Contact at or 8097636691
Responsive Ads Here

Wednesday, 31 January 2018

Trust Modeling in Social Tagging of Multimedia(2012)

Trust Modeling in Social Tagging of Multimedia(2012)

Tagging in online social networks is very popular these days, as it facilitates search and retrieval of multimedia content. However, noisy and spam annotations often make it difficult to perform an efficient search. Users may make mistakes in tagging and irrelevant tags and content may be maliciously added for advertisement or self-promotion. This article surveys recent advances in techniques for combatting such noise and spam in social tagging. We classify the state-of-the-art approaches into a few categories and study representative examples in each. We also qualitatively compare and contrast them and outline open issues for future research.
Trust modeling can be formulated as either a classification problem or a ranking problem, depending on the way of treatment. In the classification problem, the results of an algorithm can be summarized by a confusion matrix from ground-truth data and predicted labels, which contains the number of true positives, true negatives, false positives, and false negatives. From these values, classical measures such as a receiver operating characteristic (ROC), the area under the ROC curve (AUC), precision-recall (PR) curves, and F-measure can be derived.
Existing System
When information is exchanged on the Internet, malicious individuals are everywhere, trying to take advantage of the information exchange structure for their own benefit, while bothering and spamming others. Before social tagging became popular, spam content was observed in various domains: first in e-mail, and then in Web search networks have been also influenced by malicious peers, and thus various solutions based on trust and reputation have been proposed, which dealt with collecting information on peer behavior, scoring and ranking peers, and responding based on the scores . Today, even blogs are spammed. Ratings in online reputation systems, such as eBay, Amazon, and Epinions, are very similar to tagging systems and they may face the problem of unfair ratings by artificially inflating or deflating reputations. Several filtering techniques for excluding unfair ratings are proposed in the literature . Unfortunately, the countermeasures developed for e-mail and Web spam do not directly apply to social networks.
Proposed System
In a social tagging system, spam or noise can be injected at three different levels: spam content, spam tag-content association, and spammer. Trust modeling can be performed at each level separately or different levels can be considered jointly to produce trust models, for example, to assess a user’s reliability, one can consider not only the user profile, but also the content that the user uploaded to a social system. In this article, we categorize trust modeling approaches into two classes according to the target of trust, i.e., user and content trust modeling. Table 1 summarizes representative recent approaches for trust modeling in social tagging. Presented approaches are sorted based on their complexity from simple to advanced, separately for both content and user trust models.
Modules Description
Content trust modeling is used to classify content (e.g., Web pages, images, and videos) as spam or legitimate. In this case, the target of trust is a content (resource), and thus a trust score is given to each content based on its content and/or associated tags. Content trust models reduce the prominence of content likely to be spam, usually in query-based retrieval results. They try to provide better ordering of the results to reduce the exposure of the spam to users. Koutrika et al. [20] proposed that each incorrect content found in a system could be simply removed by an administrator. The administrator can go a step further and remove all content contributed by the user who posted the incorrect content, on the assumption that this user is a spammer (polluter).
The aforementioned studies consider users’ reliability as static at a specific moment. However, a user’s trust in a social tagging system is dynamic, i.e., it changes over time. The tagging history of a user is better to consider, because a consistent good behavior of a user in the past can suddenly change by a few mistakes, which consequently ruins his/her trust in tagging.
A dynamic trust score, called SocialTrust, is derived for each user. It depends on the quality of the relationship with his/her neighbors in a social graph and personalized feedback ratings received from neighbors so that trust scores are updated as the social network evolves. The dynamics of the system is modeled by including the evolution of the user’s trust score to incent long-term good behavior and to penalize users who build up a good trust rating and suddenly “defect.” It was shown that SocialTrust is resilient to the increase in number of malicious users, since the highly trusted users manage to keep them under control thanks to the trustaware feedback scheme introduced in this approach. It was also shown that SocialTrust outperforms TrustRank-based models, because SocialTrust model incorporates relationship quality and feedback ratings into the trust assessment so that bad behavior is punished.
Data sets used for development and evaluation of trust modeling techniques have a wide range of diversity in terms of content, numbers of resources, tags and users, and type of spam. Social bookmarking is the most popularly explored domain for trust modeling, especially user trust modeling.
H/W System Configuration:-
Processor               -    Pentium –III
Speed                                -    1.1 Ghz
RAM                                 -    256  MB(min)
Hard Disk                          -   20 GB
Floppy Drive                     -    1.44 MB
Key Board                         -    Standard Windows Keyboard
Mouse                                -    Two or Three Button Mouse
Monitor                              -    SVGA
S/W System Configuration:-
v   Operating System            :Windows95/98/2000/XP
v   Application  Server          :   Tomcat5.0/6.X                                                  
v   Front End                          :   HTML, Java, Jsp
v    Scripts                                :   JavaScript.
v   Server side Script             :   Java Server Pages.
v   Database                            :   Mysql
v   Database Connectivity      :   JDBC.

No comments:

Post a Comment