Method, device and system for filtering network data

A technology of network data and filtering method, applied in the field of communication, can solve the problems of low processing efficiency, poor effect, large amount of calculation, etc., and achieve the effect of improving processing efficiency, improving accuracy and filtering effect

Active Publication Date: 2017-06-16
TENCENT TECH (SHENZHEN) CO LTD
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] During the research and practice of the prior art, the inventors of the present invention found that because the words used by cheaters are ever-changing, the same type of sensitive words often have many variants. High accuracy and effect, but only relying on manual annotation to identify image data such as pornographic images requires a large amount of calculation, resulting in low processing efficiency and poor results

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
  • Method, device and system for filtering network data
  • Method, device and system for filtering network data
  • Method, device and system for filtering network data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0027] This embodiment will be described from the perspective of an apparatus for filtering network data. The apparatus for filtering network data may specifically be integrated in a network device such as a server.

[0028] A method for filtering network data, comprising: acquiring a seed bank of a target information type and user behavior data, preliminarily filtering the seed bank according to the user behavior data and preset filtering rules to obtain a candidate set, and constructing a structure based on the candidate set transfer matrix, and initialize the user scoring vector and the page scoring vector, filter the information in the candidate set according to the transfer matrix, the user scoring vector and the page scoring vector to obtain target information; add the target information to the seed library.

[0029] like figure 1 As shown, the specific process of the method for filtering network data may be as follows:

[0030] 101. Obtain the seed bank of the target i...

Embodiment 2

[0099] According to the methods described in the embodiments, the following examples will be used for further detailed description.

[0100] In this embodiment, the device for filtering network data is specifically integrated in the server, and the target information type is specifically pornographic images and texts as an example for description.

[0101] like figure 2 As shown, a method for filtering network data, the specific process can be as follows:

[0102] 201. The server obtains a seed library of pornographic images and texts, and obtains user behavior data.

[0103] Wherein, the seed bank can be constructed in advance manually, for example, some pornographic messages, website addresses, and some pornographic pictures can be manually marked, and then added to the seed bank, and so on. The seed library may be stored in the server, or may be stored in other network devices.

[0104] 202. The server performs preliminary filtering on the seed database according to the...

Embodiment 3

[0148] In order to better implement the above method, an embodiment of the present invention also provides a device for filtering network data, such as image 3 As shown, the apparatus for filtering network data may include an acquiring unit 301, a first filtering unit 302, a constructing unit 303, a second filtering unit 304 and an adding unit 305, as follows:

[0149] The obtaining unit 301 is configured to obtain the seed library of the target information type and user behavior data.

[0150] Among them, the target information type can be set according to the needs of the actual application. For example, if you need to identify pornographic images and texts, you can set the target information type as pornographic images and texts, and if you need to identify some politically sensitive images and texts, You can set the target information type to politically sensitive graphics, and so on.

[0151] Initially, the seed bank can be manually constructed in advance. For example, ...

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 embodiment of the invention discloses a filtering method, device and system for network data. By means of a seed bank which can acquire target information types and user behavior data, preliminary filtration can be conducted on the seed bank according to the user behavior data and preset filtering rules, a candidate set is obtained, a transition matrix is structured based on the candidate set, a user grading vector and a page grading vector are initialized, information of the candidate set is filtered according to the transition matrix, the user grading vector and the page grading vector, target information is obtained, and the target information is added to the seed bank. By means of the scheme, the identification accuracy rate and filtering efficiency are greatly improved, and the processing efficiency is improved.

Description

technical field [0001] The present invention relates to the field of communication technologies, in particular to a method, device and system for filtering network data. Background technique [0002] With the development of Internet technology, the information on the Internet has become more and more abundant. While this information brings us convenience, it also brings some negative effects. For example, some pictures and texts about pornography or violence are full of them. The news will have a negative impact on the growth of young people and the stability of society. Therefore, how to filter out some negative information in this huge amount of information is an extremely important issue. [0003] In the prior art, some sensitive words are generally identified manually, and then corresponding words are filtered out from various network data through text mining and analysis, so as to achieve the purpose of identifying certain types of information. For image type data, it...

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
Patent Type & AuthorityPatents(China)
IPC IPC(8): G06F17/30
CPCG06F16/9535
Inventor梁龙军文勖
OwnerTENCENT TECH (SHENZHEN) CO LTD