Dataset Information

Exploiting implicit social relationships via dimension reduction to improve recommendation system performance.

ABSTRACT: The development of Web 2.0 and the rapid growth of available data have led to the development of systems, such as recommendation systems (RSs), that can handle the information overload. However, RS performance is severely limited by sparsity and cold-start problems. Thus, this paper aims to alleviate these problems. To realize this objective, a new model is proposed by integrating three sources of information: a user-item matrix, explicit and implicit relationships. The core strategy of this study is to use the multi-step resource allocation (MSRA) method to identify hidden relations in social information. First, explicit social information is used to compute the similarity between each pair of users. Second, for each non-friend pair of users, the MSRA method is applied to determine the probability of their relation. If the probability exceeds a threshold, a new relationship will be established. Then, all sources are incorporated into the Singular Value Decomposition (SVD) method to compute the missing prediction values. Furthermore, the stochastic gradient descent technique is applied to optimize the training process. Additionally, two real datasets, namely, Last.Fm and Ciao, are utilized to evaluate the proposed method. In terms of accuracy, the experiment results demonstrate that the proposed method outperforms eight state-of-the-art approaches: Heats, PMF, SVD, SR, EISR-JC, EISR-CN, EISR-PA and EISR-RAI.

SUBMITTER: Al-Sabaawi AMA

PROVIDER: S-EPMC7176113 | biostudies-literature | 2020

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Exploiting implicit social relationships via dimension reduction to improve recommendation system performance.

Al-Sabaawi Ali M Ahmed AMA Karacan Hacer H Yenice Yusuf Erkan YE

PloS one 20200422 4

The development of Web 2.0 and the rapid growth of available data have led to the development of systems, such as recommendation systems (RSs), that can handle the information overload. However, RS performance is severely limited by sparsity and cold-start problems. Thus, this paper aims to alleviate these problems. To realize this objective, a new model is proposed by integrating three sources of information: a user-item matrix, explicit and implicit relationships. The core strategy of this stu ...[more]

PMID: 32320416

Similar Datasets

Project description:BackgroundThe race-based Implicit Association Test (IAT) was proposed to measure individual differences in implicit racial bias subsumed within social cognition. In recent years, researchers have debated the theoretical tenets underpinning the IAT, questioning whether performance on this task: (1) measures implicit attitudes that operate automatically outside of conscious awareness; (2) reflects individual differences in social cognition; and (3) can predict social behavior. One way to better address these research questions is to assess whether the race-IAT correlates with other implicit processes that are subsumed within social cognition.AimsThe current study assessed whether the race-IAT was related to other commonly used individual difference measures of implicit (and explicit) social cognition. Experiment 1 assessed whether dissociable patterns of performance on the race-IAT were related to measures of implicit imitative tendencies, emotion recognition and perspective taking toward White task actors, as well as explicit measures of trait and state affective empathy and racial bias. Overcoming limitations of task conceptual correspondence, Experiment 2 assessed whether these latter tasks were sensitive in detecting racial biases by using both White and Black task actors and again examined their relationships with the race-IAT.MethodIn two lab-based experiments, 226 and 237 participants completed the race-IAT followed by an extensive battery of social cognition measures.ResultsAcross both experiments, pro-White/anti-Black bias on the race-IAT was positively related to a pro-White bias on explicit measures of positive affective empathy. However, relationships between the race-IAT and implicit imitative tendencies, perspective taking, emotion recognition, and explicit trait and negative state affective empathy were statistically equivalent.ConclusionThe race-IAT was consistently related to explicit measures of positive state affective empathy but not to other individual difference measures of implicit social cognition. These findings are discussed with regards to the theoretical underpinnings of the race-IAT as an individual difference measure of implicit social cognition, as well as alternative explanations relating to the reliability of social cognition measures and the various combinations of general-purpose (social and non-social) executive processes that underpin performance on these tasks.

Dataset Information

Exploiting implicit social relationships via dimension reduction to improve recommendation system performance.

Publications

Exploiting implicit social relationships via dimension reduction to improve recommendation system performance.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets