基于相关性分析的网页学术性算法研究  

Research on Webpage Academic Algorithm Based on Correlation Analysis

作　　者：赵冰漫 王卫亚[1] Zhao Bingman;Wang Weiya(CHANG ANUNIVERSITY,Xi＇an Shannxi,710064)

机构地区：[1]长安大学,陕西西安710064

出　　处：《电子测试》2018年第22期70-71,共2页Electronic Test

摘　　要：本文以学术网页的识别与检索为目标,调查分析学术网页的网页特征。并以非学术文献网页作为参照,对网页特征抽取,验证所发现特征的可靠性。研究结果显示,学术文献网页在关键词词频、关键词权重和关键词相关度等特征方面与非学术文献网页具有明显差别,差异程度明显。能较好地用于区分学术文献网页与非学术文献网页,为今后系统开发学术文献网页的自动化识别工具提供了依据和理论支持。This article aims at the identification and retrieval of academic web pages, investigates and analyzes the characteristics of web pages of academic web pages. The non-academic literature webpage is used as a reference to extract the features of the webpage and verify the reliability of the discovered features. The research results show that the academic literature webpage has obvious differences with the non-academic literature webpage in terms of keyword frequency, keyword weight and keyword relevance, and the degree of difference is obvious. It can be used to distinguish between academic literature pages and non-academic literature pages, and provides a basis and theoretical support for the automatic identification tools for systematic development of academic literature pages in the future.

关 键 词：学术网页 网页特征 特征抽取 

分 类 号：TP393.092[自动化与计算机技术—计算机应用技术]