基于KNN算法与局部回归的网站无障碍采样评估  

Website accessibility sampling evaluation based on KNN and local regression

作　　者：陈荣华[1] 王鹰汉[2] 卜佳俊[3] 于智[3] 高斐 CHEN Rong-hua;WANG Ying-han;BU Jia-jun;YU Zhi;GAO Fei(College of Information Engineering,Jiangxi Vocational College of Finance&Economics,Jiujiang 332000,China;Department of Information Engineering Shangrao Vocational&Technical College,Shangrao 334109,China;Zhejiang Provincial Key Laboratory of Service Robot,College of Computer Science,Zhejiang University, Hangzhou 310027,China;College of Information Engineering,Putian University,Putian 351100,China)

机构地区：[1]江西财经职业学院信息工程学院,江西九江332000 [2]上饶职业技术学院信息工程系,江西上饶334109 [3]浙江省服务机器人重点实验室,浙江大学计算机学院,浙江杭州310027 [4]莆田学院信息工程学院,福建莆田351100

出　　处：《浙江大学学报（工学版）》2018年第9期1702-1708,共7页Journal of Zhejiang University：Engineering Science

基　　金：国家科技支撑计划资助项目(2014BAK15B02)；国家自然科学基金资助项目(61173185，61173186)；江西省教育厅科技资助项目(GJJ161399)；浙江省自然科学基金资助项目(LZ13F020001)；江西省高校人文社会科学研究资助项目(SH17203).

摘　　要：提出一个新的抽样评估方法,通过对评估样本的KNN分析,选出特定网页.因大规模网站抽样结果稀疏,KNN算法会导致高检测误差,应用一个局部回归模型提升KNN评估质量.首先在网站中随机选择一些网页进行评估,得到该网站初始无障碍得分.在此基础上,将每一个评估网页作为一个标记样例,其他网页根据KNN局部回归模型进行无障碍评估得分预测.实验结果证明:所提方法相比随机抽样算法的效果上有着显著性提升.A novel sampling evaluation algorithm was proposed for a given page based on the KNN evaluated samples.As sampling in a large website tends to be sparse,KNN may lead to a high evaluation bias and a local regression model was thus employed to improve the quality of KNN-based evaluation.First,a certain number of webpages were randomly selected from a website and evaluated to obtain an initial website accessibility score.Each evaluated webpage was treated as a labeled sample and the accessibility scores for the rest pages in the website were estimated using local regression on the KNN.The experimental results validate that the proposed algorithm has significant improvement over the random sampling algorithm in website accessibility evaluation.

关 键 词：网站无障碍 自动检测 人工检测 局部回归 随机抽样 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]