부산대학교 도서관

The purpose of the research is to compare Random Forest (RF) Algorithm methods to Extra Trees (ET) Classification algorithms for detecting the Disk Filtration attacks in air gapped computers. Materials and Methods: The sample set comprises 609 positive records and 612 negative records were taken for conducting the experiment to compare Random Forest Algorithm with Extra Trees Algorithm for Disk Filtration Attacks. This sample size was obtained using clinical analysis, which had alpha and beta values of 0.05 and 1, an enrollment ratio of 1, pre-test G power of 80%, and 0.5, respectively, with a 95% confidence level. Random Forest and ET classifiers were built using a framework for disk filtration detection. Results: The Disk Filtration attacks on the data set is detected with 99.40% accuracy by the Novel Random Forest classifier,the ET generates 99.20%, in contrast. With a 95% confidence interval, there is a statistically significant difference between the two groups (p=0.002; p0.05). Therefore, RF classifiers outperform ET classifiers. Conclusion: The outcomes demonstrate that RF performs more accurately when compared to ET.