Open AccessEvolutionary Multivariate Kernal Svm Prediction Method for Classification
Author(s) -
K. Geetha*
Publication year - 2020
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.d1923.069820
Subject(s) - support vector machine , multivariate statistics , population , rank (graph theory) , artificial intelligence , thyroid , set (abstract data type) , computer science , data mining , pattern recognition (psychology) , mathematics , machine learning , medicine , environmental health , combinatorics , programming language
Thyroid disorders are common among the world wide population. This disorders posses’ significant problems among Indians. Research studies shows that nearly 32% of Indian population suffers from various thyroid disorders. This paper deals with the thyroid data set which in turn classify into three groups as hyper thyroidisim, hypothyroidism and normal. The American Thyroid Association reported twelve percent of their citizens suffer from thyroidism in which 60% population are unaware of their condtions.. Above statistics implies the classification of thyroid disorder is crucial in global perspective too. The thyroid data set are collected from UCI repository and it is multivariate type with 21 attributes. With the 21 attributes only 10 attributes are selected based on their rank. Hybrid Differential Evolution Kernel Based SVM algorithm is used to classify the data set. It takes around 30 epochs to stabilize the errors. The classification accurancy is observed to be 67.97%.