Research On Ontology Based Dynamic Trust Model Among Software Entities

In recent years, due to the rapid development of information industry, information technology has brought unprecedented prosperity. But at the same time, against the information security incidents continue to occur, information security is faced with more and more challenges. Traditional solution is through encryption, decryption, or other technologies. These can make up for the system security features to a certain extent, but can't fundamentally solve the security problem. The hardware structure and the operating system's security is the basis of information system security, and only start from the underlying hardware and software, from a whole to take measures, we can effectively ensure the security of information systems. The adoption of trusted computing technologies, construction of trust root and trust chain can increase the credibility of the hardware platform. However, how to ensure the credibility of the network software inter-entity is still at its beginning phase.If the conduct of a software entity is always in line with expectations, it is credible. However, in the Internet environment, due to the openness, diversity and dynamic of resource itself, making the difficulty of choosing reliable service with high quality collaborative software entities increased. It is difficult to ensure the credibility of software entities. In addition, the trust relationship between software entities usually changes with the environment or changes with the interaction. It is a highly dynamic and evolutional relationship. Therefore, in the Internet environment, the credibility of software entities is faced with many new problems and challenges. This paper aim at three urgent problems, that is the complexity, diversity and dynamic of trust relationship between network software entities, mainly completed the following tasks:1. Trust Computing. Combined with the idea of trust graph, carry out a reasonable way to abstract and express the trust relationship between entities. Quoted the concept of ontology, discuss how to set the initial value of trust entities, how to integrate a variety of factors to calculate trustworthiness and how to improve the calculation method.2. Trust Evolution. Evolution is occurred by the stimulation of the interaction between entities (can be seen as a data stream). There will be a result after interaction between any two entities, such as success or fail. According to these results, we can learn. Through the deeply study of trust model evolutionary algorithm, improve the evolution method of trust model.3. At last, this paper do some experiments and compare the new model with typical models. The experimental results show that compared with existing models, the new model has higher degree of accuracy and security in the trustworthiness calculation of software entities.