Evaluation And Implementation Of Component-based User Feedback

Along with the increasing maturity of component-based software development technology, the various types of software component development tools are becoming increasingly popular. So the ordinary crowd of who produce ordinary software and component has been expanded from the traditional developers with professional knowledge to the general population who just familiar with the development tools and have a certain programming ability. The developments of network technology and network resource sharing have made the quantity and species of components completely meet all kinds of users with different needs. Users can order components from component manufacturers, but users tend to find the necessary components from major online component library. With the expansion of the number of components, the following problem is the uneven quality of the components,In addition, the requirements of component from the modern user are not only focus on the complete functions, high efficiency or other rational rigid quality, but also the component’s emotional experience. Therefore, it is necessary to solve the problem about how to evaluate and select a component from the massive components to meet user needs.Most of the existing components evaluation researches are aimed to single component. The evaluation index based on component is built by referencing existing software quality model, which mainly assessed the quality of the objective and rational aspects of the components. Thus the subjective feedback of the components’ actual situation is lacking. In this dissertation, firstly, the similarities and differences between software attributes and component attributes are analyzed deeply. Secondly, components evaluation index system based on some mainstream software component quality model is established. Thirdly, a secondary user evaluation index system is established by analyzing users who participate in the component evaluation. Then, samples of expert feedback value and user feedback value are collected. A variety of subjective and objective weighting algorithm is used to determine the component index weights and used the relationship between users to determine the user weigh. At last, user feedback value is combined to get the evaluation results of the components. This method established a more complete evaluation index system, combining subjective and objective weighting methods, considering the relationship between users which making the components evaluation more reasonable. Finally a component evaluation tool is designed and implemented, and the model and evaluation method’s reasonability is verified through a lot of experiments.