FUSE BEE: Fusion of Subjective Opinions through Behavior Estimation

Information is critical in almost all decision makingprocesses. Therefore, it is important to get the right informationat the right time from the right sources. However, informationsources may behave differently while providing information i.e., they may provide unreliable, erroneous, noisy, or misleadinginformation deliberately or unintentionally. Motivated by thisobservation, in this paper, we propose a statistical informationfusion approach based on behavior estimation. Our approachtransforms the conveyed information into more useful form bytempering them with the estimated behaviors of sources. Throughextensive simulations, we have shown that our approach has alower computational complexity, and achieves significantly lowbehavior estimation and fusion errors. 18th International Conference on Information Fusion , 06 Jul 2015, 09 Jul 2015,