A Multi-Dimension Dynamic Trust Evaluation Model Based on GA

Abstract

As entities in distributed environments lack sufficient global information, it is difficult to evaluate trust accurately. By considering characteristics of resources in distributed environments such as openness, dynamicity and uncertainty, we proposed a new dynamic trust evaluation model in this paper. The model describes factors that can affect trust evaluation based on a multi-dimensional view that include static attributes of entities, time-based behavior speculation, context awareness and communication conditions. Genetic algorithm (GA) is then introduced to do trust computation in the proposed model, which is important due to the lack of mathematical tools in traditional trust models for multi-dimensional data processing. Simulation results show that our proposed model is better compared to other trust evaluation models.