MALWARE CATEGORY PREDICTION USING KNN AND SVM CLASSIFIERS
Journal: International Journal of Mechanical Engineering and Technology(IJMET) (Vol.10, No. 2)Publication Date: 2019-02-27
Authors : UDAYAKUMAR N AYUSH MISHRA SHIVANG MISHRA; SUBBULAKSHMI.T;
Page : 787-797
Keywords : Malware; Malware prediction; K-Nearest Neighbors; Support Vector Machines.;
Abstract
The emergence of the vulnerability databases around the world are serving the purpose of a double edged sword. The malware researchers, industry members and end users are aware of them to initiate better prevention strategies. The dark world hackers are using them to lure into systems through the points mentioned in the vulnerability databases. Hence, it is highly necessary to predict the malware at the early stage to avoid further loss. The objective of this research work is to predict the malware using the classifiers Logistic Regression, K–Nearest Neighbors (KNN) and Support Vector Machines (SVM). We found that the appropriate use of these classifiers have resulted great improvement in prediction accuracy. Feature selection is also done to further improve the accuracy to 99% with polynomial kernel function.
Other Latest Articles
- RESEARCH ABOUT PHASE TRANSFORMATION OF LOW CARBON STEEL WHEN DEFORMATION AND HEAT TREATMENT
- IMPLEMENTATION OF PORTABLE SOLAR PLANTS SUPPLY (DC -WATER PUMP) AND EMERGENCY UNIT POWER SUPPLY
- ІНТЕГРАЦІЯ НАЦІОНАЛЬНИХ ПІДПРИЄМСТВ У МІЖНАРОДНІ ВИРОБНИЧІ МЕРЕЖІ
- ПОДАТОК НА ПРИБУТОК У СТИМУЛЮВАННІ ІНВЕСТИЦІЙНОЇ АКТИВНОСТІ ПІДПРИЄМСТВ
- ОБЛІК ОБ’ЄКТІВ ПРАВА ІНТЕЛЕКТУАЛЬНОЇ ВЛАСНОСТІ
Last modified: 2019-05-27 21:53:45