The document presents a machine learning model for diagnosing erythemato-squamous diseases using an innovative approach that combines Catfish Binary Particle Swarm Optimization (CatfishBPSO) and Kernelized Support Vector Machines (K-SVM) with Association Rules for feature selection. The proposed model demonstrates a high classification accuracy of 99.09% with 24 features from a dataset sourced from the University of California, Irvine. The research highlights the effectiveness of the AR-CatfishBPSO-KSVM model compared to traditional methods in accurately diagnosing these challenging dermatological conditions.