摘  要：一個普通的網頁可以被分成正文和噪聲兩個部分，噪聲影響了對網頁進行正文提取、聚類等處理，因此快速準確地去除網頁中的噪聲是網頁信息處理的關鍵技術之一。本文根據網頁文本格式屬性的相似性，提出一種基于網頁標簽屬性的去噪算法，并將通過此算法處理的網頁用于K-MEANS聚類算法。實驗結果表明本文提出的去噪算法是有效的，并且聚類結果的準確性有了較好的改進。

關鍵詞: 網頁去噪；標簽屬性；DOM樹；極大相容類

An Approach for Noise Reduction in Web Pages Based on Tag Attributes

JIANG Kun¹  YANG Yue-xiang²  FANG Hong²

(1. School of Computer Science, National University of Defense Technology, Changsha, 410073;

2. Information Center, National University of Defense Technology, Changsha, 410073)

Abstract: A common web page could be separated into two categories:valuable segments and noise segments, which affects web page extraction, clustering and other processing, so eliminating noise accurately and efficiently is a key technique in web disposal. According to the similarity of web page's text format, we present a new approach of noise reduction algorithm based on the tag attributes of web page, and apply this algorithm to the K-MEANS clustering  experiment. The experimental results show that the proposed algorithm is effective, and the accuracy of clustering has been improved.

Key words: Noise Reduction; Tag Attributes; DOM Tree; Maximal Compatible Classes

 

作者簡介：姜琨（1984-），男，碩士研究生，主要研究方向為計算機網絡與安全；楊岳湘，教授、博士生導師；方宏，高級工程師。