Return to search

Machine Learning for Botnet Detection: An Optimized Feature Selection Approach

Yes / Technological advancements have been evolving for so long, particularly
Internet of Things (IoT) technology that has seen an increase
in the number of connected devices surpass non IoT connections.
It has unlocked a lot of potential across different organisational
settings from healthcare, transportation, smart cities etc. Unfortunately,
these advancements also mean that cybercriminals are
constantly seeking new ways of exploiting vulnerabilities for malicious
and illegal activities. IoT is a technology that presents a
golden opportunity for botnet attacks that take advantage of a
large number of IoT devices and use them to launch more powerful
and sophisticated attacks such as Distributed Denial of Service
(DDoS) attacks. This calls for more research geared towards the detection
and mitigation of botnet attacks in IoT systems. This paper
proposes a feature selection approach that identifies and removes
less influential features as part of botnet attack detection method.
The feature selection is based on the frequency of occurrence of the
value counts in each of the features with respect to total instances.
The effectiveness of the proposed approach is tested and evaluated
on a standard IoT dataset. The results reveal that the proposed
feature selection approach has improved the performance of the
botnet attack detection method, in terms of True Positive Rate (TPR)
and False Positive Rate (FPR). The proposed methodology provides
100% TPR, 0% FPR and 99.9976% F-score.

Identiferoai:union.ndltd.org:BRADFORD/oai:bradscholars.brad.ac.uk:10454/18862
Date05 April 2022
CreatorsLefoane, Moemedi, Ghafir, Ibrahim, Kabir, Sohag, Awan, Irfan U.
Source SetsBradford Scholars
LanguageEnglish
Detected LanguageEnglish
TypeConference paper, Accepted manuscript
Rights© 2021 Association for Computing Machinery. Reproduced in accordance with the publisher's self-archiving policy. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from permissions@acm.org., Unspecified

Page generated in 0.0018 seconds