CN118227871A - Method and robot for controlling network content based on artificial intelligence of user type - Google Patents
Method and robot for controlling network content based on artificial intelligence of user type Download PDFInfo
- Publication number
- CN118227871A CN118227871A CN202410026645.6A CN202410026645A CN118227871A CN 118227871 A CN118227871 A CN 118227871A CN 202410026645 A CN202410026645 A CN 202410026645A CN 118227871 A CN118227871 A CN 118227871A
- Authority
- CN
- China
- Prior art keywords
- network content
- information
- identified
- content
- network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 26
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000012545 processing Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000012549 training Methods 0.000 description 24
- 238000012360 testing method Methods 0.000 description 20
- 230000000694 effects Effects 0.000 description 11
- 230000006399 behavior Effects 0.000 description 9
- 238000013136 deep learning model Methods 0.000 description 8
- 238000012552 review Methods 0.000 description 3
- 235000013601 eggs Nutrition 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005067 remediation Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The method and the robot for managing the network content based on the artificial intelligence of the user type utilize the artificial intelligence model to learn the internal association between the network content and the user type and the producer type of the network content, then predict the bad content in the network content according to the user type and the producer type, can accelerate the detection speed of the bad content, and can improve the accuracy of predicting the bad content detection.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a method and a robot for controlling network content based on artificial intelligence of user types.
Background
The review of web content is often based on the content itself.
In the process of implementing the present invention, the inventor finds that at least the following problems exist in the prior art: the prior art only focuses on the network content itself, but not on the type of user, when detecting the network content.
Accordingly, the prior art is still in need of improvement and development.
Disclosure of Invention
Based on the above, it is necessary to provide a method and a robot for controlling network content based on artificial intelligence of user type, which can rapidly screen and detect bad content in the network content according to user type, thereby improving speed and efficiency of bad content detection.
In a first aspect, an embodiment of the present invention provides an artificial intelligence method, the method including:
The network content type identification first model use step: if the information of the user watching the network content to be identified can be obtained, the information of the user watching the network content to be identified is taken as input, and the output obtained through calculation of the network content type identification first model is taken as the type label of the network content to be identified.
Preferably, the method further comprises:
The network content type identification second model use step: if the information of the user watching the network content to be identified and the information of the platform of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification second model is taken as the type label of the network content to be identified.
Preferably, the method further comprises:
The network content type identification third model using step: if the information of the user watching the network content to be identified and the information of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification third model is taken as the type label of the network content to be identified.
Preferably, the method further comprises:
And a bad content treatment step: if the type label of the network content to be identified is unhealthy content, notifying a user that if the type label of the network content to be identified is unhealthy content, timely processing is needed.
In a second aspect, embodiments of the present invention provide an artificial intelligence system, the system comprising:
The network content type identification first model use module: if the information of the user watching the network content to be identified can be obtained, the information of the user watching the network content to be identified is taken as input, and the output obtained through calculation of the network content type identification first model is taken as the type label of the network content to be identified.
Preferably, the system further comprises:
The network content type identification second model use module: if the information of the user watching the network content to be identified and the information of the platform of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification second model is taken as the type label of the network content to be identified.
Preferably, the system further comprises:
The network content type identification third model use module: if the information of the user watching the network content to be identified and the information of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification third model is taken as the type label of the network content to be identified.
Preferably, the system further comprises:
Bad content management module: if the type label of the network content to be identified is unhealthy content, notifying a user that if the type label of the network content to be identified is unhealthy content, timely processing is needed.
In a third aspect, embodiments of the present invention provide an artificial intelligence device comprising a module of a system according to any of the embodiments of the second aspect.
In a fourth aspect, embodiments of the present invention provide a computer readable storage medium having stored thereon a computer program which when executed by a processor implements the steps of the method according to any of the embodiments of the first aspect.
