摘要
研究目的是挖掘搜索引擎中用户兴趣偏好,实现个性化搜索引擎技术。研究方法采用识别用户输入查询串,通过查询进行挖掘用户兴趣类别,但有时用户输入查询串短,或者出现查询词歧义等。由于查询会返回一系列文档,将相关文档分类处理,能够更清晰识别用户兴趣偏好。结果显示通过文档关系矩阵,将用户查询映射到对应类别,发现用户兴趣爱好。对于兼类查询等问题可以通过扩展查询解决。结论是该模型通过查询串和相关文档之间关系,进而实现用户偏好的辨别。该技术为搜索引擎信息推荐等技术打下良好基础。
The goal of the researches is digging user interest and realizing personalized search. The method of research is finding user query. Digging user class by query ,but sometime the query is short or query is ambiguity. Query will reaun some documents and then class the document finding user interest. The result shows query is mapped to the class by document matrix. Query expansion effectively solves the question of query short and ambiguity class. The result is that the input of the model is user query and the document,the output is user interest that provides the foundation for sorting technology.
出处
《计算机技术与发展》
2011年第11期153-155,共3页
Computer Technology and Development
基金
国家自然科学基金项目(60736044
60773070)