Browsing by Author "Etuk, A."

Now showing 1 - 3 of 3

Metadata only
How to trust a few among many
(Springer International Publishing, 2017) Etuk, A.; Norman, T. J.; Şensoy, Murat; Srivatsa, M.; Computer Science; ŞENSOY, Murat
The presence of numerous and disparate information sources available to support decision-making calls for efficient methods of harnessing their potential. Information sources may be unreliable, and misleading reports can affect decisions. Existing trust and reputation mechanisms typically rely on reports from as many sources as possible to mitigate the influence of misleading reports on decisions. In the real world, however, it is often the case that querying information sources can be costly in terms of energy, bandwidth, delay overheads, and other constraints. We present a model of source selection and fusion in resource-constrained environments, where there is uncertainty regarding the trustworthiness of sources. We exploit diversity among sources to stratify them into homogeneous subgroups to both minimise redundant sampling and mitigate the effect of certain biases. Through controlled experiments, we demonstrate that a diversity-based approach is robust to biases introduced due to dependencies among source reports, performs significantly better than existing approaches when sampling budget is limited and equally as good with an unlimited budget.
Metadata only
Strategies for truth discovery under resource constraints
(ACM, 2015) Etuk, A.; Norman, T. J.; Oren, N.; Şensoy, Murat; Computer Science; ŞENSOY, Murat
We present a decision-theoretic approach for sampling information sources in resource-constrained environments, where there is uncertainty regarding source trustworthiness. We exploit diversity among sources to stratify the population into homogeneous subgroups to both minimise redundant sampling and mitigate the effect of source collusion. We show through empirical evaluation that our model is as effective as existing truth discovery approaches with respect to accuracy, while significantly reducing sampling cost.
Metadata only
TIDY: A trust-based approach to information fusion through diversity
(IEEE, 2013) Etuk, A.; Norman, T. J.; Şensoy, Murat; Bisdikian, C.; Srivatsa, M.; Computer Science; ŞENSOY, Murat
Trust and reputation are significant components in open dynamic systems for making informed and reliable decisions. State-of-the-art information fusion models that exploit these mechanisms generally rely on reports from as many sources as possible. Situations exist, however, where seeking evidence from all possible sources is unrealistic. Querying information sources is costly especially in resource-constrained environments, in terms of time and bandwidth. In addition, reports from multiple sources expose one to the risk of double-counting evidence, introducing an extra challenge of distinguishing fact from rumour. This paper describes TIDY (Trust-based Information fusion through DiversitY), a trust-based approach to information fusion that exploits diversity among information sources in order to select a small number of candidates to query for evidence, and to minimise the effect of correlated evidence and bias. We demonstrate that reliable decisions can be reached using evidence from small groups of individuals. We show empirically that our approach is robust in contexts of variable trust in information sources, and to a degree of deception.