A trust model based on cloud model and Bayesian networks

The Internet has been becoming the most important infrastructure for distributed applications which are composed of online services. In such open and dynamic environment, service selection becomes a challenge. The approaches based on subjective trust models are more adaptive and efficient than traditional binary logic based approaches. Most well known trust models use probability or fuzzy set theory to hold randomness or fuzziness respectively. Only cloud model based models consider both aspects of uncertainty. Although cloud model is ideal for representing trust degrees, it is not efficient for context aware trust evaluation and dynamic updates. By contrast, Bayesian network as an uncertain reasoning tool is more efficient for dynamic trust evaluation. An uncertain trust model that combines cloud model and Bayesian network is proposed in this paper.