Accurate Human Palm Recognition System in Cybercrime Analysis using Naive Bayes in comparison with Polynomial Regression

Aim: The proposed model helps in recognizing accurately human palm in cybercrime analysis using Naive Bayes (NB) and Polynomial Regression (PR) and palm recognition is used to identify the person easily. Materials and Methods: Algorithm which is proposed here is Naive Bayes as well as the compared one is Polynomial Regression. Both the algorithms work on human palm recognition for accuracy. Accuracy is analyzed for human palm recognition. Naive Bayes is an act of processing technique based on Bayes' theorem. The framework receives the data where the computation of result is done based on the features used as input. Both training as well as data for testing are employed for generation of results. Palm recognition process gets carried over using Naive Bayes whose size of sample is (N=23) as well as Polynomial Regression whose size of sample is (N=23). G power computation is done and it attains 80% of value using SPSS. Result: Naive Bayes (NB) accuracy is 93.136% whereas Polynomial Regression (PR) has accuracy as 92.200%. Significant difference exists between 2 groups as their significance value becomes 0.225 (p>0.05). Conclusion: Naive Bayes (NB) performs better than the Polynomial Regression (PR) in accuracy of human palm recognition in cybercrime analysis.