摘要
对多媒体数字网络恶意信息的分层识别,能够有效保障多媒体数字网络安全稳定的运行。对恶意信息的分层优化识别,需要将网络文档中特征词集合进行切分,对数字网络中发音相似的字符串进行匹配,完成信息分层识别。传统方法提取网络信息的细粒度特征,对网络信息的分散度和个数进行分析,但忽略了对网络发音相似字符的匹配,导致信息识别精度偏低。提出多媒体数字网络恶意信息分层识别方法,采用切分标志法对多媒体数字网络中的文档进行切词处理,将多媒体数字网络中的句子切分成短小的字段,最终将多媒体数字网络中的文档切分特征词的集合;对采集到的多媒体数字网络中的信息进行分析,对多媒体数字网络中发音相似的字符串进行匹配,完成多媒体数字网络中恶意信息的分层识别。仿真结果表明,所提方法对多媒体数字网络中的恶意信息进行识别时,有效性高、识别效率好。
The hierarchical recognition of malicious information in multimedia digital network can effectively en- sure the security and stability of multimedia digital network. In traditional methods, the fine - grained feature of net- work information was extracted, and the degree of dispersion and number of network information was analyzed, but the matching of similar characters of network pronunciation was ignored, resulting in low accuracy of information recognition. This article focuses on hierarchical recognition method for malicious information in muhimedia digital network. This research used the method of segmentation sign to segment words of document in muhimedia digital network, and segment sentences into short fields. Finally, this method segmented documents into the set of characteristic words. In addition, the research analyzed collected information and matched alphabetic string with similar pronunciation. Thus, we completed the hierarchical identification of malicious information in multimedia digital network. Simulation results show that the proposed method has high effectiveness and good recognition efficiency in identifying the malicious information in multimedia digital network.
作者
石春爽
SHI Chun - shuang(College of Art and Design, Dalian Polytechnic University,Dalian Liaoning 116034, China)
出处
《计算机仿真》
北大核心
2018年第6期207-210,214,共5页
Computer Simulation
基金
辽宁省社会科学界联合会课题(2014lslhzkt-18)
关键词
多媒体数字网络
恶意信息
分层识别
Multimedia digital network
Malicious information
Hierarchical recognition