Supercharge Your Innovation With Domain-Expert AI Agents!

Semi-supervised SQL (Structured Query Language) injection attack detection method based on few abnormal tags

A technology for injecting attack and detection methods, applied in the field of information security, can solve problems such as detection performance degradation, achieve stable detection performance, meet the needs of actual scenarios, and enhance model detection capabilities and generalization capabilities.

Active Publication Date: 2022-03-11
SICHUAN UNIV
View PDF6 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] Aiming at the above-mentioned deficiencies in the prior art, the present invention provides a semi-supervised SQL injection attack detection method based on a small number of abnormal labels, which solves the problem that the detection performance of the existing SQL injection detection method decreases when the data labels are unbalanced

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Semi-supervised SQL (Structured Query Language) injection attack detection method based on few abnormal tags
  • Semi-supervised SQL (Structured Query Language) injection attack detection method based on few abnormal tags
  • Semi-supervised SQL (Structured Query Language) injection attack detection method based on few abnormal tags

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0067] Such as figure 1 As shown, a semi-supervised SQL injection attack detection method based on a small number of abnormal labels includes the following steps:

[0068] S1. Construct an original data set according to collected normal SQL samples and abnormal SQL samples;

[0069] S2. Preprocessing the original data set to obtain an encoded SQL sample corresponding to each SQL sample in the original data set;

[0070] S3. According to the encoded SQL sample, construct and train a semi-supervised detection model based on the attention mechanism;

[0071] S4. The unknown SQL sample is detected through the semi-supervised detection model based on the attention mechanism, and the detection of the SQL injection attack is completed.

[0072] In this embodiment, in order to solve the problems that the traditional SQL injection attack detection framework adopts syntax analysis, word embedding code may hide key features of SQL injection and introduce noise, etc., the present invent...

Embodiment 2

[0087] The attention mechanism-based semi-supervised detection model generated in step S3 of Embodiment 1 includes a data distribution learning module and a detector module connected to each other.

[0088] In this example, if figure 2 As shown, the data distribution learning module uses a model based on data distribution learning to provide the detector module with high-level features such as the central tendency and degree of dispersion of sample features; and the detector module fuses the high-level features and sample encoding features provided by the data distribution learning module As an input, the feature extraction sub-module and the attention sub-module are used to effectively extract and focus on the features, and then judge whether the SQL sample contains SQL injection attacks.

[0089] In this embodiment, the data distribution learning module is implemented using the auxiliary deep generative model proposed in "Auxiliary deep generative models", which enhances th...

Embodiment 3

[0121] For the semi-supervised detection model based on the attention mechanism in Example 2, the experimental results of the ablation experiment are used.

[0122] The detection effect of the model is improved by introducing high-level information such as the central trend of the sample feature, the degree of dispersion, and the attention sub-module. In order to verify the influence of the two factors on the performance of the model, ablation experiments are carried out.

[0123] In order to evaluate the influence of introducing sample feature central tendency and discrete degree information on model performance, the method of the present invention is compared with the influence of introducing sample feature posterior distribution information on model performance. Experimental results such as image 3 As shown, ASDM is the method of the present invention, and ASDM-Z is the method of introducing the posterior distribution of sample features; using high-level features such as t...

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

No PUM Login to View More

Abstract

The invention discloses a semi-supervised SQL (Structured Query Language) injection attack detection method based on a small number of abnormal tags, and aims to solve the problem that the detection performance is reduced when data tags are unbalanced in the existing SQL injection detection method and meet the actual scene requirements. An SQL injection attack detection framework based on bit coding is provided, and the framework does not need to pre-train a word embedding model and grammar rule analysis; a semi-supervised SQL injection attack detection model based on an attention mechanism is provided based on the framework, and the model has excellent feature attention ability and generalization ability; and the method also has certain detection performance for novel SQL injection attacks.

Description

technical field [0001] The invention belongs to the technical field of information security, and in particular relates to a semi-supervised SQL injection attack detection method based on a small number of abnormal labels. Background technique [0002] SQL (Structured Query Language) is a general-purpose, highly functional relational database language, a standard interface for relational data access, and the basis for interoperability between different database systems. It integrates data query, data operation, data definition, and data control functions, and is used to interact between web application data and background database data. [0003] SQL injection (SQL injection) refers to inserting malicious SQL commands into Web forms to submit or input query strings for domain names or page requests, tricking the server into executing malicious SQL commands to achieve the purpose of stealing and destroying data. SQL injection attack has the characteristics of strong concealmen...

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
no application Login to View More
IPC IPC(8): G06F21/55G06F21/56G06K9/62
CPCG06F21/554G06F21/56G06F18/214G06F18/253Y02D10/00
Inventor 周颖杰赵伟杨松吕建成
Owner SICHUAN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com