Treextrust: topic-awarecomputational trust based oninteraction experience, reputationof users with similarity, and pathalgebra of graph in social networks

• Autorzy: Tran Dinh Que

Identyfikatory

DOI

10.7494/csci.2025.26.2.5723

• Języki publikacji: EN

Abstrakty

EN

The trust measure is the confidence or reliability among users or peers, whichhas been studied widely in online social networks. Most trust models are cur-rently based on the concepts of interaction trust and reputation trust; however,various forms of interactions and analyses of the interaction contexts have notbeen considered fully for trust estimation. Moreover, the mechanism for com-puting reputation trust based on propagation lacks a clear foundation and isexpensive in computation. The purpose of this paper is to present a family ofcomputational trust models (called TreeXTrust) to estimate the trust degree ofa usertrusteron another usertrustee. Our model is a mathematical formulationthat is based on an aggregation of topic-aware experience trust with variousforms of interactions and topic-aware reputation trust with users’ similarityand operators on path algebra in a graph. We conducted experiments to eval-uate the impacts of interaction forms and users’ interests on experience trustand the correlation of experience trust and reputation trust on overall trustestimation. Our experimental results demonstrated the following: (i) interestdegrees influenced experience trust more than interaction ones did; (ii) a com-munity’s evaluation of some trustee affected an overall trust estimation morethan a truster’s individual evaluation did. Our family of models outperformedthe state-of-the-art methods that have been presented in the literature and isa framework for selecting and implementing a suitable model of computationaltrust for our problem at hand.

Słowa kluczowe

EN

computational trust; social network; interaction; interests; reputation

• Wydawca: Wydawnictwa AGH
• Czasopismo: Computer Science
• Rocznik: 2025
• Tom: T. 26 (2)
• Opis fizyczny: Bibliogr. 43 poz., rys., tab., wykr.

Twórcy

autor

Tran Dinh Que

• Department of Information Technology, Posts, and Telecommunications Institute of Technology, Ha Noi, Vietnam