基于分类排名的网站可信度分析

A rankclass-based algorithm for the Web trust analysis

提出一种考虑网站分类排名的可信度分析算法,即在分析可信度时,通过引入训练集,对不同类别的信息进行分析,再利用分类信息来提高网站可信度分析的结果.实验结果表明,改进后的算法可以提高可信度分析结果的实用性和准确性.

We propose an algorithm based on the ranking of each category in the Web sites. It means that importing the training set, we can analyze the class information of the objects and then can use the information to improve the trust analysis. Results show that the trust analysis, with the help of the trust of different categories in Web sites, can be improved.