Shulou(Shulou.com)06/03 Report--

Under the background of big data's surging attack, people's attention has gradually shifted from how to grasp the huge data information to how to realize the deep mining of these data, and then make them value-added. Through the integration of a large amount of information and the analysis of massive data, enterprises can have a deeper understanding of their own business, achieve new needs insight, and better make business decisions.

Here the editor has built a big data learning exchange QQ group: 529867072, I have sorted out the latest big data advanced materials and advanced development tutorials myself. If you want to do anything, you can join the group to study and communicate together.

Picture description

For enterprises, information security serves for informatization, and informatization serves business growth. Therefore, the use of big data to improve the level of enterprise information security protection can indirectly bring benefits to enterprises. Big data's application in the field of information security will evolve into a part of the development trend of IT business intelligence, that is, the combination of security data and business data can provide enterprises with a more reliable strategy basis, help enterprises to judge a variety of potential threats and predict business development trends.

This paper introduces the related technologies around the application of big data in four aspects: security detection, data mining, network awareness and video surveillance, including the integration of security detection and big data, the mining technology for security big data, the network situation awareness technology based on big data, and the mining technology of video surveillance data. Strive to make everyone understand big data's application in the field of information security.

The Fusion of Safety Inspection and big data

The integration of security detection and big data can detect potential threats in time, provide security analysis and trend prediction, and strengthen the ability to deal with threats. First of all, it is necessary to classify, filter and filter the data, then use information security detection technology to detect the system environment and data environment, and then build a security threat model through association analysis and data mining. the security trend is predicted through data analysis.

1. Data extraction and processing

After collecting the original massive data, it is necessary to classify, filter, filter and process the data. The data should be preprocessed according to the sensitivity of the data, the scope of influence, the application scenario, and the data requirements of business cooperation. Then filter sensitive data, confidential data and illegal data to get valuable and usable data. The performance requirements of data filtering and filtering include massive processing scale, multi-field filtering, intelligent filtering, efficient filtering and so on. Therefore, the intelligent filtering and content audit for big data can quickly and easily match a large number of custom keywords and words, intelligently filter the content that violates national laws and regulations and infringe upon the rights and interests of users, and ensure the security of the information content.

(1) Intelligent matching of keywords and words

Big data's intelligent matching requires that after users enter keywords and words, the system can automatically match and calculate the relevant information about the word. At the same time, it ensures that there is no limit to the number of user-defined keywords and can support millions of keywords concurrently. According to the keyword matching methods defined by Google, Baidu and other companies, big data intelligent matching roughly includes extensive matching, phrase matching, accurate matching, multi-keyword matching and so on.

● extensive matching can match any order of keywords, trigger queries that contain other words, and trigger similar words (including plural and synonyms).

● phrase matching is triggered only if the search term exactly matches the phrase order, and allows other search terms to exist around the phrase.

● precise matching will only trigger the keywords that match perfectly, limit the trigger range and reduce the number of triggers.

● multi-keyword matching is a technology that quickly matches multiple keywords (multiple patterns) from a large amount of data, which requires pre-processing of the text.

(2) Intelligent logical relation operation

Big data's intelligent logic relation operation should support the complex matching of keywords and words, including the commonly used "and, OR, no", as well as the complex algorithms of "NEAR" proximity relations. According to the user-defined category system, sort out the filtered information content, according to the characteristics of their own business, customize the content filtering system, and output the custom categories.

(3) sample machine learning

When there are no keywords, the machine automatically learns the skills to achieve the similar classification of information text. The information is provided by big data environment, and the learning part realizes the information conversion, memorizes it in a form that can be understood, and obtains useful information from it. The learning process is closely related to the reasoning process. According to the number of reasoning used in learning, the strategies used in machine learning can be divided into four types: mechanical learning, imparting learning, analogical learning and case-based learning. The more reasoning is used in learning, the stronger the ability of the system is.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.