Effective Implementation of Accurate Human Iris Eye Recognition System using Polynomial Regression in comparison with Naive Bayes algorithm for improved Accuracy

Abstract

The Effective implementation of an accurate human iris eye recognition system in the field of cyber crime using Polynomial Regression in the comparison with Naive Bayes algorithm for improved accuracy. It is used to identify people easily. Two groups of algorithms are proposed. Naive Bayes algorithm is compared with polynomial regression. Both algorithms work on human eye recognition for accuracy. Accuracy is analyzed for Human iris Eye recognition. Naive Bayes is a processing technique based on Bayes theorem. Polynomial Regression calculates the result values based on input features from the data in the work. The algorithm uses the properties of training data to create a model, which it then uses to estimate the value of new data. The Naive Bayes sample size (N=23) and Polynomial Regression sample size (N=23) algorithms are used to recognise iris. The significance value of the data set was predicted using SPSS with an G-power value 80%. Naive Bayes accuracy is 94.366% which is comparatively higher than Polynomial Regression with accuracy of 92.364%. There is a significant difference among two groups with a significance value of 0.048(p<0.05). aive Bayes algorithm performs better than the Polynomial Regression accuracy of human eye recognition in the field of cyber crime.