WO2024000554A1

WO2024000554A1 - Artificial intelligence-based method for configuring image for information, device, medium, and program product

Info

Publication number: WO2024000554A1
Application number: PCT/CN2022/103226
Authority: WO
Inventors: 李明宇; 庄建家; 王嘉楠; 吴志伟; 郑泽伟
Original assignee: 富途网络科技(深圳)有限公司
Priority date: 2022-07-01
Filing date: 2022-07-01
Publication date: 2024-01-04
Also published as: CN115298660A

Abstract

The present application discloses an artificial intelligence-based method and apparatus for configuring an image for information, a device, a medium, and a program product. The method comprises: obtaining information content; determining an information theme label of the information content on the basis of a preset algorithm model; determining whether a target theme label matching the information theme label is present in a first image library, wherein images in the first image library are labeled with corresponding theme labels; and if a target theme label matching the information theme label is present in the first image library, using an image corresponding to the target theme label as a cover image of the information content.

Description

Information mapping methods, equipment, media and program products based on artificial intelligence

Technical field

This application relates to the field of computer technology, specifically to an information mapping method, device, equipment, medium and program product based on artificial intelligence.

Background technique

In financial information operations, the work of configuring cover images is often tedious. Faced with massive amounts of information every day, operators need to find pictures with suitable copyright and quality to configure cover images. The work is repetitive and ineffective.

The current information matching methods mainly include two types: random matching of cover pictures and manual configuration of cover pictures. Among them, when randomly matching the cover image, some pictures without a specific theme are randomly used as the cover image. The main disadvantage is that the theme of the picture is not clear and it seems to have nothing to do with the cover image. The user feels that the picture and text have nothing to do with it, and the list reading effect is not good. good. When manually configuring the cover image, you manually find pictures related to the theme of a single article and upload them as the information cover image. The main disadvantage is that the work is too repetitive and cumbersome, and the matching of pictures for a large amount of information requires a lot of manpower.

Contents of the invention

Embodiments of the present application provide an information mapping method, device, equipment, media and program product based on artificial intelligence, which can realize intelligent configuration of cover images, reduce labor costs, enhance the degree of fit between information content and graphics, and improve the quality of information The efficiency and effectiveness of illustrations.

On the one hand, embodiments of the present application provide an information mapping method based on artificial intelligence. The method includes:

Obtain information content;

Determine the information topic tag of the information content based on a preset algorithm model;

Determine whether there is a target topic tag matching the information topic tag in the first image gallery, wherein the pictures in the first image gallery are marked with corresponding topic tags;

If there is a target topic tag matching the information topic tag in the first image gallery, the picture corresponding to the target topic tag is used as the cover image of the information content.

On the other hand, embodiments of the present application provide an information mapping device based on artificial intelligence. The device includes:

Acquisition unit, used to obtain information content;

A determining unit, configured to determine the information topic tag of the information content based on a preset algorithm model;

A judging unit configured to judge whether there is a target topic tag matching the information topic tag in the first gallery, wherein the pictures in the first gallery are marked with corresponding topic tags;

A processing unit configured to, if there is a target topic tag matching the information topic tag in the first gallery, use the picture corresponding to the target topic tag as a cover image of the information content.

On the other hand, embodiments of the present application provide a computer device. The computer device includes a processor and a memory. A computer program is stored in the memory. The processor calls the computer program stored in the memory. Used to execute the information mapping method based on artificial intelligence as described in any of the above embodiments.

On the other hand, embodiments of the present application provide a computer-readable storage medium that stores a computer program, and the computer program is suitable for loading by a processor to execute the steps described in any of the above embodiments. Information mapping method based on artificial intelligence.

On the other hand, embodiments of the present application provide a computer program product that includes computer instructions. When the computer instructions are executed by a processor, the artificial intelligence-based information mapping method described in any of the above embodiments is implemented.

The embodiment of the present application obtains information content; determines the information topic tag of the information content based on a preset algorithm model; and determines whether there is a target topic tag matching the information topic tag in the first gallery, where the pictures in the first gallery are marked with The corresponding topic tag; if there is a target topic tag matching the information topic tag in the first gallery, the picture corresponding to the target topic tag will be used as the cover image of the information content. In the embodiment of the present application, an algorithm based on artificial intelligence is used to identify information topic tags, which improves the accuracy of information positioning for information of various contents; and tag retrieval and matching is performed in the first image gallery, and the information in the first image gallery is The picture corresponding to the target topic tag that matches the topic tag is used as the cover image of the information content, which can realize intelligent configuration of the cover image, reduce labor costs, enhance the fit between the information content and the accompanying images, and improve the efficiency and effect of the information matching.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, other drawings can also be obtained based on these drawings without exerting creative efforts.

Figure 1 is a schematic flowchart of an information mapping method based on artificial intelligence provided by an embodiment of the present application.

Figure 2 is a schematic diagram of the first application scenario provided by the embodiment of the present application.

Figure 3 is a schematic diagram of the second application scenario provided by the embodiment of the present application.

Figure 4 is a schematic diagram of the third application scenario provided by the embodiment of the present application.