In a fifth aspect, an embodiment of the present invention provides a robot, including a memory, a processor, and an artificial intelligence robot program stored on the memory and executable on the processor, the processor implementing the steps of the method according to any one of the embodiments of the first aspect when the processor executes the program.
According to the method and the robot for controlling the network content based on the artificial intelligence of the user type, the network content is learned by the artificial intelligence module, the internal association between the network content and the user type and the producer type is utilized, then the bad content in the network content is predicted according to the user type and the producer type, the detection speed of the bad content can be accelerated, and meanwhile the accuracy of predicting the bad content detection can be improved.
Drawings
FIG. 1 is a flow chart of an artificial intelligence abatement method provided by an embodiment of the invention;
FIG. 2 is a block diagram of an artificial intelligence abatement system in accordance with one embodiment of the invention.
Detailed Description
The following describes the technical scheme in the embodiment of the present invention in detail in connection with the implementation mode of the present invention.
Basic embodiment of the invention
In a first aspect, as shown in FIG. 1, an embodiment of the present invention provides an artificial intelligence method including a network content type identification first model use step. In a preferred embodiment, the method further comprises a web content type identification second model use step. In a preferred embodiment, the method further comprises a third model use step of web content type identification. In a preferred embodiment, the method further comprises a bad content remediation step.
In a second aspect, as shown in FIG. 2, an embodiment of the present invention provides an artificial intelligence system that includes a network content type identification first model use module. In a preferred embodiment, the system further comprises a web content type identification second model use module. In a preferred embodiment, the system further comprises a web content type identification third model use module. In a preferred embodiment, the system further comprises a bad content governance module.
PREFERRED EMBODIMENTS OF THE PRESENT INVENTION
Where there are bad eggs, where there are flies, whereas where there are flies, where there are bad eggs. For example, if no user exists in the illegal network audio/video, the network audio/video has no living space.
And tracing out the users watching the network content to establish the correlation between the users and the network content, and then rapidly predicting the types of the network content according to the types and the number of the users to perform preliminary screening so as to perform large-scale census on the network content. The type of web content affects the type of users, e.g. puberty and male users may be more concerned with yellow content, of course how many users are watching, which types of users are watching, also depends on the type of platform, e.g. more viewers of the platform with a large traffic, and also on the information of the web content itself, e.g. the duration of the release, the duration of the web content, the title of the web content.
The technical effects are as follows: different users like to watch different network contents, so that the type of the network contents can be predicted according to the users, and the user information is very simple compared with the network contents, particularly audio and video, so that the network contents are predicted based on the user information, and the bad contents can be rapidly screened out compared with the network contents.
A type 1 tag set acquisition step: and obtaining a type tag set of the network content, wherein the type tag set comprises healthy content, bad content, illegal content and the like, and the bad content comprises yellow content, low custom content, violent content and the like.
2, Acquiring a training test data set of the network content type identification model: acquiring type tags of network contents (the network contents comprise characters, audio, video and the like) for training and testing, and information of users watching the network contents (if a plurality of users watching the network contents exist, the information of the users comprises information of a plurality of users; the information of the users who view the web content includes a type of each user who views the web content, the type of each user who views the web content includes a sex, an age, a occupation, an academy, etc. of each user who views the web content, the type of each user who views the web content includes a sex, an age, an occupation, an academy, etc. of each user who views the web content, a behavior of the user who views the web content (if there are a plurality of users who view the web content, the information of the user includes information of a plurality of users; the behavior of the user watching the network content comprises behavior information of each user watching the network content, wherein the behavior information comprises browsing time length, forwarding times, collection or not, and the like), information of a platform where the network content is located (the information of the platform comprises registered user quantity, access user quantity and the like of the platform where the network content is located), information of the network content (the information of the network content comprises time length from release to date of the network content, network content capacity (comprising word number of words, time length of audio and video) and title of the network content).
3, A first model training testing step of network content type identification: initializing a deep learning model as a network content type identification model; and taking information of a user watching the network content as input, taking the type label of the network content as expected output, and training and testing a network content type identification model. The technical effects are as follows: only the information of the user is needed, the requirement on the acquired data is low, and the data can be used under the condition of only the information of the user.
