Detection of Malware in Cloud Storage Data using Naive Bayes Algorithm Comparing K-Nearest Neighbors Algorithm to Reduce False Detection

Abstract

Aim: To enhance the accuracy in detection of Novel Cloud malware in cloud storage data Using  K-Nearest Neighbors Algorithm comparing Naive Bayes Algorithm to reduce false detection. Materials and Methods: This research work we are considering two groups,one group is K-Nearest Neighbors Algorithm (KNN) comparing group 2 Naive Bayes Algorithm (NB).  Each group consists of a sample size of 30. Their accuracies are compared with each other using different sample sizes also. Results: By running algorithms for various iterations the following results are obtained. SPSS was used to calculate the sample size. The pre-test analysis was maintained at 80%. G-power is used to calculate sample size.  K-Nearest Neighbors Algorithm is 99.4% more accurate than the Naive Bayes Algorithm of 62.8% in detection of malware in cloud storage data which reduces the false detection rate (p=0.001). Conclusion: Through this, we are able to prove that the prediction novel cloud Malware Analysis done using K-Nearest Neighbors (KNN) model is significantly better than the Naive Bayes in identifying Malware detection in cloud storage data. It can be also considered as a better option for the classification of malware detection.