[雙語翻譯]網(wǎng)絡隱私保護外文翻譯--使用個性化網(wǎng)絡搜索的客戶端隱私保護（英文）

1、 Procedia Computer Science 79 ( 2016 ) 1029 – 1035 Available online at www.sciencedirect.com1877-0509 © 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND licens

2、e (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the Organizing Committee of ICCCV 2016 doi: 10.1016/j.procs.2016.03.130 ScienceDirect7th International Conference on Communicat

3、ion, Computing and Virtualization 2016 Client side Privacy Protection Using Personalized Web Search Mrs. Sharvari V. Malthankar , Prof. Shilpa Kolte PG Student,Saraswati College of Engg.Kharghar,NaviMumbai,India. Assis

4、tant Professor, Saraswati College of Engg.Kharghar,NaviMumbai,India. Abstract We are providing a Client-side privacy protection for personalized web search.. Any PWS captures user profiles in a hierarchical taxonomy.

5、The system is performing online generalization on user profiles to protect the personal privacy without compromising the search quality and attempt to improve the search quality with the personalization utility of th

6、e user profile. On other side they need to hide the privacy contents existing in the user profile to place the privacy risk under control. User privacy can be provided in form of protection like without compromising th

7、e personalized search quality. In general we are working for a trade off between the search quality and the level of privacy protection achieved from generalization. © 2016 The Authors. Published by Elsevier B.V.

8、 Peer-review under responsibility of the Organizing Committee of ICCCV 2016. Keywords: UPS,Privacy Protection, Greedy DP,Greedy IL; 1. Introduction The web search engine has long become the most important portal for or

9、dinary people looking for useful information on the web. users may experience failure when search engines return irrelevant results that do not meet their real and expected intentions. Such irrelevant think is largely

10、due to the enormous variety of users’ contexts and backgrounds, as well as the ambiguity of the texts. Personalized web search (PWS) is one general search techniques aiming to providing better search results, which are

11、 tailored to individual user needs. At the expense, user information has to be collected and analyzed to figure out the user intention behind the issued query. PWS can generally into two types ? Click-log-based method

12、s and ? Profile-based methods © 2016 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under res

13、ponsibility of the Organizing Committee of ICCCV 20161031Sharvari V. Malthankar and Shilpa Kolte / Procedia Computer Science 79 ( 2016 ) 1029 – 1035 search services on the Internet. However, evidences show that u

14、sers' reluctance to disclose their private information during search has become a major barrier to the wide proliferation of PWS [4]. ? We study a large-scale evaluation framework for personalized search based on

15、query logs and then evaluate five personalized search algorithms (including two clicks-based ones and three topical-interest- based ones) using 12-day query logs of Windows Live Search. By analyzing the results, we reve

16、al that personalized Web search does not work equally well under various situations [5]. ? Long-term search history contains rich information about a user’s search preferences, which can be used as search context to i

17、mprove retrieval performance. The user profiles for particular users are stored on the clients, thus preserving the privacy of the users. The design adopts the server-client model in which user queries are forwarded to

18、 a server for processing the training and re-ranking quickly [6]. ? The proposed introduces vector quantization approach piecewise on the datasets which segment, each row of datasets and quantization approach is perfor

19、med on each segment, using the proposed approach which later are again united to transformed data set [7]. ? We study private safety in pws applications that representation user desire as hierarchical user profiles. We

20、 are providing a private requirement using a pws framework ups. Two predictive metrics utility of personalization and the privacy risk are used for build – up of the profile. In the generalization process we use greed

21、y DP and the greedy IL algorithm. The innovative outcome tells that greedy IL obviously outperforms greedy DP in terms of efficiency [8]. ? We propose a method that, given a query submitted to a search engine, suggests

22、 a list of related queries. The related queries are based on previously issued queries, and can be issued by the user to the search engine to tune or redirect the search process. The method proposed is based on a query

23、 clustering process in a group of semantically similar queries are identified. [9]. ? We proposed the reliability of implicit feedback generated from click through data in WWW search. Analyzing the users’ decision pro

24、cess using eye tracking and comparing implicit feedback against manual relevance judgments, we conclude that clicks are informative but biased. We show that relative preferences derived from clicks are reasonably accur

25、ate on average [10]. ? We propose a novel context-aware query suggestion approach.. In which steps for in an offline model learning step, to address data sparseness, queries are summarized into concepts by clustering a

26、 click- through bipartite. Then, from session data a concept sequence suffix tree is constructed as the query suggestion model [11]. 3. PROBLEM DEFINITION To protect user privacy in profile-based PWS, we have to conside

27、r two contradicting effects during the search process. On the one hand, they attempt to improve the search quality with the personalization utility of the user profile. They need to hide the privacy contents in existin

28、g user profile to place the privacy risk under control. Significant gains can be obtained by personalization at the expense of only a small and less-sensitive portion of the user profile, namely a generalized profile.

29、Thus, user privacy can be protected without compromising the personalized search quality The existing profile-based Personalized Web Search does not support runtime profiling. A user profile is typically generalized fo

30、r only once offline, and used to personalize all queries from a same user indiscriminatingly. Such “one profile is fits all” strategy certainly has drawbacks given the variety of queries. One evidence reported in is tha