Unlock instant, AI-driven research and patent intelligence for your innovation.

Content risk control method and device

A risk control and risk technology, applied in the security field, can solve the problems of low efficiency, single content detection method, easy to be missed detection within the risk, etc., to achieve the effect of improving the accuracy and recall rate

Pending Publication Date: 2021-07-02
BEIJING WODONG TIANJUN INFORMATION TECH CO LTD +1
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The current content detection method is relatively simple, such as through image blacklist detection, sensitive lexicon detection and other detection methods
[0004] In the process of implementing this application, the inventor found that when using the above content risk detection method to detect risky content and perform message interception processing, a large number of risky content is easily missed, and the efficiency is low

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
  • Content risk control method and device
  • Content risk control method and device
  • Content risk control method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0049] In this embodiment, real-time control is performed on content risks.

[0050] see figure 1 , figure 1 It is a schematic diagram of the content risk control process in Embodiment 1 of this application. The specific steps are:

[0051] Step 101, monitor messages between devices.

[0052] The monitoring of messages between devices here will monitor all messages, but due to the timeliness of processing, all messages cannot be processed in real time, and only the messages that can be processed in real time can be processed here.

[0053] Messages between devices, including messages between client and client, messages between client and server, messages between server and server, such as instant messaging messages between buyers and sellers in e-commerce .

[0054] Step 102, based on a preset recognition model and a preset sample library, perform content recognition on the monitored messages respectively.

[0055] For a message, content recognition is performed based on...

Embodiment 2

[0068] In the embodiment of the present application, quasi-real-time processing is performed on the pictures sent by users with risky behaviors analyzed in the first embodiment.

[0069] see figure 2 , figure 2 It is a schematic diagram of the content risk control process in Embodiment 2 of this application. The specific steps are:

[0070] Step 201, periodically acquire pictures in messages with risky content and pictures sent by users with risky behaviors.

[0071] In the first embodiment, the identified messages with risky content and users with risky behaviors are marked.

[0072] In this embodiment, identification processing is performed on pictures in messages marked as having risky content and pictures sent by users with risky behaviors.

[0073] Here, the period may be N hours, etc., and there is no limitation on this, and the length of the period interval is determined according to the needs of actual scenarios.

[0074] Step 202, perform optical character dete...

Embodiment 3

[0083] In this embodiment, messages between devices are periodically acquired, and parallel quasi-real-time policy processing is performed.

[0084] see image 3 , image 3 It is a schematic diagram of the content risk control process in Embodiment 3 of this application. The specific steps are:

[0085] Step 301, periodically obtain messages between devices that are determined to be other than risk messages, and determine risk messages in the messages based on a preset classification model.

[0086] Wherein, the risk message includes a message determining that there is a risky content, and a message sent by a user with a risky behavior.

[0087] The cycle in this embodiment can be the same as or different from the cycle in embodiment two. The time length corresponding to the cycle in embodiment two and embodiment three can be greater than the time length of the preset time in embodiment one. Limit.

[0088] Step 302: Analyze the behavior characteristics of the messages se...

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 provides a content risk control method and device. The method comprises the following steps: monitoring messages between devices; based on a preset identification model and a preset sample library, performing content identification on the monitored message; if it is identified that risk content exists in the message based on the preset identification model and / or the preset sample library, intercepting the message; acquiring a message sent by a user who sends the message with the risk content within a preset time; carrying out behavior characteristic analysis on the obtained message based on a preset analysis model; and if the existence of the risk behavior is determined through analysis, intercepting the message sent by the user with the risk behavior. The method can improve the accuracy and recall rate of risk detection.

Description

technical field [0001] The present invention relates to the field of security technology, in particular to a content risk control method and device. Background technique [0002] Content risk detection is to use computer natural language processing technology and visual technology to judge whether there are risks such as pornography and advertisement drainage in text or pictures. [0003] The current content detection methods are relatively simple, such as through image blacklist detection, sensitive lexicon detection and other detection methods. [0004] In the process of implementing the present application, the inventor found that when using the above-mentioned content risk detection method to detect risky content and perform message interception processing, a large amount of risky content is likely to be missed, and the efficiency is low. Contents of the invention [0005] In view of this, the present application provides a content risk control method and device, whic...

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): G06F16/33G06K9/00G06K9/62G06Q10/06
CPCG06F16/3344G06Q10/0635G06V30/40G06F18/214
Inventor 康思源吴磊
Owner BEIJING WODONG TIANJUN INFORMATION TECH CO LTD