Topic: Decision Making Based on Dempster-Shafer Theory of Evidence
Speaker: Dr. Mohammad Abdullah-Al-Wadud
Abstract: The available training data for different classification problems are usually imprecise and incomplete, which leads to uncertainty in classifications as traditional probability-based classifiers requires complete knowledge of priors and conditional probabilities. This requires a robust fusion framework to combine available information sources with some degree of certainty. The Dempster–Shafer theory of evidence provides a method for combining evidences from different sources without prior knowledge of their distributions. In this method, it is possible to assign probability values to sets of possibilities rather than to single events only, and it is not needed to distribute all the probability values among the events, thus modeling more naturally certain classes of problems. Dempster’s rules for combination give a numerical procedure for fusing together multiple pieces of measurements from different (unreliable) observers. This talk addresses the employment of the Dempster-Shafer Theory of evidence in few practical applications.