Psoriatic Disease Classification Based on Exploratory Data Analysis

Abstract

Psoriasis is caused because of the problem in immune system that that causes a rash with itchy, scaly patches, commonly occur on the knees, elbows, trunk and scalp. This study aims to diagnose Psoriasis and its type through clinical data. A framework has been developed to carry out this task in three phases. Initially preprocessing operation had been performed followed by Exploratory Data Analysis (EDA) is used to gain knowledge about the dataset by using statistical graphics and data visualization mechanism. Here, promising features are selected using various discriminatory techniques Conjoint Analysis, Correlation analysis, Interdependence of pairs, Discriminant Analysis, Analysis of Variance. Best discriminant features were given input to Adaboost classifier to classify into six psarosisi types. Finally, performance analysis had been made and compared with recent works with the accuracy of 98%