Principles of subjective networks
Audun Jøsang, Lance Kaplan · International Conference on Information Fusion · 2016
This paper focuses on representation and reasoning in conditional inference networks, combined with trust networks, thereby introducing subjective networks as graph-based structures of variables combined with conditional opinions. Subjective networks generalize Bayesian networks from being based on probability calculus, to being based on subjective logic. In addition, subjective networks generalise Bayesian networks from assuming a global view of input arguments by a single analyst, to taking subjective view of input arguments by multiple agents who might have conflicting opinions. The result is a highly flexible and expressive framework for modelling and analysing realistic situations, which is also backwards compatible with traditional Bayesian networks.