ISSN 2394-5125
 


    MACHINE LEARNING TECHNIQUES FOR CYBER ATTACKS DETECTIONS (2023)


    L.C.Usha Maheswari, P.Sruthi, Ch.Hamsini, G.Keerthika, M.Sreeja
    JCR. 2023: 206-211

    Abstract

    Improvements in PCs and correspondence have brought about broad changes in comparison to the past. The use of new technologies gives individuals, companies and governments unbelievable benefits, even when they are messing against them. For eg, security of major data, safety of data stadiums, accessibility of information etc. Digital fear-based oppression is, depending on these questions, one of the biggest problems of our day. Digital fear, which has brought many problems to citizens and organizations, has come to an extent that could threaten the openness and protection of the nation through numerous events, such as criminal groups, professional individuals and digital activists. In this sense, IDS systems were built to maintain a strategic distance from digital attacks. Intrusion detection systems (IDS) At the moment, learning the computation of the SVM (Bolster Support Vector Machine) was used to identify port sweep efforts that depend on a new data set of CICIDS 2017, which achieved 69.79% individually. We should implement other algorithms, including CNN, ANN and Random Forest, rather than SVM, so that accuracy such as SVM � 93.29, CNN - 63.52, Random Forest � 99.93, ANN � 99.11 can be acquired.

    Description

    » PDF

    Volume & Issue

    Volume 10 Issue-7

    Keywords