SensitiveFileDetectionMethodBasedonCNN

doi:10.3969/j.issn.1006-2475.2018.07.006

Abstract

Abstract: Inrecentyears,thepowerindustryinformationconstructionhasmadegreatachievements.Moreandmoreofficedocuments,projectdocuments,projectcontractsandotherdocumentsinvolvingindustrysecrettransmitonInternet,onthetransmissionprocess,enterprise-classsensitivedocumentsmayhavebeenleaked.Traditionalsensitivedatarecognitionmethodbasedonsensitivelexiconforfeaturedetectioncangetdetectionresultquickly,butthereisalowaccuracy,highfalsenegativesrateandfalsepositivesrate.ThispaperproposesasensitivefiledetectionmethodbasedonDeepLearning.Themethodreferstowordembeddingandconvolutionneuralnetworkalgorithmtorealizetheaccurateclassificationofsensitivedocuments.Theapproachinthispapermakesenterprisesensitivefilesdetectionindependentoffeaturekeywords,andreducesthefalsenegativerateandfalsepositiverate.

Key words: sensitivewordtable, wordembedding, convolutionneuralnetwork, deeplearning, sensitivefiledetection

CLC Number:

TP309.2

LINXue-feng1,XIAYuan-yi2,GUOJin-long1,YUXiao-wen1. SensitiveFileDetectionMethodBasedonCNN[J]. Computer and Modernization, doi: 10.3969/j.issn.1006-2475.2018.07.006.

References

［1］徐建忠,罗准辰,张亮.语义扩展技术在敏感数据识别中的应用研究［J］.现代电子技术,2016,39(12):80-82.
［2］李扬,潘泉,杨涛.基于短文本情感分析的敏感信息识别［J］.西安交通大学学报,2016,50(9):80-84.
［3］于东,荀恩东.基于WordEmbedding语义相似度的字母缩略术语消歧［J］.中文信息学报,2014,28(5):51-59.
［4］蔡慧苹,王丽丹,段书凯.基于WordEmbedding和CNN的情感分类模型［J］.计算机应用研究,2016,33(10):2902-2905.
［5］KrizhevskyA,SutskeverI,HintonGE.ImageNetclassificationwithdeepconvolutionalneuralnetworks［C］//Proceedingsofthe25thInternationalConferenceonNeuralInformationProcessingSystems.2012:1097-1105.
［6］CollobertR,WestonJ,BottonL,etal.Naturallanguageprocessing(almost)fromscratch［J］.JournalofMachineLearningResearch,2011,12(1):2493-2537.
［7］ShenYelong,HeXiaodong,GaoJianfeng,etal.LearningsemanticrepresentationsusingconvolutionalneuralnetworksforWebsearch［C］//Proceedingsofthe23rdInternationalConferenceonWorldWideWeb.2014:373-374.
［8］KalchbrennerN,GrefenstetteE，BlunsomP.Aconvolutionalneuralnetworkformodellingsentences［C］//Proceedingsofthe52ndAnnualMeetingoftheAssociationforComputationalLinguistics.2014:655-665.
［9］MikolovT,SutskeverT,ChenKai,etal.Distributedrepresentationsofwordsandphrasesandtheircompositionality［C］//Proceedingsofthe26thInternationalConferenceonNeuralInformationProcessingSystems.2013:3111-3119.
［10］BengioY,DucharmeR,VincentP,etal.Aneuralprobabilisticlanguagemodel［J］.JournalofMachineLearningResearch,2003,3(6):1137-1155.
［11］MikolovT,KarafiátM,BurgetL,etal.Recurrentneuralnetworkbasedlanguagemodel［C］//Proceedingsofthe11thAnnualConferenceoftheInternationalSpeechCommunicationAssociation.2010:1045-1048.
［12］HintonGE,SalakhyurdinovRR.Reducingthedimensionalityofdatawithneuralnetworks［J］.Science,2006,313(5786):504-507.
［13］HintonGE,OsinderoS,TehYW.Afastlearningalgorithmfordeepbeliefnets［J］.NeuralComputation,2006,18(7):1527-1554.
［14］GoodfelloI,BengioY,CourvilleA.深度学习［M］.赵申剑,符天凡,李凯,等译.北京:人民邮电出版社，2017.
［15］DomingosP.终极算法［M］.黄芳萍译.北京:中信出版集团股份有限公司,2017.
［16］HarringtonP.机器学习实战［M］.李锐,李鹏,曲亚东译.北京:人民邮电出版社,2013.
［17］郑泽宇,顾思宇.TensorFlow：实战Google深度学习框架［M］.北京:电子工业出版社，2017.
［18］何宇健.Python与机器学习实战：决策树、集成学习、支持向量机与神经网络算法详解及编程实现［M］.北京:电子工业出版社,2017.

[1]	HEJun-jie,XIAOKe,LIUChang,CHENSong-yan. TFT-LCDCircuitDefectsDetectionBasedonFasterR-CNN [J]. Computer and Modernization, 2018, 0(07): 33-.
[2]	ZHANGWen. Multi-taskFaceAttributeRecognitionBasedonUnbalancedClasses [J]. Computer and Modernization, 2018, 0(06): 58-.