基于关联规则的隐私保护算法研究-计算机应用专业论文.docx

研究生优秀毕业论文 硕士学位论文 硕士学位论文 基于关联规则的隐私保护算法研究 Abstract As the cominuous deVelopment of inf0咖ation tecllIlology，data showing explosive 群。坝h，it encourages data mining technology has been deVeloped fapidly．Howeverwith也e a刚ication of data miIling teclmiques data埘vaCy problems ensued．Wim people’s attention degree of priVacy increasingly high，priVacy protection technology also gradually become a hot research． PriVacy preserving data mining tecllIlology is sensitiVe data pmtection application method in data mining．The main pu叩ose of mis tecllIlology is to ensure the accuracy of me data mining results while impmVing to sensitiVe data，knowledge of privacy protection．It is also now the main direction of t11e people in me study of pdVacy protection． This paper mainly studies on association mles mining algorithm privacy protection．The paper introduces tlle aSsociation mle mining aIld priVacy-related concepts and teclulologies， And it analyzes me existing two kinds of PriVacy Protection aIgoritlⅢls baSed on aSsociation rules．For the lack of time emciency aIld rules loss rate on the algorit№s，、ve propose improved methods： (1)In mis paperwe make a detailed analysis based on Randomized Response with Partial Hiding(对冲H)method．For the inadequate of the emciency of support reconstruction on this method， we use diVide and conquer strategy to 叩timize t|le solving of the transfomation matrix in support the reconstruction．And we use the set operations methods to optimize the counting of the distoning data set support．Finally，according to experiments and analysis，demonstrate the ea’ectiVeness of the improved algorithm． (2)In this p印er，we analysis the priVacy protection of aSsociation rule mining(PPARM) algorit№．Found that the loss rate of rules is relatiVely high，while the privacy protection method hiding SensitiVe rules．Tb s。lVe fhis problem，￡his paper proposes a pf如aey pfotection