A. Swaminathan, Balamurali Ramakrishnan, K. M, S. R
{"title":"Prediction of Cyber-attacks and Criminality Using Machine Learning Algorithms","authors":"A. Swaminathan, Balamurali Ramakrishnan, K. M, S. R","doi":"10.1109/3ICT56508.2022.9990652","DOIUrl":null,"url":null,"abstract":"Cyber-attacks are quickly becoming one of the world's most serious problems. This crisis can be avoided by using real-time data to identify an attack and its perpetrator. The information can be obtained from the implementations of individuals who were subjected to cyber-attacks in forensic units. The information includes the criminal activity, the perpetrator's gender, impairment, and attack methods. This work use supervise Machine Learning (ML) methods to investigate cybercrime in four distinct models and predict the effects of the defined traits just on identification of the threat technique and the perpetrator. In our system, utilized three machine learning methods and predicted that their precision ratios would be close. In this paper, investigate digital misdoings in three distinct models using ML techniques, and forecast the impact of the characterized attributes on the spot of the electronic assault tactic and the perpetrator. In this investigation, will use three ML algorithms, Logistic Regression, Random Forest, and K-Nearest Neighbor (KNN) and compare their efficacy in two different models before concluding with the model that has the best survivability for every type of information index. Machine learning allows cyber security systems to assess and learn from patterns in order to detect and prevent terrorist acts and adapting to different behavior. It can assist cyber security teams in being more proactive in terms of preventing threats as well as reacting to malicious activities in real time.","PeriodicalId":361876,"journal":{"name":"2022 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 International Conference on Innovation and Intelligence for Informatics, Computing, and Technologies (3ICT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/3ICT56508.2022.9990652","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Cyber-attacks are quickly becoming one of the world's most serious problems. This crisis can be avoided by using real-time data to identify an attack and its perpetrator. The information can be obtained from the implementations of individuals who were subjected to cyber-attacks in forensic units. The information includes the criminal activity, the perpetrator's gender, impairment, and attack methods. This work use supervise Machine Learning (ML) methods to investigate cybercrime in four distinct models and predict the effects of the defined traits just on identification of the threat technique and the perpetrator. In our system, utilized three machine learning methods and predicted that their precision ratios would be close. In this paper, investigate digital misdoings in three distinct models using ML techniques, and forecast the impact of the characterized attributes on the spot of the electronic assault tactic and the perpetrator. In this investigation, will use three ML algorithms, Logistic Regression, Random Forest, and K-Nearest Neighbor (KNN) and compare their efficacy in two different models before concluding with the model that has the best survivability for every type of information index. Machine learning allows cyber security systems to assess and learn from patterns in order to detect and prevent terrorist acts and adapting to different behavior. It can assist cyber security teams in being more proactive in terms of preventing threats as well as reacting to malicious activities in real time.