Trust Modeling in Recommendation

Date

2021-12-09

Author

Demirci, Mehmet Utku
Karagöz, Pınar

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

251
views

0
downloads

In social networks, trust is a fundamental notion aecting the na-ture and the strength of ties between individuals. It is also a pieceof useful auxiliary information for improving the performance ofrecommendation systems. The number of ratings given by a user isminimal compared to all items in popular, widely-used e-commercesites. Therefore, the user-item matrix that is used in collaborativeltering suers from data sparsity, resulting in poor recommenda-tion quality. Another issue is the cold start problem, which occursfor the inclusion of new users and new items to the system. Trustnotion is helpful for alleviating the eect of these problems byproviding additional relationships between the users and pointingout strong relationships. Information as to the trust between userscan be explicitly available. However, such information is not widelyavailable, and hence implicit trust models have been employed.This work analyzes two sub-problems under trust modeling forrecommendation: (1) What is the relationship between explicit andimplicit trust scores, are they replaceable? (2) Can we model ex-plicit trust in a trust network? For the rst problem, we present animplicit trust model and analyze the compatibility of implicit andexplicit trust scores. For the second problem, we model explicit trustmodeling as a link prediction problem and analyze the performanceof the prediction models we generate on a set of benchmark datasets

Subject Keywords

Trust modeling, Implicit trust, Explicit trust, Recommendation, Rec-ommender systems, Supervised learning

URI

http://dx.doi.org/10.1145/3444757.3485105
https://hdl.handle.net/11511/95429

DOI

https://doi.org/10.1145/3444757.3485105

Conference Name

MEDES '21: Proceedings of the 13th International Conference on Management of Digital EcoSystem

Collections

Department of Computer Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Explicit and implicit trust modeling compatibility and explicit trust link prediction in trust-based recommendation systems Demirci, Mehmet Utku; Karagöz, Pınar; Department of Computer Engineering (2022-1-31) In social networks, trust is a fundamental notion affecting the nature and the strength of ties between individuals. It is also a piece of useful auxiliary information for improving the performance of recommendation systems. The number of ratings given by a user is minimal compared to all items in popular, widely-used e-commerce sites. Therefore, the user-item matrix that is used in collaborative filtering suffers from data sparsity, resulting in poor recommendation quality. Another issue is the cold start ...
Trust in principals Zayim Kurtay, Merve (Oxford University Press, 2020-09-01) Trust is generally considered as the basis of relationships between major school stakeholders and contributes not only to productive attitudes and behaviors but also to the accomplishment of collective goals. The principal is the key person in creating the atmosphere and the conditions conducive to trust in which trust-based relationships among school constituencies can flourish. Specifically, school culture characterized by collegiality, professionalism, and consideration generally leads to higher trust in...
The relationship between regulatory focus and trust: The roles of regulatory closure and propensity to trust Emirza, Sevgi; Karagonlar, Gökhan (Orta Doğu Teknik Üniversitesi (Ankara, Turkey), 2020-6) Drawing on regulatory focus theory and regulatory closure argument, this study argues that self-regulation mechanisms influence trust among people during anonymous interactions. Prevention focus is generally associated with risk aversion and quest for security and hence expected to induce less trust compared to promotion focus. We proposed and tested the hypothesis that the effect of regulatory focus on trust is moderated by regulatory closure (presence or absence of goal fulfillment). In support of th...
TRUST IN BANKS: A MULTI-LEVEL ANALYSIS Öztürk, Yasemin; Öztürk, Yasemin; Department of Financial Mathematics (2023-1-06) Trust is required for a healthy and well-functioning financial system. However, limited evidence is present on what affects confidence towards the banks. Utilizing a sample drawn from the latest waves of World Value Survey (WVS), this thesis constructs multilevel- logit models to account for factors at both the micro and macro level of individuals. The findings confirm the nested structure of trust in banks. The findings of this study show that the majority of socio-economic characteristics of individuals e...
Trust-Driven Entrepreneurship for Community Well-Being of Refugees and Their Local Hosts Baktir, Zeynep; Watson, Forrest (SAGE Publications, 2020-07-01) This article explores community well-being (CWB) outcomes of institutional trust (and distrust) through entrepreneurship in the context of a refugee-hosting society in Turkey. Existing studies show the positive relationship between institutional trust and quality of life (QOL) as well as entrepreneurship and QOL in subsistence contexts. This research, however, explores the relationship between institutional (dis)trust and entrepreneurship on a path to CWB with a special emphasis on refugee and local interac...

Citation Formats

M. U. Demirci and P. Karagöz, “Trust Modeling in Recommendation,” presented at the MEDES ’21: Proceedings of the 13th International Conference on Management of Digital EcoSystem, Tunus, 2021, Accessed: 00, 2022. [Online]. Available: http://dx.doi.org/10.1145/3444757.3485105.