本文選題：協(xié)同過濾 + 隱私保護(hù)��； 參考：《北京交通大學(xué)》2017年碩士論文

【摘要】：推薦系統(tǒng)是應(yīng)用在電子商務(wù)系統(tǒng)中的一門非常成功的技術(shù),它能有效緩解由于互聯(lián)網(wǎng)飛速發(fā)展所帶來(lái)的信息超載問題,并根據(jù)人們的行為、偏好等特點(diǎn)從海量數(shù)據(jù)中挖掘用戶的潛在需求,為用戶提供個(gè)性化的推薦服務(wù)。協(xié)同過濾算法作為其中最為廣泛應(yīng)用的一類算法,它的基本思想是根據(jù)和目標(biāo)用戶具有相同愛好的用戶的偏好為目標(biāo)用戶提供預(yù)測(cè)。然而,協(xié)同過濾算法也易受到惡意用戶的攻擊,常見的攻擊模型有托攻擊模型和kNN攻擊模型。在托攻擊模型中,攻擊者會(huì)構(gòu)造一定數(shù)目的概貌特征接近真實(shí)用戶的虛假用戶來(lái)擾亂推薦算法的預(yù)測(cè),提高或是降低某些商品的預(yù)測(cè)評(píng)分;在kNN攻擊模型中,攻擊者會(huì)構(gòu)造一定數(shù)目的概貌特征和目標(biāo)用戶接近的虛假用戶來(lái)獲取用戶的隱私信息。不論哪一種攻擊,都將損害用戶的切身利益,使得用戶喪失對(duì)推薦系統(tǒng)的信任。因此推薦算法的隱私保護(hù)問題成為當(dāng)今的研究熱點(diǎn)。本文分別針對(duì)推薦算法中常見的托攻擊模型和kNN攻擊模型展開深入研究,提出解決方案,主要研究成果如下:第一,針對(duì)托攻擊模型的實(shí)現(xiàn)方式以及攻擊特點(diǎn),對(duì)當(dāng)前存在的抵抗托攻擊的主要算法展開深入研究。當(dāng)前的解決方案主要是攻擊檢測(cè)方法和魯壯性的協(xié)同過濾算法,為解決這些算法中的假正率較高、預(yù)測(cè)不準(zhǔn)確等缺點(diǎn),本文提出一種軟決策處理方法,首先應(yīng)用支持向量機(jī)方法獲取每個(gè)用戶可疑程度,然后構(gòu)建選擇鄰居的變長(zhǎng)分區(qū),最后在保證給定的安全度量標(biāo)準(zhǔn)的前提下,選擇與目標(biāo)用戶最相似的鄰居。該方法通過標(biāo)記可疑用戶而不是直接刪除他們,能有效的使被錯(cuò)誤判斷為虛假用戶的正常用戶在相似性計(jì)算中做出貢獻(xiàn),進(jìn)而降低假正率。實(shí)驗(yàn)結(jié)果表明,該算法在抵抗托攻擊時(shí)能取得較優(yōu)異的預(yù)測(cè)準(zhǔn)確性。第二,針對(duì)kNN攻擊模型的特點(diǎn),對(duì)當(dāng)前存在的隱私保護(hù)協(xié)同過濾算法展開研究。當(dāng)前的隱私保護(hù)協(xié)同過濾算法主要以加密方法、隨機(jī)擾亂方法、模糊處理方法為主,針對(duì)這些方法的計(jì)算成本高、數(shù)據(jù)實(shí)用性低、噪音量級(jí)難調(diào)節(jié)等缺點(diǎn),本文主要研究k-匿名方法在隱私保護(hù)協(xié)同過濾算法中的應(yīng)用,結(jié)合推薦算法數(shù)據(jù)集的數(shù)據(jù)特點(diǎn),提出一種新的匿名化準(zhǔn)則應(yīng)用到推薦算法中,該方法基于重要性劃分改進(jìn)微聚集算法來(lái)提高匿名化后等價(jià)類中用戶間的同質(zhì)性,以達(dá)到較好的數(shù)據(jù)實(shí)用性效果;并提出(p,l)-多樣性和(p,l,α)-多樣性模型增加用戶間的差異性,提升用戶敏感數(shù)據(jù)的隱私保護(hù)水平,其中p指攻擊者所擁有的背景知識(shí),l和(l,α)指用戶間的多樣性。實(shí)驗(yàn)結(jié)果表明,該算法能在較低信息損失量的前提下確保較高隱私保護(hù)水平。
[Abstract]:Recommendation system is a very successful technology applied in electronic commerce system. It can effectively alleviate the problem of information overload caused by the rapid development of the Internet, and according to the behavior of people, Preferences and other features mine the potential needs of users from massive data and provide personalized recommendation services for users. As one of the most widely used collaborative filtering algorithms, the basic idea of collaborative filtering algorithm is to provide prediction for target users based on the preferences of users with the same interests as the target users. However, collaborative filtering algorithms are also vulnerable to malicious users. The common attack models include trust attack model and kNN attack model. In the trust attack model, the attacker will construct a certain number of false users whose profile features are close to real users to disrupt the prediction of the recommendation algorithm and improve or lower the prediction score of some items; in the kNN attack model, An attacker will construct a certain number of profile features and a false user close to the target user to obtain user privacy information. Either attack will damage the interests of the user and make the user lose trust in the recommendation system. Therefore, the privacy protection of recommendation algorithms has become a hot topic. In this paper, we have carried out in-depth research on the common proxy attack model and kNN attack model in recommendation algorithm, and put forward the solutions. The main research results are as follows: first, aiming at the implementation mode and attack characteristics of the proxy attack model, The main algorithms of resisting trust attack are studied deeply. The current solutions are mainly attack detection methods and robust collaborative filtering algorithms. In order to solve the shortcomings of these algorithms, such as high false positive rate and inaccurate prediction, a soft decision processing method is proposed in this paper. Firstly, support vector machine (SVM) is applied to obtain the suspicious degree of each user, then the variable length partition of selecting neighbor is constructed. Finally, the neighbor that is the most similar to the target user is selected on the premise of guaranteeing the given security metric. By tagging suspicious users instead of deleting them directly, the method can effectively make the normal users who are wrongly judged as false users contribute to the similarity calculation, and then reduce the false positive rate. The experimental results show that the algorithm can achieve better prediction accuracy when resisting the support attack. Secondly, according to the characteristics of kNN attack model, the existing privacy protection collaborative filtering algorithm is studied. The current privacy protection collaborative filtering algorithms mainly include encryption method, random scrambling method and fuzzy processing method, aiming at the disadvantages of these methods, such as high calculation cost, low practicability of data, difficult to adjust the noise level, and so on. This paper mainly studies the application of k-anonymity method in privacy protection collaborative filtering algorithm. Considering the data characteristics of recommendation algorithm data set, a new anonymous criterion is proposed to apply to recommendation algorithm. This method improves the homogeneity among users in the anonymous equivalent class based on the importance partition and improves the homogeneity of the users, so as to achieve better data practicability, and proposes a model to increase the difference between users. In order to improve the privacy protection level of user sensitive data, p refers to the diversity of users with the background knowledge of attackers. Experimental results show that the proposed algorithm can ensure a high level of privacy protection under the premise of low information loss.
【學(xué)位授予單位】：北京交通大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP391.3

