随着信息技术的发展,互联网数据急剧增长,导致网络服务的种类多种多样.由于网络服务内容多种多样,文本长度长短不一,使得传统分类方法难以有效解决大规模网页分类问题.文章设计并实现了一个面向海量网络服务的实时自动分类系统(AWCS),包括自反馈的系统体系结构、多维度多层次的网络服务分类标准、主动与被动方式相结合的网络服务发现采集技术、自修正的网络服务自动分类技术.性能测试显示,该系统的分类准确率明显高于已有算法.AWCS系统对构建实时大规模网络数据分类系统具有重要的参考和借鉴价值.%With the development of the Internet and communication technology, the Internet data growth rapidly, and the types of network services vary. Due to the variety of Web content and text length, the traditional classification methods can't effectively solve the problem of large-scale Web pages classification. This paper designs and implements a real-time automatic Web services classiifcation system (AWCS), including self-feedback system architecture, multi-dimensional network services classiifcation standard, active and passive combining network service discovery and collection technology, and automatic self-correction network service classiifcation techniques. Performance tests show that the classiifcation accuracy of AWCS is signiifcantly higher than the traditional algorithms. AWCS offers a promising approach for large-scale real-time network data classiifcation system.
展开▼
