Semantic Representation and Fusion of Uncertain Information and
Modelling Trust in the Fused Information
ABSTRACT
The volumes of information streamed and collected from disparate sources have increased dramatically in recent years.
Fusing such multimodal information have increased the decision-maker’s ability to make informed decision in rapidly
changing complex environments which would not have been possible if the information were taken individually. While such
approaches have drawn a considerable attention from the community, the trustworthiness of the fused information stands as
a critical issue yet to be addressed.
Yrd. Doç. Dr.
Murat Şensoy
Computer Science
DEPARTMENT
Computer Science
CONTACT
[email protected]
FUNDING SCHEME
TÜBİTAK 3501
START DATE
01.10.2013
DURATION
36 months
OZU BUDGET
130,000.00 TL
2013 National Grants
The purpose of this project is to model trustworthiness of information sources, detect conflicts in information, and estimate
if the provided information is misleading. Then, this project will use these models, detected conflicts, and estimations
to calculate trustworthiness of the fused information. Furthermore, this project goes beyond calculating trust and will
propose mathematical approaches for trustworthy information fusion. In this context, the project is composed of 3 main
work packages: 1) Semantic representation of uncertain information, 2) Modeling trust in the fused information, and 3)
Trustworthy information fusion. The listed scientific contributions will make a significant impact on different areas of
research and applications where information from diverse and uncontrolled sources should be fused for decision-making.
On the Web, service consumers are overloaded with genuine and bogus information (e.g., ratings and reviews) about
potential service providers from different sources. In a forensic investigation, the Police collect possibly conflicting and
misleading reports from multiple witnesses about an incident. In military settings, information from controlled sensors (e.g.,
satellites, UAVs) and intelligence from uncontrolled sources (e.g., ally countries) should be carefully fused before making
critical decisions. Failures in fusion may lead to serious problems in decision-making, which may lead to disasters. These
are only a few examples where the outcomes of this project can contribute to the quality of the resulting fused information.
37