4, A second model training testing step of network content type identification: initializing a deep learning model as a network content type identification model; and training and testing a network content type identification model by taking information of a user watching the network content and information of a platform where the network content is located as input and taking a type label of the network content as expected output. The technical effects are as follows: the information of the user is considered, and the information affecting the platform where the network content of the user is located is also considered to be taken as the basis for predicting the type of the network content, so that the accuracy of prediction can be further improved.
After the step 3 of the process, the process is carried out,
5, A third model training test step of network content type identification: initializing a deep learning model as a network content type identification model; and training and testing a network content type identification model by taking information of a user watching the network content and information of the network content as input and taking a type label of the network content as expected output. The technical effects are as follows: the information of the user is considered, and the information affecting the network content of the user is also considered to be taken as the basis for predicting the network content type, so that the accuracy of prediction can be further improved.
6, Identifying the network content type and testing the fourth model training: initializing a deep learning model as a network content type identification model; and taking information of a user watching the network content, information of a platform where the network content is located and information of the network content as input, taking a type label of the network content as expected output, and training and testing a network content type identification model. The technical effects are as follows: the information of the user is considered, the information of the platform where the network content of the user is located and the information of the network content are considered, and the information of the network content are taken as the basis for predicting the type of the network content, so that the accuracy of prediction can be further improved.
7, A fifth model training test step of network content type identification: initializing a deep learning model as a network content type identification model; and training and testing a network content type identification model by taking information of a user watching the network content, behaviors of the user watching the network content, information of a platform where the network content is located and information of the network content as input and taking a type label of the network content as expected output. The technical effects are as follows: the method and the device not only consider the information of the user, but also consider the behavior of the user, the information of the platform where the network content of the user is located and the information of the network content, and the information is taken as the basis for predicting the type of the network content together, so that the accuracy of prediction can be further improved.
After the step 7 of the process, the process is carried out,
8, The network content type identification model uses a data acquisition step: acquiring information of a user watching network content to be identified, behavior of the user watching the network content, information of a platform of the network content and information of the network content;
9, training test data acquisition of a user information prediction model: acquiring information of a user for training and testing and information of each network content watched by the user;
10, training and testing a user information prediction model: taking information of each network content watched by the user (including information of a plurality of network contents if the user watches a plurality of network contents) as input, taking the information of the user as expected output, and training and testing a user information prediction model;
11, user information acquisition: if the information of the user watching the network content cannot be obtained, taking the information of each network content watched by the user (including the information of a plurality of network contents if the user watches a plurality of network contents) as input, and taking the output obtained through calculation of a user information prediction model as the obtained information of the user watching the network content; the technical effects are as follows: the information of the user of the network content can not be obtained, and the information of the user can be predicted according to the user information prediction model according to the information of the network content watched by the user in turn;
12 network content type identification fifth model use step: if the information of the user watching the network content to be identified, the behavior of the user watching the network content, the information of the platform of the network content and the information of the network content can be obtained, the information of the user watching the network content to be identified, the behavior of the user watching the network content, the information of the platform of the network content and the information of the network content are taken as input, and the output obtained through the calculation of the network content type identification fifth model is taken as a type label of the network content to be identified and is taken as a first label.
13 Network content type identification fourth model use step: if only the information of the user watching the network content to be identified, the information of the platform of the network content and the information of the network content can be obtained, the information of the user watching the network content to be identified, the information of the platform of the network content and the information of the network content are taken as input, and the output obtained through the calculation of the fourth model for identifying the network content type is taken as the type label of the network content to be identified and is taken as the first label.
14 Network content type identification third model use step: if only the information of the user watching the network content to be identified and the information of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification third model is taken as a type label of the network content to be identified and is taken as a first label.
