Building a web of trust without explicit trust ratings

Young Ae Kim, Minh Tam Le, Hady W. Lauw, Ee Peng Lim, Haifeng Liu, Jaideep Srivastava

Research output: Chapter in Book/Report/Conference proceedingConference contribution

25 Citations (Scopus)

Abstract

A satisfactory and robust trust model is gaining importance in addressing information overload, and helping users collect reliable information in online communities. Current research on trust prediction strongly relies on a web of trust, which is directly collected from users based on previous experience. However, the web of trust is not always available in online communities and even though it is available, it is often too sparse to predict the trust value between two unacquainted people with high accuracy. In this paper, we propose a framework to derive degree of trust based on users' expertise and users' affinity for certain contexts (topics), using users rating data which is available and much more dense than direct trust data. In experiments with a real-world dataset, we show that our model can predict trust connectivity with a high degree of accuracy. With this framework, we can predict trust connectivity and degree of trust without a web of trust and then apply it to online community applications, e.g. e-commerce environments with users rating data.

Original languageEnglish
Title of host publicationProceedings - International Conference on Data Engineering
Pages531-536
Number of pages6
DOIs
Publication statusPublished - 2008
Externally publishedYes
Event2008 - IEEE 24th International Conference on Data Engineering Workshop, ICDE'08 - Cancun
Duration: 7 Apr 200812 Apr 2008

Other

Other2008 - IEEE 24th International Conference on Data Engineering Workshop, ICDE'08
CityCancun
Period7/4/0812/4/08

Fingerprint

Experiments

ASJC Scopus subject areas

  • Software
  • Engineering(all)
  • Engineering (miscellaneous)

Cite this

Kim, Y. A., Le, M. T., Lauw, H. W., Lim, E. P., Liu, H., & Srivastava, J. (2008). Building a web of trust without explicit trust ratings. In Proceedings - International Conference on Data Engineering (pp. 531-536). [4498374] https://doi.org/10.1109/ICDEW.2008.4498374

Building a web of trust without explicit trust ratings. / Kim, Young Ae; Le, Minh Tam; Lauw, Hady W.; Lim, Ee Peng; Liu, Haifeng; Srivastava, Jaideep.

Proceedings - International Conference on Data Engineering. 2008. p. 531-536 4498374.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Kim, YA, Le, MT, Lauw, HW, Lim, EP, Liu, H & Srivastava, J 2008, Building a web of trust without explicit trust ratings. in Proceedings - International Conference on Data Engineering., 4498374, pp. 531-536, 2008 - IEEE 24th International Conference on Data Engineering Workshop, ICDE'08, Cancun, 7/4/08. https://doi.org/10.1109/ICDEW.2008.4498374
Kim YA, Le MT, Lauw HW, Lim EP, Liu H, Srivastava J. Building a web of trust without explicit trust ratings. In Proceedings - International Conference on Data Engineering. 2008. p. 531-536. 4498374 https://doi.org/10.1109/ICDEW.2008.4498374
Kim, Young Ae ; Le, Minh Tam ; Lauw, Hady W. ; Lim, Ee Peng ; Liu, Haifeng ; Srivastava, Jaideep. / Building a web of trust without explicit trust ratings. Proceedings - International Conference on Data Engineering. 2008. pp. 531-536
@inproceedings{218ebeff03374925b852b86afc7dd595,
title = "Building a web of trust without explicit trust ratings",
abstract = "A satisfactory and robust trust model is gaining importance in addressing information overload, and helping users collect reliable information in online communities. Current research on trust prediction strongly relies on a web of trust, which is directly collected from users based on previous experience. However, the web of trust is not always available in online communities and even though it is available, it is often too sparse to predict the trust value between two unacquainted people with high accuracy. In this paper, we propose a framework to derive degree of trust based on users' expertise and users' affinity for certain contexts (topics), using users rating data which is available and much more dense than direct trust data. In experiments with a real-world dataset, we show that our model can predict trust connectivity with a high degree of accuracy. With this framework, we can predict trust connectivity and degree of trust without a web of trust and then apply it to online community applications, e.g. e-commerce environments with users rating data.",
author = "Kim, {Young Ae} and Le, {Minh Tam} and Lauw, {Hady W.} and Lim, {Ee Peng} and Haifeng Liu and Jaideep Srivastava",
year = "2008",
doi = "10.1109/ICDEW.2008.4498374",
language = "English",
isbn = "9781424421626",
pages = "531--536",
booktitle = "Proceedings - International Conference on Data Engineering",

}

TY - GEN

T1 - Building a web of trust without explicit trust ratings

AU - Kim, Young Ae

AU - Le, Minh Tam

AU - Lauw, Hady W.

AU - Lim, Ee Peng

AU - Liu, Haifeng

AU - Srivastava, Jaideep

PY - 2008

Y1 - 2008

N2 - A satisfactory and robust trust model is gaining importance in addressing information overload, and helping users collect reliable information in online communities. Current research on trust prediction strongly relies on a web of trust, which is directly collected from users based on previous experience. However, the web of trust is not always available in online communities and even though it is available, it is often too sparse to predict the trust value between two unacquainted people with high accuracy. In this paper, we propose a framework to derive degree of trust based on users' expertise and users' affinity for certain contexts (topics), using users rating data which is available and much more dense than direct trust data. In experiments with a real-world dataset, we show that our model can predict trust connectivity with a high degree of accuracy. With this framework, we can predict trust connectivity and degree of trust without a web of trust and then apply it to online community applications, e.g. e-commerce environments with users rating data.

AB - A satisfactory and robust trust model is gaining importance in addressing information overload, and helping users collect reliable information in online communities. Current research on trust prediction strongly relies on a web of trust, which is directly collected from users based on previous experience. However, the web of trust is not always available in online communities and even though it is available, it is often too sparse to predict the trust value between two unacquainted people with high accuracy. In this paper, we propose a framework to derive degree of trust based on users' expertise and users' affinity for certain contexts (topics), using users rating data which is available and much more dense than direct trust data. In experiments with a real-world dataset, we show that our model can predict trust connectivity with a high degree of accuracy. With this framework, we can predict trust connectivity and degree of trust without a web of trust and then apply it to online community applications, e.g. e-commerce environments with users rating data.

UR - http://www.scopus.com/inward/record.url?scp=50249085086&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=50249085086&partnerID=8YFLogxK

U2 - 10.1109/ICDEW.2008.4498374

DO - 10.1109/ICDEW.2008.4498374

M3 - Conference contribution

AN - SCOPUS:50249085086

SN - 9781424421626

SP - 531

EP - 536

BT - Proceedings - International Conference on Data Engineering

ER -