Anael Bonneton, D. Migault, S. Sénécal, Nizar Kheir
{"title":"基于时间序列决策树的DGA机器人检测","authors":"Anael Bonneton, D. Migault, S. Sénécal, Nizar Kheir","doi":"10.1109/BADGERS.2015.016","DOIUrl":null,"url":null,"abstract":"This paper introduces a behavioral model for botnet detection that leverages the Domain Name System (DNS) traffic in large Internet Service Provider (ISP) networks. More particularly, we are interested in botnets that locate and connect to their command and control servers thanks to Domain Generation Algorithms (DGAs). We demonstrate that the DNS traffic generated by hosts belonging to a DGA botnet exhibits discriminative temporal patterns. We show how to build decision tree classifiers to recognize these patterns in very little computation time. The main contribution of this paper is to consider whole time series to represent the temporal behavior of hosts instead of aggregated values computed from the time series. Our experiments are carried out on real world DNS traffic collected from a large ISP.","PeriodicalId":150208,"journal":{"name":"2015 4th International Workshop on Building Analysis Datasets and Gathering Experience Returns for Security (BADGERS)","volume":"89 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"DGA Bot Detection with Time Series Decision Trees\",\"authors\":\"Anael Bonneton, D. Migault, S. Sénécal, Nizar Kheir\",\"doi\":\"10.1109/BADGERS.2015.016\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper introduces a behavioral model for botnet detection that leverages the Domain Name System (DNS) traffic in large Internet Service Provider (ISP) networks. More particularly, we are interested in botnets that locate and connect to their command and control servers thanks to Domain Generation Algorithms (DGAs). We demonstrate that the DNS traffic generated by hosts belonging to a DGA botnet exhibits discriminative temporal patterns. We show how to build decision tree classifiers to recognize these patterns in very little computation time. The main contribution of this paper is to consider whole time series to represent the temporal behavior of hosts instead of aggregated values computed from the time series. Our experiments are carried out on real world DNS traffic collected from a large ISP.\",\"PeriodicalId\":150208,\"journal\":{\"name\":\"2015 4th International Workshop on Building Analysis Datasets and Gathering Experience Returns for Security (BADGERS)\",\"volume\":\"89 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2015-11-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2015 4th International Workshop on Building Analysis Datasets and Gathering Experience Returns for Security (BADGERS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/BADGERS.2015.016\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 4th International Workshop on Building Analysis Datasets and Gathering Experience Returns for Security (BADGERS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/BADGERS.2015.016","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
This paper introduces a behavioral model for botnet detection that leverages the Domain Name System (DNS) traffic in large Internet Service Provider (ISP) networks. More particularly, we are interested in botnets that locate and connect to their command and control servers thanks to Domain Generation Algorithms (DGAs). We demonstrate that the DNS traffic generated by hosts belonging to a DGA botnet exhibits discriminative temporal patterns. We show how to build decision tree classifiers to recognize these patterns in very little computation time. The main contribution of this paper is to consider whole time series to represent the temporal behavior of hosts instead of aggregated values computed from the time series. Our experiments are carried out on real world DNS traffic collected from a large ISP.