[Expert Viewpoint] Pay attention to the "hidden corner fall" in big data -the security of stealth data and their challenges

Author: Min Jinghua, CLP Great Wall Network System Application Co., Ltd.

Big data has the characteristics of large volume, many types, fast generations, and more changes. The value contained in it. With the rapid development of new generation of information technology such as cloud computing, artificial intelligence, and mobile Internet, it has been continuously excavated and given to humans to humans Life and work have brought tremendous changes. At the same time, the security issues of big data are accompanied by each other. The wider the application of big data, the more severe the security issues.

The value of big data is different from traditional data. The value of traditional data is generally presented directly on the data that can be visible within a limited range, such as transaction data, scientific and technological papers, etc., and the value presentation of big data requires a large range of continuous excavation, such as customer consumption preferences, hotspots of scientific and technological development, and hotspots of scientific and technological development and the development of scientific and technological development. Trends, etc. It can be seen that there are two forms of data in big data, one is explicit data and the other is invisible data. The so -called explicit data refers to the data that can be seen directly, that is, the data is in front of the eyes; the so -called invisible data refers to the data that needs to be seen after excavation (ie, analysis, association, and generation), that is, this is this kind of kind of Data did not know where it was at first. When needed, it was only to be tapped to make it explicit data. The data of these two forms are valuable, and they are facing various security issues, and they need to be protected. For explicit data, traditional data protection measures (such as encryption and access control) are valid, but it is difficult to play for stealth data, because I do n’t know where the data that needs to be protected, so I will not make all the data containing invisible data. Encryption or control of access to it, which not only protects the cost, but also hinders the reasonable circulation and application of the data, which is not worth it.

People usually pay attention to explicit data, but often ignore the stealth data, which can be described as a "hidden corner" in big data, and still lacks enough attention and attention. How to protect stealth data in big data? This is a new challenge that is different from traditional data security protection. Traditional data security protection objects are visible explicit data, and the protection measures are directly on the data that needs to be protected, but this protection measure is not suitable for invisible stealth data. Although the target of stealth data security protection is not appropriate to be directly invisible stealth data itself, you can consider the excavation process, methods and tools of invisible data, and take security control measures, that is, the data mining process and review data mining from the perspective of security perspectives, reviewing the data mining process and reviewing data mining Methods and control data mining tools to achieve the purpose of protecting stealth data. To this end, new ideas, methods and technical means are needed to discover hidden dangers of stealth data and prevent it. At the same time, in big data security education and training, people's consciousness and related knowledge and skills of hidden data security protection are increased.

The "Data Security Law of the People's Republic of China" stipulates that the data is classified and classified for data, formulated important data catalogs, and strengthened the protection of important data. Big data explicit or implicitly contains important data, which is manifested as an important data or an important data of an invisible data. Important data directory must not only list important data, but also depict invisible important data. Similarly, the protection of important data must not only protect visible and important data, but also protect invisible invisible important data. Risk management is the basic method of data security. Therefore, risk assessment must identify important data on dynamics, but also identify important data of invisible data, and provide a basis and implementation object for data security protection measures used in risk disposal.

The comparison of the explicit data in the table big data and the invisible data

The above table compares the explicit data in the big data and the invisible data.

In summary, stealth data security is a major security challenge brought by big data. Traditional data security protection methods and means are difficult to apply for it, and it needs to be re -recognized and continuously innovated. I believe that new generation of information technology such as big data, cloud computing, artificial intelligence, and mobile Internet is also an effective way to solve the security problems brought by itself.

(Author brief introduction: Min Jinghua, doctoral, professor -level senior engineer. Graduated from Tsinghua University, he has obtained a bachelor's degree and master's degree, and has taught at the Department of Computer Science and Technology of Tsinghua University. He studied in Japan and graduated from the University of Tokyo, Japan. He has worked at the Central Research Institute of Hitachi Production Institute as HIVIPS (Hitachi Senior Visit Scholars). After returning to China, he successively worked in Beijing Tsinghuaden Technology Co., Ltd. (later renamed Beijing Tongfang Information Security Technology Co., Ltd.) and served as senior researcher; Shanghai Three -Zero Guard Information Security Co., Ltd., which belongs to the 30th Research Institute of China Electronics Technology Group, is the chief security strategic consultant. Standardized chief. The research field is information technology. The research direction is information security, cloud computing, big data, smart cities, artificial intelligence, etc. The research expertise is the standardization of information technology.)

- END -