Towards Leader Based Recommendations

Esslimani, Ilham; Brun, Armelle; Boyer, Anne

doi:10.1007/978-3-7091-1346-2_20

Cited by 3 publications

(11 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 4 shows the MAE values obtained by testing of our approach on TijuanaRestaurant dataset and comparing it with degree centrality approach (Esslimani et al, 2013). The latter has been applied on the same dataset.…”

Section: Mae Resultsmentioning

confidence: 99%

“…As the scores of the credibility components can exceed their thresholds, we have normalised them using sigmoid function. The latter varies between [0, 1], and is defined by (Esslimani et al, 2013):…”

Section: The Personality Of the Usermentioning

confidence: 99%

“…This technique aims to identify the set of most important delegates, called 'opinion leaders', who can represent as many users as possible (Victor et al, 2008;Esslimani et al, 2013) to effectively address the cold start challenges. Indeed, opinion leaders are highly confident users with strong opinions (Al-Sharawneh and Williams, 2010) and effective connections.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SCOL: similarity and credibility-based approach for opinion leaders detection in collaborative filtering-based recommender systems

Chekkai¹,

Chorfi²,

Meshoul

et al. 2020

IJRIS

View full text Add to dashboard Cite

Her current research interests include social networks, information retrieval, recommender systems, natural language processing, graph theory, big data, elastic research, machine learning, context aware systems, plagiarism detection and statistical modelling.Ilyes Chorfi was an Associate Professor from the Badji Mokhtar Annaba University. He studied mathematics and obtained his PhD from the latter, with the collaboration of the University of Strasbourg where he did three research internships at the Institute of Advanced Mathematical Research IRMA. In 2017, he obtained his mathematics qualification in France then joined the Department of Mathematics at the AIX-Marseille University, France, in the same year. He has published many articles on the theory of probabilities and financial mathematics fields, and regularly takes part in conferences and congresses in financial mathematics, probability and statistics. His current research interests include the modelling applications in actuarial science as well as the theory of similarity that intervenes in the artificial intelligence.

show abstract

Section: Mae Resultsmentioning

confidence: 99%

Section: The Personality Of the Usermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SCOL: similarity and credibility-based approach for opinion leaders detection in collaborative filtering-based recommender systems

Chekkai¹,

Chorfi²,

Meshoul

et al. 2020

IJRIS

View full text Add to dashboard Cite

show abstract

“…Some approaches in this direction have been proposed. For example, in the frame of neighborhood-based models, Esslimani et al (2013) proposes to discover seed users based on their connectivity and average similarity. Rashid (2007) proposes to evaluate the importance of users as the negative influence rendered on the quality of recommendations, when these users are removed from the system.…”

Section: Cold-start Problem In Collaborative Filteringmentioning

confidence: 99%

“…These users should represent the interests of the whole population as fully as possible and/or be capable to influence the preferences of others. Such set of users is referred to as seed users or seeds , representative users , influential users (Rashid 2007), power users (Seminario and Wilson 2014) or leaders (Esslimani et al 2013).…”

Section: Cold-start Problem In Collaborative Filteringmentioning

confidence: 99%

Identifying representative users in matrix factorization-based recommender systems: application to solving the content-less new item cold-start problem

et al. 2016

Self Cite

View full text Add to dashboard Cite

Matrix factorization has proven to be one of the most accurate recommendation approaches. However, it faces one major shortcoming: the latent features that result from the factorization are not directly interpretable. Providing interpretation for these features is important not only to help explain the recommendations presented to users, but also to understand the underlying relations between the users and the items. This paper consists of 2 contributions. First, we propose to automatically interpret features as users, referred to as representative users. This interpretation relies on the study of the matrices that result from the factorization and on their link with the original rating matrix. Such an interpretation is not only performed automatically, as it does not require any human expertise, but it also helps to explain the recommendations. The second proposition of this paper is to exploit this interpretation to alleviate the content-less new item cold-start problem. The experiments conducted on several benchmark datasets confirm that the features discovered by a NonNegative Matrix Factorization can be interpreted as users and that representative users are a reliable source of information that allows to accurately estimate ratings on new items. They are thus a promising way to solve the new item cold-start problem.

show abstract