国家教育考试网上有害信息自动监测模型研究

Research on the Model of Internet Harmful Information Auto-detection about National Education Examinations

随着信息技术的飞速发展,互联网已成为国家教育考试有害信息传播的主要途径。为净化涉考网络环境,保障教育考试的公平和安全,在国家教育考试期间,各级考试机构安排专人利用百度、搜狗等搜索引擎,人工搜索有害信息,并上报有关部门进行处置。然而该方式存在工作效率低、搜索范围小、信息分析程度低等问题。针对这些问题,本文利用主题搜索、文本处理等信息技术手段,提出了以领域知识库为核心的有害信息自动监测模型,自动对互联网信息进行采集、去重、分类等处理,实时提供分类统计、热点分析、来源分析等基础数据。最后给出该模型与现有人工监测的协作方式,两者之间相互补充,从而形成覆盖面大、实时性强、精确度高的有害信息监控体系。

With the rapid development of information technology, the Internet has become the main way of the harmful information on National Education Examinations. In order to ensure fair and safe examination, the examination staff search harmful information using search engines, such as Baidu and Sogou. The search results will be submitted to the relevant departments for disposal. However, this method has some problems, such as low efficiency, small searching scope and low information analysis. In order to solve these problems, this paper presents an Internet harmful information auto-detecting model based on domain knowledge. The model can automatically collect information of the Internet, de- duplicate, classify, and provide the basis data for classification, hot spot analysis and source analysis. Finally, this paper presents the model of cooperation with the existing manual monitoring to form a harmful information monitoring system with large coverage, real-time performance and high accuracy.