Improving Prediction Accuracy in Trust-Aware Recommender Systems
Citations
273 citations
Cites background from "Improving Prediction Accuracy in Tr..."
...In order to enhance robustness, Zhang and Xu (2007) recommend utilizing topic-level trust-based prediction algorithm....
[...]
...Trust-aware recommender systems are more robust prediction schemes than traditional ones against shilling attacks (Massa and Avesani 2007; Ray and Mahanti 2010)....
[...]
233 citations
Cites background or methods from "Improving Prediction Accuracy in Tr..."
...In addition, as indicated by [27] and as a general belief, even trusted users may not share similar preference and so does the social similarity....
[...]
...In addition, most previous works are only evaluated on a single data set [4,8,25,27]....
[...]
...Nevertheless, we adopt the weighted average because the two most related works [25,27] also take the same equation....
[...]
...Another recent work using the explicit trust network is proposed by [27]....
[...]
...RN denotes the approach proposed by [27] that predicts item ratings by reconstructing the trust networks....
[...]
205 citations
Cites methods from "Improving Prediction Accuracy in Tr..."
...For example, the Epinions dataset and the Extended Epinions dataset compiled by [92] have been used to analyze the performance of trust models concerned with bootstrapping and collaborative recommendation [46], [93]–[95]; in [33], data from the FilmTrust social network are used to analyze the performance of the proposed SUNNY socio-cognitive trust model; rating information from eBay is used in [38]; the web spam dataset from Yahoo! is used in [49] to evaluate their model for propagating trust and distrust in the web; and data...
[...]
99 citations
78 citations
Cites background from "Improving Prediction Accuracy in Tr..."
...Trust-aware recommender systems are more robust prediction schemes than traditional ones against shilling attacks (Lacoste-Julien et al. 2013; Ray and Mahanti 2010)....
[...]
References
8,634 citations
"Improving Prediction Accuracy in Tr..." refers background in this paper
...There are two primary approaches which are used to build collaborative filtering (CF) memory based recommender systems, user-based CF [5] and item-based CF [12]....
[...]
1,487 citations
1,137 citations
"Improving Prediction Accuracy in Tr..." refers background in this paper
...Recent research on trust aware recommender systems [4, 6, 7, 8] has shown that they are more robust against shilling attacks and are more capable of generating recommendations for new users in the system....
[...]
...Initial research [11] on trust-aware recommender systems used trust values derived from ratings, subsequent research on trust-ware recommender systems [4,6,7] used explicitly made trust statements....
[...]
897 citations
Additional excerpts
...Initial research [11] on trust-aware recommender systems used trust values derived from ratings, subsequent research on trust-ware recommender systems [4,6,7] used explicitly made trust statements....
[...]
636 citations
"Improving Prediction Accuracy in Tr..." refers background or methods in this paper
...A detailed analysis of the data set can be found at [6, 8]....
[...]
...As trust statement values in the dataset were all 1, we use a linear decay approach for propagating trust [6]....
[...]
...Similarly in [6] superiority of trust-based systems has been experimentally shown on data from Epinions....
[...]
...Recent research on trust aware recommender systems [4, 6, 7, 8] has shown that they are more robust against shilling attacks and are more capable of generating recommendations for new users in the system....
[...]
...Trust statements are weighted, subjective, context dependent and asymmetric[6]....
[...]