Comparative Analysis of K-Nearest Neighbours Algorithm and Naive Bayes Algorithm for Prediction of Storm Warning
Main Article Content
Article Sidebar
Abstract
The primary aim of this research was to enhance the accuracy of storm warnings by employing the novel KNearest Neighbours algorithm and comparing it to the Naive Bayes method. This investigation divided
participants into two groups: the Novel K-Nearest Neighbours and the Naive Bayes Algorithm, each
comprising ten representatives. The mean accuracy was determined using the ClinCalc software tool in a
supervised learning setting, considering an alpha value of 0.05, a G-Power of 0.8, and a 95% confidence
interval. The K-Nearest Neighbours algorithm showcased a notable accuracy rate of 68.20%, outstripping the
57.31% accuracy of the Naive Bayes. The difference between the two was statistically significant (p=0.000).
In conclusion, the K-Nearest Neighbours approach substantially surpassed the Naive Bayes.
How to Cite
Article Details
Novel K-Nearest Neighbours, Machine Learning, Naive Bayes Algorithm