Parallel Outlier Detection for Streamed Data Using Non-Parameterized Approach

Int. J. Synth. Emot. Pub Date : 2017-07-01 DOI:10.4018/IJSE.2017070102

H. D. Markad, S. Sangve

{"title":"Parallel Outlier Detection for Streamed Data Using Non-Parameterized Approach","authors":"H. D. Markad, S. Sangve","doi":"10.4018/IJSE.2017070102","DOIUrl":null,"url":null,"abstract":"Outlierdetectionisusedinvariousapplicationslikedetectionoffraud,networkanalysis,monitoring trafficovernetworks,manufacturingandenvironmentalsoftware.Thedatastreamswhicharegenerated arecontinuousandchangingovertime.Thisisthereasonwhyitbecomesnearlydifficulttodetect theoutliersintheexistingdatawhichishugeandcontinuousinnature.Thestreameddataisreal timeandchangesovertimeandhenceitisimpracticaltostoredatainthedataspaceandanalyze itforabnormalbehavior.Thelimitationsindataspacehasledtotheproblemofrealtimeanalysis ofdataandprocessingit inFCFSbasis.Theresultsregardingtheabnormalbehaviorhavetobe doneveryquicklyandinalimitedtimeframeandonaninfinitesetofdatastreamscomingoverthe networks.Toaddresstheproblemofdetectingoutliersonareal-timebasisisachallengingtaskand hencehastobemonitoredwiththehelpoftheprocessingpowerusedtodesignthegraphicsofany processingunit.Thealgorithmusedinthispaperusesakernelfunctiontoaccomplishthetask.It producestimelyoutcomeonhighspeedmulti-dimensionaldata.Thismethodincreasesthespeed ofoutlierdetectionby20timesandthespeedgoesonincreasingwiththeincreasewiththenumber ofdataattributesandinputdatarate. KEywORDS Anomaly Intrusion Detection, Compute Unified Device Architecture (CUDA), Gaussian Detection Scheme, Graphics Processing Unit (GPU), Outlier Detection, Parallel Execution","PeriodicalId":272943,"journal":{"name":"Int. J. Synth. Emot.","volume":"40 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"4","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Int. J. Synth. Emot.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4018/IJSE.2017070102","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 4

Abstract

Outlierdetectionisusedinvariousapplicationslikedetectionoffraud,networkanalysis,monitoring trafficovernetworks,manufacturingandenvironmentalsoftware.Thedatastreamswhicharegenerated arecontinuousandchangingovertime.Thisisthereasonwhyitbecomesnearlydifficulttodetect theoutliersintheexistingdatawhichishugeandcontinuousinnature.Thestreameddataisreal timeandchangesovertimeandhenceitisimpracticaltostoredatainthedataspaceandanalyze itforabnormalbehavior.Thelimitationsindataspacehasledtotheproblemofrealtimeanalysis ofdataandprocessingit inFCFSbasis.Theresultsregardingtheabnormalbehaviorhavetobe doneveryquicklyandinalimitedtimeframeandonaninfinitesetofdatastreamscomingoverthe networks.Toaddresstheproblemofdetectingoutliersonareal-timebasisisachallengingtaskand hencehastobemonitoredwiththehelpoftheprocessingpowerusedtodesignthegraphicsofany processingunit.Thealgorithmusedinthispaperusesakernelfunctiontoaccomplishthetask.It producestimelyoutcomeonhighspeedmulti-dimensionaldata.Thismethodincreasesthespeed ofoutlierdetectionby20timesandthespeedgoesonincreasingwiththeincreasewiththenumber ofdataattributesandinputdatarate. KEywORDS Anomaly Intrusion Detection, Compute Unified Device Architecture (CUDA), Gaussian Detection Scheme, Graphics Processing Unit (GPU), Outlier Detection, Parallel Execution

查看原文本刊更多论文

基于非参数化方法的流数据异常点并行检测

Outlierdetectionisusedinvariousapplicationslikedetectionoffraud，networkanalysis，monitoring trafficovernetworks，manufacturingandenvironmentalsoftware。Thedatastreamswhicharegenerated arecontinuousandchangingovertime。Thisisthereasonwhyitbecomesnearlydifficulttodetect theoutliersintheexistingdatawhichishugeandcontinuousinnature。Thestreameddataisreal timeandchangesovertimeandhenceitisimpracticaltostoredatainthedataspaceandanalyze itforabnormalbehavior。Thelimitationsindataspacehasledtotheproblemofrealtimeanalysis ofdataandprocessingit inFCFSbasis。Theresultsregardingtheabnormalbehaviorhavetobe doneveryquicklyandinalimitedtimeframeandonaninfinitesetofdatastreamscomingoverthe网络。Toaddresstheproblemofdetectingoutliersonareal-timebasisisachallengingtaskand hencehastobemonitoredwiththehelpoftheprocessingpowerusedtodesignthegraphicsofany processingunit.Thealgorithmusedinthispaperusesakernelfunctiontoaccomplishthetask。It producestimelyoutcomeonhighspeedmulti-dimensionaldata。Thismethodincreasesthespeed ofoutlierdetectionby20timesandthespeedgoesonincreasingwiththeincreasewiththenumber ofdataattributesandinputdatarate。关键词异常入侵检测，计算统一设备架构(CUDA)，高斯检测方案，图形处理单元(GPU)，离群点检测，并行执行

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Int. J. Synth. Emot.

自引率

0.00%

发文量