of the tree used for the classification of uncertain data or huge data. A DT consists of root nodes and leaves which shows the dataset dependent attributes and their respective values. In Data mining there are several decision tree such as using ID3 (Iterative Dichotomiser), CART (Classification & Regression Tree), Random Forest, J48graft DT (J48 graft Decision Tree), etc.