A classification evaluation model for software trustworthiness based on trustworthiness evolution

Due to the outstanding of the problems related to software trustworthiness, software trustworthiness evaluation has become a research focus area in recent years. In this paper, based on the analysis of the intension of software trustworthiness, we analyze the complexity and the dynamic evolutionary process of software trustworthiness, and propose a classification evaluation model for software trustworthiness which reproduces the software trustworthiness's evolutionary process to evaluate software's trustworthiness level. The model defines the software trustworthiness attribute model and the classification levels, analyses the intrinsic and extrinsic evolutionary relations between users' trustworthiness expectation, trustworthiness attribute, trustworthiness index, trustworthiness evidence and trustworthiness level, and establishes an effective reference model for finding software trustworthiness index, the trustworthiness evidence, and their relationships. Thus, the classification evaluation model for software trustworthiness provides an effective and convenient method for software trustworthiness evaluation. As a spin-off, a multi-dimensional and multi-scale trustworthiness evaluation index system for software of the same type is established in a XML database so as to guide trustworthiness evaluation for other software of the same type.