本文選題：社會(huì)網(wǎng)絡(luò) + 隱私保護(hù)��； 參考：《河南大學(xué)》2014年碩士論文

【摘要】：近年來(lái)，隨著互聯(lián)網(wǎng)技術(shù)的不斷發(fā)展，社交網(wǎng)絡(luò)產(chǎn)品也在不斷的融入我們的生活中。從QQ、人人網(wǎng)到微博、微信，，社交網(wǎng)絡(luò)漸漸成為我們生活中不可或缺的一部分。但是，社交網(wǎng)絡(luò)在提供給我們便利的同時(shí)也對(duì)我們個(gè)人隱私保護(hù)及社會(huì)關(guān)系隱私保護(hù)提出了新的挑戰(zhàn)�，F(xiàn)階段在傳統(tǒng)的關(guān)系型數(shù)據(jù)庫(kù)隱私保護(hù)研究領(lǐng)域已經(jīng)有了很多科研成果，但是由于社會(huì)網(wǎng)絡(luò)的數(shù)據(jù)模型是類似于計(jì)算機(jī)圖論中圖的結(jié)構(gòu)，因此我們?cè)谔幚砩鐣?huì)網(wǎng)絡(luò)隱私保護(hù)問(wèn)題時(shí)顯然不能直接套用針對(duì)傳統(tǒng)關(guān)系型數(shù)據(jù)庫(kù)的隱私保護(hù)方法。然而，伴隨著大數(shù)據(jù)時(shí)代的到來(lái)，我們一般處理的社會(huì)網(wǎng)絡(luò)數(shù)據(jù)也是海量的，人工處理顯然不現(xiàn)實(shí)，因此社會(huì)網(wǎng)絡(luò)的隱私保護(hù)問(wèn)題必是當(dāng)前研究的熱點(diǎn)問(wèn)題也是未來(lái)計(jì)算機(jī)技術(shù)必然的研究趨勢(shì)。目前我們使用社會(huì)網(wǎng)絡(luò)軟件主要是為了與他人共享或者交換信息資源，單純的個(gè)人信息隱私保護(hù)已經(jīng)不能滿足需求，對(duì)個(gè)人社會(huì)關(guān)系隱私保護(hù)的研究是目前的熱門(mén)研究領(lǐng)域。 本文主要從數(shù)據(jù)挖掘的角度對(duì)k-匿名算法進(jìn)行研究。首先介紹了現(xiàn)階段社會(huì)網(wǎng)絡(luò)隱私保護(hù)研究的國(guó)內(nèi)外現(xiàn)狀及其概念和特點(diǎn)，針對(duì)性的分析了攻擊社會(huì)網(wǎng)絡(luò)的幾種方式，并對(duì)現(xiàn)階段幾種匿名算法進(jìn)行了介紹。以此為基礎(chǔ)，借鑒他人已有的研究思想，對(duì)原k-對(duì)稱匿名算法給予改進(jìn)，并設(shè)計(jì)出一種有效地還原算法，找出一個(gè)推導(dǎo)出k值的公式。k-對(duì)稱匿名方法是一種隱私保護(hù)算法，對(duì)社會(huì)網(wǎng)絡(luò)中的節(jié)點(diǎn)進(jìn)行對(duì)稱處理，使得等價(jià)類的結(jié)果中每個(gè)集合都包括k個(gè)節(jié)點(diǎn)，這就使得攻擊者識(shí)別目標(biāo)個(gè)體的概率不高于1/k。還針對(duì)k-對(duì)稱匿名方法的可用性分析提出一種能還原出原社會(huì)網(wǎng)絡(luò)圖的還原算法。最后，論文基于微信討論組的社會(huì)網(wǎng)絡(luò)數(shù)據(jù)，實(shí)現(xiàn)了k-對(duì)稱匿名發(fā)布，評(píng)估了這種匿名發(fā)布方法的可用性，并且驗(yàn)證了有效性。
[Abstract]:In recent years, with the continuous development of Internet technology, social network products are constantly integrated into our lives. From QQ, Renren to Weibo, WeChat, social networks are becoming an integral part of our lives. However, social networks not only provide us with convenience, but also pose new challenges to our privacy protection and social privacy protection. At present, a lot of achievements have been made in the field of privacy protection of traditional relational database, but the data model of social network is similar to the structure of graph in computer graph theory. Therefore, when we deal with the social network privacy protection problem, we obviously can not directly apply the traditional relational database privacy protection method. However, with the arrival of the big data era, the social network data we generally deal with is also massive, and manual processing is obviously not realistic. Therefore, the privacy protection of social networks must be the hot topic of current research and the inevitable research trend of computer technology in the future. At present, we use social network software mainly to share or exchange information resources with others. The simple privacy protection of personal information can no longer meet the needs. The research on privacy protection of personal social relations is a hot research field at present. In this paper, the k-anonymity algorithm is studied from the point of data mining. Firstly, this paper introduces the present situation, concepts and characteristics of the research on privacy protection of social networks at home and abroad, analyzes several ways of attacking social networks, and introduces several anonymous algorithms at the present stage. On this basis, the original k-symmetric anonymous algorithm is improved, and an effective algorithm is designed to find a formula to deduce k value. The method of k-symmetric anonymity is a privacy protection algorithm. The nodes in the social network are treated symmetrically so that each set of the equivalent class includes k nodes which makes the probability of the attacker to identify the target individual not more than 1 / 1 / k. Based on the availability analysis of k-symmetric anonymous method, an algorithm is proposed to restore the original social network graph. Finally, based on the social network data of WeChat discussion Group, the paper implements k-symmetric anonymous publishing, evaluates the availability of this anonymous publishing method, and verifies its effectiveness.
【學(xué)位授予單位】：河南大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TP393.08

【參考文獻(xiàn)】

相關(guān)期刊論文 前3條

1 王歡,郭玉錦;網(wǎng)絡(luò)社區(qū)及其交往特點(diǎn)[J];北京郵電大學(xué)學(xué)報(bào)(社會(huì)科學(xué)版);2003年04期

2 蘭麗輝;鞠時(shí)光;金華;;社會(huì)網(wǎng)絡(luò)數(shù)據(jù)發(fā)布中的隱私保護(hù)研究進(jìn)展[J];小型微型計(jì)算機(jī)系統(tǒng);2010年12期

3 蘭麗輝;鞠時(shí)光;金華;;用二分圖實(shí)現(xiàn)社會(huì)網(wǎng)絡(luò)的匿名發(fā)布[J];小型微型計(jì)算機(jī)系統(tǒng);2011年10期

本文編號(hào)：1887307