摘要
经典的基于链接结构的PageRank算法,它主要是依据页面之间的链接关系进行排序,容易出现主题漂移、忽视专业站点、偏重旧网页等缺点。针对这些问题,从超文本相关性、基于网站权威性权重因子和时间权重方面提出改进。实验结果表明,与传统的PageRank排序算法相比,改进算法能有效提高查准率,提高用户对排序结果的满意度。
Analyzes the classical algorithm on PageRank which is based on the existing link structure. The algorithm mostly works on interlinks a-mong Web pages and then presents some disadvantages of this algorithm. Those disadvantages are prone to theme-drift, ignoring special sites and preferring to old pages. Aiming at theses disadvantages, describes the improved algorithm. The experimental results show that, compared with the traditional PageRank ranking algorithm, the improved algorithm can both improve the retrieves accuracy ratio effec-tively and the satisfactory of the users.
出处
《现代计算机》
2014年第2期15-18,29,共5页
Modern Computer
关键词
搜索引擎
页面排序
链接结构
PageRank
Search Engine
Web Page Ranking
PageRank
Link Structure