【參考文獻(xiàn)】

相關(guān)期刊論文 前5條

1 郭云飛;方耀寧;扈紅超;;基于Logistic函數(shù)的社會(huì)化矩陣分解推薦算法[J];北京理工大學(xué)學(xué)報(bào);2016年01期

2 賀毅朝;宋建民;張敬敏;茍海燕;;利用遺傳算法求解靜態(tài)與動(dòng)態(tài)背包問題的研究[J];計(jì)算機(jī)應(yīng)用研究;2015年04期

3 陳小輝;高燕;劉漢燁;;基于歸一化方法的協(xié)同過濾推薦算法[J];電子設(shè)計(jì)工程;2014年14期

4 張富國(guó);;Preventing Recommendation Attack in Trust-Based Recommender Systems[J];Journal of Computer Science & Technology;2011年05期

5 丁振國(guó),陳靜;基于關(guān)聯(lián)規(guī)則的個(gè)性化推薦系統(tǒng)[J];計(jì)算機(jī)集成制造系統(tǒng)-CIMS;2003年10期

相關(guān)博士學(xué)位論文 前1條

1 何賢芒;隱私保護(hù)中k-匿名算法和匿名技術(shù)研究[D];復(fù)旦大學(xué);2011年

相關(guān)碩士學(xué)位論文 前5條

1 牛曉龍;推薦系統(tǒng)中托攻擊防御方法研究[D];廣東工業(yè)大學(xué);2014年

2 黃�；�;基于用戶概貌和評(píng)分序列的混合托攻擊檢測(cè)[D];重慶大學(xué);2014年

3 段明秀;層次聚類算法的研究及應(yīng)用[D];中南大學(xué);2009年

4 張娜;電子商務(wù)環(huán)境下的個(gè)性化信息推薦服務(wù)及應(yīng)用研究[D];合肥工業(yè)大學(xué);2007年

5 史今馳;背包問題的實(shí)用求解算法研究[D];山東大學(xué);2005年

，

本文編號(hào)：1941190