Davranış tahmini ile öznel fikirleri tümleştirme

Information is significantly important in almost all decision-making process. A decision- maker agent collects information from diverse sources. Thus, it should correctly fuse opinions, which are shared from different information sources. However, some of infor- mation sources may be unreliable and malicious. That is, some of information sources may behave differently while sharing their opinions. Thus, the decision-maker agent needs to eliminate opinions that these opinions are collected from such kind of infor- mation sources. Motivated by this observation, in this thesis, we propose a statistical information fusion approach based on behavior estimation. In this approach, before estimation of fusion, we estimate behavior of information sources based on their sta- tistical values. Then, we enhance information fusion process based on our estimation for behavior of information sources. Through extensive simulations, we have shown that our approach has a low computational complexity, and achieves significantly low behavior estimation and fusion errors.