Enhancing the trust-based recommendation process with explicit distrust

Victor, Patricia; Verbiest, Nele; Cornelis, Chris; Cock, Martine De

doi:10.1145/2460383.2460385

Cited by 36 publications

(32 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At first, this prediction is based on (3). If this equation is not able to predict the rating i for the user u, prediction of this rating will be given to (6), but before doing the prediction step, it is needed to calculate the probability values of selecting trusted friends and finding trusted friends and this is done according to (4) and (5). Pheromone updating is provided to increase the trust rate of the target user for users who participated in the prediction and later, these users will be selected with higher probability in the future predictions.…”

Section: Proposed Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

A New Strategy in Trust-Based Recommender System using K-Means Clustering

Sani¹,

Tabriz²

2017

ijacsa

View full text Add to dashboard Cite

Abstract-Recommender systems are among the most important parts of online systems, including online stores such as Amazon, Netflix that have become very popular in the recent years. These systems lead users to finding desired information and goods in electronic environments. Recommender systems are one of the main tools to overcome the problem of information overload. Collaborative filtering (CF) is one of the best approaches for recommender systems and are spreading as a dominant approach. However, they have the problem of coldstart and data sparsity. Trust-based approaches try to create a neighborhood and network of trusted users that demonstrate users' trust in each other's opinions. As such, these systems recommend items based on users' relationships. In the proposed method, we try to resolve the problems of low coverage rate and high RMSE rate in trust-based recommender systems using kmeans clustering and ant colony algorithm (TBRSK). For clustering data, the k-means method has been used on MovieLens and Epinion datasets and the rating matrix is calculated to have the least overlapping.

show abstract

Section: Proposed Algorithmmentioning

confidence: 99%

“…Then, users who are the most similar to the active user are selected as his neighbors [3]. Finally, using the neighbor ensemble, the user's interests in unrated items is predicted [4].…”

Section: Introductionmentioning

confidence: 99%

A New Strategy in Trust-Based Recommender System using K-Means Clustering

Sani¹,

Tabriz²

2017

ijacsa

View full text Add to dashboard Cite

show abstract

“…Three other credibility algorithms are selected to compare with the improved algorithm proposed in this paper, which named as: BN [6], TCF [7], SRP-CCF [2]. The algorithm accuracy is denoted in Figure 3 and Figure 4.…”

Section: The Comparison Of Some Credibility Algorithmsmentioning

confidence: 99%

Credibility Evaluation Algorithm Based on Deep Learning

Liu¹,

Li²

2017

ACM

View full text Add to dashboard Cite

Abstract:The credibility of a recommendation system is a hot focus nowadays in the field of personalized recommendation research. However, it is difficult to carry out effective credibility evaluation for the users in the presence of a false recommendation system, say nothing of eliminating suspicious users and further more improve the security and reliability of the system. This paper proposed a new method of reliability assessment based on deep learning. According to the users' rating database, community of users with average scores is constructed and traditional credibility algorithm is used to calculate the initial credibility of the users. With the average users' reliability value as a criterion, the second assessment to the credibility based on deep learning algorithm is applied to other users, the results of which are arranged in ascending order. Then suspicious users ranking top-L will be removed and a trustfully adjacent group for the target users will be created. Experiments show that the improved algorithm can optimize the recommendation system with better security, accuracy and reliability as well.

show abstract

“…Golbeck's formula (8) for generating recommendation in presence of trust has been modified by Patricia [15] in order to highlight the distrust angle also. The modified formula is given by equation (12) where all the neighbors who have rated the target product are considered in the process of recommendation generation but their importance is …”

Section: Incorporating Distrust In Recommendation Systemmentioning

confidence: 99%