International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 2 Issue: 5 1214– 1219 _______________________________________________________________________________________________ 1214 IJRITCC | May 2014, Available @ http://www.ijritcc.org _______________________________________________________________________________________ Reputation Management using Trust Based Decision Making System through Temporal and Correlation Analysis Dr. S. Jabeen Begum, Associate Professor & Head/ IT, Velalar College of Engineering and Technology, Erode – 638012, Tamilnadu, India. e-mail : sjabeenbegum@gmail.com Mr. G. Rajesh Kumar Assistant Professor (Sr.Gr.) Velalar College of Engineering and Technology, Erode – 638012, Tamilnadu, India. e-mail:grajesh.grk@gmail.com R. Varanambigai IT Final Year IT Student, Velalar College of Engineering and Technology, Erode – 638012, Tamilnadu, India. Abstract — With the rapid development of online reputation systems in various online social networks, manipulations against such online reputation systems are evolving quickly. Due to the anonymity of the Internet, it is very difficult for normal users to evaluate a stranger’s trustworthiness and quality, which makes online interactions risky. TATA, the abbreviation of joint Temporal And Trust Analysis, which protects online reputation systems from a new angle: the combination of time domain anomaly detection and Dempster–Shafer theory-based trust computation. The problem is how the online participants protect themselves by judging the quality of strangers or unfamiliar items beforehand. To address this problem, Online Reputation System (ORS) have been built up. The goal is to create large-scale virtual word-of- mouth networks where individuals share opinions and experiences, in terms of reviews and ratings, on various items, including products, services, digital contents and even other people. Keywords-component; formatting; style; styling; insert (key words) __________________________________________________*****_________________________________________________ I. INTRODUCTION The Internet has been very beneficial to daily life by providing vast information and convenient services. For instance, electronic mail reduces the message sending time and makes communication much easier. Online shopping makes it possible to purchase at home. Search engines can get the relevant information immediately. Moreover, the Internet has enabled the proliferation of online business and interpersonal interactions between individuals who have never interacted before. Usually, these interactions are not completed without a certain concern given that private information as well as the exchange of money and goods are involved. An online reputation system is an approach to systematically evaluate opinions of online community members on various issues (e.g., products, services, events, etc.) and their opinions on the trustworthiness of other community members. Online reputation systems first collect and combine all relevant opinions, draw conclusions about the trustworthiness of all opinions from the subjective perspective of a given user and calculate the trustworthiness of all opinions referring to certain issues. Then, all opinions referring to a particular issue are combined according to their trustworthiness, and the result is returned to the requesting user or application, where it can be used to make a decision, e.g., to recommend the highest ranked restaurant. The use of online reputation systems has been proposed for various applications, for example to validate the trustworthiness of sellers and buyers in online auctions, to detect free-riders in peer-to-peer networks and to ensure the authenticity of signature keys in a web of trust. As more people use the Internet for entertainment, building personal relationships, and conducting businesses, the Internet has created vast opportunities for online interactions. However, due to the anonymity of the Internet, it is very difficult for normal users to evaluate a stranger’s trustworthiness and quality, which makes online interactions risky. To address this problem, online reputation systems have been built up. The goal is to create large-scale virtual word-of-mouth networks where individuals share opinions and experiences, in terms of reviews and ratings, on various items, including products, services, digital contents and even other people. These opinions and experiences, which are called users’ feedback, are collected as evidence, and are analysed, aggregated, and disseminated to general users. The disseminated results are called reputation score. Such systems are also referred to as feedback-based online reputation systems. A reputation defense scheme, named TATA, for feedback-based online reputation systems. Here, TATA is the abbreviation of joint Temporal And Trust Analysis. It contains two modules: a time domain anomaly detector and a trust model based on the Dempster–Shafer theory. Specifically, considering the ratings to a given item as a time sequence, and a time domain anomaly detector is introduced to detect suspicious time intervals where anomaly occurs. A trust analysis is then conducted based on the anomaly detection results. The concept of user behavior uncertainty from the Dempster–Shafer theory to model users’ behavior patterns, and evaluate whether a user’s rating value to each item is reliable or not. The performance of TATA, two other representative reputation schemes, and previous scheme TATA is evaluated against real user attack data collected through a cyber- competition. TATA demonstrates significant advantages in