15 Network content type identification second model use step: if only the information of the user watching the network content to be identified and the information of the platform of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification second model is taken as a type label of the network content to be identified and is taken as a first label.
16 Network content type identification first model use step: if only the information of the user watching the network content to be identified can be obtained, the information of the user watching the network content to be identified is taken as input, and the output obtained through the calculation of the network content type identification first model is taken as the type label of the network content to be identified and is taken as the first label. The technical effects are as follows: in consideration of the difficulty of acquiring different data, different models are called for identification according to the acquired data, and the applicability of the models can be greatly improved.
After the step 7 of the process, the process is carried out,
17, Acquiring a training test data set of the network content type rechecking model: network content for training and testing, and type tags of the network content are acquired.
18, Training and testing the network content type rechecking model: initializing a deep learning model as a network content type rechecking model; and taking the network content as input, taking the type label of the network content as expected output, and training and testing a network content type rechecking model.
19, Using a network content type rechecking model: if the type label of the network content to be identified is unhealthy content (unhealthy content includes bad content, illegal content, etc.), inputting the network content to be identified into a network content type rechecking model, and taking the calculated output as the type label of the network content to be identified as a second label. The technical effects are as follows: because most of the network content is healthy content, and only a small part of the network content is of other types, unhealthy suspected content is firstly and rapidly screened out based on a deep learning model of a user, and then the suspected content is rechecked by using a time-consuming deep learning model based on the network content, so that the accuracy of identification is improved.
After the step 19 of the process, the process is carried out,
20, Manually rechecking: if the second label is consistent with the first label, returning the second label to the user as the type label of the network content to be identified; if the second label is inconsistent with the first label, requesting for manual review, and acquiring a type label obtained by manual review as a third label; the technical effects are as follows: the detection speed of the network content is increased and the detection accuracy can be ensured by quickly identifying the network content based on the user information, checking the network content based on the user information and checking the network content based on the manual work.
And (3) performing incremental training on the network content type rechecking model: if the third label is inconsistent with the first label, the information of a user watching the network content to be identified, the information of a platform of the network content and the information of the network content are taken as input, the third label is taken as expected output, and incremental training is carried out on the network content type identification model; if the third label is inconsistent with the second label, the network content to be identified is taken as expected output, and incremental training is carried out on the network content type rechecking model. The technical effects are as follows: the network content type identification model and the network content type rechecking model are subjected to incremental training through predicting the actual type of inaccurate network content while the network content type prediction is carried out through the network content type identification model and the network content type rechecking model, so that the accuracy of the model can be continuously improved in the use process of the model.
After the step 7 of the process, the process is carried out,
22 Bad content management steps: if the type label of the network content to be identified is unhealthy content, notifying a user that if the type label of the network content to be identified is unhealthy content, timely processing is needed.
The foregoing examples illustrate only a few embodiments of the invention and are described in detail herein without thereby limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the spirit of the invention, which are within the scope of the invention. Accordingly, the scope of protection of the present invention is to be determined by the appended claims.
Claims (10)
1. An artificial intelligence governance method, the method comprising:
The network content type identification first model use step: if the information of the user watching the network content to be identified can be obtained, the information of the user watching the network content to be identified is taken as input, and the output obtained through calculation of the network content type identification first model is taken as the type label of the network content to be identified.
2. The artificial intelligence governance method of claim 1, further comprising:
The network content type identification second model use step: if the information of the user watching the network content to be identified and the information of the platform of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification second model is taken as the type label of the network content to be identified.
3. The artificial intelligence governance method of claim 1, further comprising:
The network content type identification third model using step: if the information of the user watching the network content to be identified and the information of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification third model is taken as the type label of the network content to be identified.
4. The artificial intelligence governance method of claim 1, further comprising:
And a bad content treatment step: if the type label of the network content to be identified is unhealthy content, notifying a user that if the type label of the network content to be identified is unhealthy content, timely processing is needed.
