The Trusted Measurement Method Of Software Process

With the rapid development of software technology, people are more likely to relay on the software. And the complexity of software is also spreading in an unprecedented scale. And it extends continuously. People began to pay more attention to the trustworthy of the software. Both domestic and foreign government, scientific research institutions, industrial enterprises have gradually realize the importance of software reliable research. And they have put forward the targeted research program. Among these programs, software trustworthy metric is a very important problemSoftware product is the artifact of the process; The quality of software products has a large extent depends on the software process. It is a useful method for the trustworthy software to use the standardization process methods at the very beginning of the software life cycle and using it throughout the course of the project. At present, it is rare to measure the software trusted attributes on the perspective of process. In this paper, a definition of software trusted process was given by the method of using fonnalization. And on the other hand, we get trusted evidence form the process and using this evidence to measure the creditability of the process. In this method, we may find a way to ensure the software to have a high quality.Based on these methods, the thesis solved the following questions:1. Defined the software trustworthy evidence, software trustworthy metrics and software measuring data. We obtain evidence from the evolution of instances2. Presents a hierarchical software process model. The author find that the trusted attributes is hierarchical when reading the papers. Besides the well-known software quality evaluation model is also a hierarchical model. Hierarchical model provides a foundation for the next stage of evaluation of the credibility of the software process.3. Lists three very important trustworthy attribute of software as a key trustworthy attributes. In the development process, through extracting evidence of software in real- time way and using the metrics of software process to give a qualitative or quantitative data of metrics of software process, an actual measurement model of a key trustworthy of process is constructed.4On the basis of the hierarchical measurement model of process, it does the consistency conversion for quantitative and qualitative measurement data of process. Especially for qualitative measurement data, because it is given by the expert then has a certain untrustworthiness, so the author uses the method of information entropy to evaluate the trustworthiness of data of experts, which prevents untrusted data to disrupt the trustworthiness of credible measurement of process.In summary, letting process as metric unit, this thesis builds a hierarchical measur ement model of process by extracting credible evidence and measures the trustworthin ess of software process by fuzzy hierarchical analysis. The goal is to find an approach to measure the software trustworthy process and foresee the quality of this software.