Building a Prediction Model for Diagnose Dermatology Diseases using Classification Technique

Abstract

Differential diagnosis means the process of differentiating between two or more diseases that share similar signs or symptoms, the differentiating in such cases consider as a challenge in the field of dermatology, a real Sudanese data was built for the purposes of this research which had been collected from the medical reports in Omdurman Military Hospital department of dermatology. Three models were built using three classification algorithms Naïve Bayes, j48 and IBK .The research aimed to build a model classifying four dermatology diseases which have high similarity in their symptoms, these diseases are: 1- Psoriasis 2 - seboreic dermatitis 3. lichen planus 4- cronic dermatitis The classification models had an accuracy in the range of%90.6 to %99.4 ,the results showed that IBK algorithm gave the highest accuracy ( %99.4 ) and less time to construct the model.