5. An artificial intelligence governance system, the system comprising:
The network content type identification first model use module: if the information of the user watching the network content to be identified can be obtained, the information of the user watching the network content to be identified is taken as input, and the output obtained through calculation of the network content type identification first model is taken as the type label of the network content to be identified.
6. The artificial intelligence abatement system of claim 5, further comprising:
The network content type identification second model use module: if the information of the user watching the network content to be identified and the information of the platform of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification second model is taken as the type label of the network content to be identified.
7. The artificial intelligence abatement system of claim 5, further comprising:
The network content type identification third model use module: if the information of the user watching the network content to be identified and the information of the network content can be obtained, the information of the user watching the network content to be identified and the information of the platform of the network content are taken as input, and the output obtained through the calculation of the network content type identification third model is taken as the type label of the network content to be identified.
8. The artificial intelligence abatement system of claim 5, further comprising:
Bad content management module: if the type label of the network content to be identified is unhealthy content, notifying a user that if the type label of the network content to be identified is unhealthy content, timely processing is needed.
9. A robot comprising a memory, a processor and an artificial intelligence robot program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1-4 when the program is executed by the processor.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the steps of the method according to any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410026645.6A CN118227871A (en) | 2024-01-08 | 2024-01-08 | Method and robot for controlling network content based on artificial intelligence of user type |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410026645.6A CN118227871A (en) | 2024-01-08 | 2024-01-08 | Method and robot for controlling network content based on artificial intelligence of user type |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118227871A true CN118227871A (en) | 2024-06-21 |
Family
ID=91510222
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410026645.6A Pending CN118227871A (en) | 2024-01-08 | 2024-01-08 | Method and robot for controlling network content based on artificial intelligence of user type |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118227871A (en) |
-
2024
- 2024-01-08 CN CN202410026645.6A patent/CN118227871A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110275958B (en) | Website information identification method and device and electronic equipment | |
| CN110222512B (en) | Software vulnerability intelligent detection and positioning method and system based on intermediate language | |
| CN105283866B (en) | Include the use of the application tracking method and system of the optimization analysis of similar frequencies | |
| CN111783016B (en) | Website classification method, device and equipment | |
| CN112183758A (en) | Method and device for realizing model training and computer storage medium | |
| CN110175851B (en) | Cheating behavior detection method and device | |
| CN110163376B (en) | Sample detection method, media object identification method, device, terminal and medium | |
| US20100030717A1 (en) | Framework to evaluate content display policies | |
| Limberg et al. | Advancing the counseling profession by examining relationships between variables | |
| CN111160783B (en) | Digital asset value evaluation method and system and electronic equipment | |
| CN113988044B (en) | Method for judging error question reason type | |
| CN116361104A (en) | Big data-based application fault prediction method, device, equipment and storage medium | |
| CN115147353A (en) | Defect detection model training method, device, equipment, medium and program product | |
| CN115100739A (en) | Man-machine behavior detection method, system, terminal device and storage medium | |
| CN111199172A (en) | Terminal screen recording-based processing method and device and storage medium | |
| CN110659954A (en) | Cheating identification method and device, electronic equipment and readable storage medium | |
| CN113869427A (en) | Scene analysis method and device, electronic equipment and storage medium | |
| CN116883909A (en) | Live sensitive image identification method, live sensitive image identification device, computer equipment and storage medium | |
| McKibben et al. | To model or not to model: false positive detection error in camera surveys | |
| CN118227871A (en) | Method and robot for controlling network content based on artificial intelligence of user type | |
| CN111382383A (en) | Method, device, medium and computer equipment for determining sensitive type of webpage content | |
| CN114722280A (en) | User portrait based course recommendation method, device, equipment and storage medium | |
| CN113673589A (en) | Label selection self-adaptive increment detection method and system based on frame distance measurement | |
| CN112131418A (en) | Target labeling method, target labeling device and computer-readable storage medium | |
| CN114610982B (en) | Computer network data acquisition, analysis and management method, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination |