基于网络文本大数据的信息隐藏方法  

Information Hiding Method Based on Web Text and Big Data

作　　者：师尚伟 黄永峰[1] 王烨[2] 

机构地区：[1]清华大学电子工程系,北京100084 [2]中国移动通信集团公司网络部,北京100033

出　　处：《小型微型计算机系统》2017年第2期227-231,共5页Journal of Chinese Computer Systems

基　　金：教育部-中国移动科研基金项目(MCM20123041)资助;国家自然科学基金项目(61472092;61402115;U1405254;61271392)资助

摘　　要：文本作为语言的视觉形式是人类最重要的交流工具,基于文本的信息隐藏算法具有很高的实用价值.文本的一个明显特点是高度凝练,信息冗余少,因而文本隐藏的容量较低.另一方面,传统的文本隐藏建立在修改载体的基础上,而文本语义对于修改非常敏感,微小的修改可能引起明显异常,因而文本隐藏的隐蔽性较差.针对以上问题,深入分析了网络文本大数据的特点,据此设计了一种基于网络文本大数据的信息隐藏算法,把对载体的修改转化成对载体的检索,并通过位置信息定位秘密消息,从而不需要修改载体即可嵌入信息,隐蔽性大大提高.另外,实验结果显示本算法具有接近通用字符编码效率的嵌入率(18比特/字符),是一种高效的信息隐藏算法.As the visual form of language, text is the most important media of communication for human beings. Text based informa- tion hiding algorithm has high practical value in many fields. The hiding capacity of text is low because it is highly concise with less redundant information. On the other hand, traditional text information hiding is based on modifying carder text which is so sensitive to changes that even minor modifications can cause obvious abnormalities, making hiding easy to be detected. In this paper, we analyze the characteristics of Web text and big data, and design a novel information hiding algorithm based on it. Modification of carder is transformed into search of carrier,and position information can locate secret message. In this way,we can embed information without modifying the carder so that undetectability is greatly improved. Experimental results show that the embedding efficiency is close to general character encoding efficiency （18bits/character）, so this algorithm is an efficient information hiding method.

关 键 词：网络文本 大数据 信息隐藏 无修改隐藏 文本隐藏 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]