本系统通技术运用主要采用Servlet实现，前台页面使用jsp技术，首先接受所要处理的数据集以及参数设置，对用户的搜索内容进行分词、去除停用词，形成用户的兴趣模型；进行用户兴趣模型匿名化：从用户兴趣模型集中选取模型，计算该模型与所有等价组模型的相似性以及计算该模型与其他所有模型的相似性，根据相似性的大小进行聚合成等价组，然后再计算各自兴趣模型的权值，以达到用户兴趣模型的匿名化以及权值的平衡，然后提交到后台进行匿名化操作，最后返回匿名化结果。-The system through the use of Servlet implementation front page using jsp technology, the first to accept a set of data to be processed, and parameter settings of a user' s search segmentation, removal of stop words, the formation of user interest model　user interest anonymization models: from the user interest model concentrated select a model to calculate the similarity of this model and all equivalents group model and to calculate the similarity of the model and all other model, based on the similarity of the size of the polymerization into an equivalent group, and then calculate the weights of the respective interest model, to achieve anonymous user interest model as well as the right balance of value, and then submitted to the background the anonymization operation, and finally return anonymized results.