Big data platform unknown threat detection method based on deep transfer learning
A technology of big data platform and transfer learning, which is applied in the field of unknown threat detection on big data platform, can solve the problems of threat small sample data sets, etc., achieve the effect of improving prediction effect, improving generalization ability and threat detection ability
Active Publication Date: 2020-04-10
CHINA ELECTRONICS TECH CYBER SECURITY CO LTD
View PDF10 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] 1) How to detect unknown threats against big data platforms and solve the problem of small sample data sets of threats on big data platforms;
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0022] 1. Implementation plan
[0023] The solution proposed by the present invention realizes a threat detection framework based on deep migration learning, which mainly includes data collection and processing, migration learning, and deep learning threat detection. Through different calling methods, a complete threat detection system for big data platforms based on deep migration learning is formed. figure 1 A schematic diagram of a threat detection scheme based on deep migration learning is given. The following is based on figure 1 Explain how it works.
[0024] Such as figure 1 As shown, the core content of the present invention is to obtain corresponding data from the execution of other malicious programs, construct a large threat sample set in the source field, and use the rich supervision information of the sample set to help the training of the deep learning model in the target field.
[0025] In terms of data collection and processing, information collection probe devices a...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a big data platform unknown threat detection method based on deep transfer learning. The method comprises the following steps: step 1, constructing a source domain sample set;step 2, collecting sample data of a target domain by adopting the same method as the step 1, then expanding the sample data by adopting a data enhancement method, and constructing a target domain sample set; and step 3, constructing a threat detection model based on deep transfer learning. Compared with the prior art, the method has the positive effects that 1, the problem of insufficient generalization ability of the deep learning model is solved through data enhancement in the target field, so that the prediction effect of the deep learning model is improved; and 2, through transfer learningfor massive threat samples of the Internet, unknown threats which do not appear are effectively detected on the premise of not reducing the known threat detection rate; and 3, the behavior characteristics of different dimensions are fused through the deep neural network fused with the characteristics, so that the recognition accuracy of the model is improved.
Description
Technical field [0001] The invention relates to a method for detecting unknown threats on a big data platform based on deep migration learning. Background technique [0002] With the increasing development and popularization of Internet technology and mobile communication technology, big data platforms are also facing network threats and data security issues. The amount of information exchange between big data platforms and users has greatly increased, and data security and risk prevention in the field of big data platforms are more complex than traditional networks. Especially for big data platforms integrated with mobile business systems, such as Hadoop, currently lack security protection means, or adopt insufficient security defenses, face security threats such as data theft, data integrity, and identity forgery, and have fewer threat samples. Aiming at the shortcomings of existing big data platform threat detection methods, this paper proposes an unknown threat detection met...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): H04L29/06G06N3/04G06N3/08
CPCH04L63/1416H04L63/1425H04L63/145G06N3/08G06N3/045
Inventor 孙治周玉金刘正军李春林陈剑锋徐锐饶志宏
Owner CHINA ELECTRONICS TECH CYBER SECURITY CO LTD