Figure 5 is a schematic diagram of the fourth application scenario provided by the embodiment of the present application.

Figure 6 is a schematic diagram of the fifth application scenario provided by the embodiment of the present application.

Figure 7 is a flow sequence diagram of the artificial intelligence-based information mapping method provided by the embodiment of the present application.

Figure 8 is a schematic structural diagram of an information mapping device based on artificial intelligence provided by an embodiment of the present application.

Figure 9 is a schematic structural diagram of a server provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts fall within the scope of protection of this application.

Embodiments of the present application provide an information mapping method, device, terminal equipment and storage medium based on artificial intelligence. Specifically, the information mapping method based on artificial intelligence in the embodiment of the present application can be executed by a computer device, where the computer device can be a terminal or a server. The terminal can be a smartphone, tablet, laptop, desktop computer, smart TV, smart speaker, wearable smart device, smart vehicle terminal and other devices. The terminal can also include a client, which can be a financial client, browser server client or instant messaging client, etc. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also provide cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, and middleware. Cloud servers include software services, domain name services, security services, content distribution network services, and basic cloud computing services such as big data and artificial intelligence platforms, but are not limited to these.

Each is explained in detail below. It should be noted that the description order of the following embodiments is not used to limit the priority order of the embodiments.

Please refer to Figures 1 to 7. Figure 1 is a schematic flow chart of the artificial intelligence-based information mapping method provided by the embodiment of the present application. Figures 2 to 6 are schematic diagrams of application scenarios provided by the embodiment of the present application. Figure 7 is a schematic diagram of the application scenario. A flow sequence diagram of the artificial intelligence-based information mapping method provided in the application embodiment. The information mapping method based on artificial intelligence in the embodiment of the present application can be applied to the server. The method includes the following steps:

Step 110: Obtain information content.

In some embodiments, the obtained information content includes:

Crawl the information content based on the information content source and store the information content into the information content library; or

In response to the storage request for the information content sent by the background management device, the information content is obtained, and the information content is stored in the information content library.

Among them, an indispensable part of information operations is information mapping. The acquisition method of an information content can include crawling the content source, automatically reviewing it, or manually creating a new one and adding it to the database.

For example, on the basis of copyright cooperation with the original information website, crawler tools are used to crawl information content from the information content source. The crawler tool is a program or script that automatically crawls World Wide Web information according to certain rules. The crawler tool initiates a request to the target site through the HTTP library, that is, it sends a Request. The request can contain additional headers and other information and waits for the server to respond; if the server can respond normally, it will get a Response, and the content of the Response is the content of the page to be obtained. , the type may include HTML, Json string, binary data (such as pictures and videos), etc.; the obtained content can be HTML, which can be parsed using regular expressions and web page parsing libraries; the obtained content can also be Json, which can be directly converted It is parsed for Json objects, usually binary data, which can be saved or further processed; the information crawled by the crawler tool can be saved as text, saved to a database, or saved as a file in a specific format.

For example, the information content can also be manually added to the database, and the information content can be obtained in response to a database entry request for the information content sent by the background management device, and the information content can be stored in the information content database.

In some embodiments, after obtaining the information content, the method further includes:

Conduct information review on the information content based on preset review rules, where the preset review rules at least include sensitive word matching and filtering rule verification;

If the information content does not match the sensitive word and the information content does not match the filtering rule, then the step of determining the information topic tag of the information content based on the preset algorithm model is performed.

In some embodiments, the method further includes:

If the information content hits a sensitive word, and/or the information content hits a filtering rule, a first prompt message indicating that the review fails is generated, and the information content and the first prompt message are sent to the backend management equipment.

For example, take the information content obtained by crawling the information content source as an example. The crawling of the information content source occupies a large part of the source of the information content. Most of the information stored in this way does not have pictures, so the follow-up needs to be based on Information content obtains information topic tags and automatically generates accompanying images.

The information content crawled by the content source will be stored in the information content library first. Before step 120, the information content needs to be subjected to information review processes such as sensitive word matching and filtering rule verification through the information review process.

Among them, the sensitive thesaurus and the filtering rule thesaurus are pre-selected and built thesaurus. When the information content is entered into the database, text matching is performed on the title, information source, text and other fields of the information content. If the sensitive word or the filter word is hit, Then it is automatically determined that the review has failed, and the operator needs to conduct manual review before determining whether to further perform step 120. If the sensitive word is not hit and the filter word is not hit, it is automatically determined that the review is passed, and step 120 is further executed.

Step 120: Determine the information topic tag of the information content based on a preset algorithm model.

In some embodiments, determining the information topic tag of the information content based on a preset algorithm model includes:

Obtain the information title text corresponding to the information text data in the information content, and process the information title text based on the preset algorithm model to determine the information topic tag of the information content.

In some embodiments, the preset algorithm model includes an entity extraction model and a keyword extraction model, and the information title text is processed based on the preset algorithm model to determine the information topic tag of the information content. ,include:

The information title text is processed based on an entity extraction model to obtain the entity type label of the information content, wherein the entity extraction model is used to extract company, industry and person name information in the information title text;

Process the information title text based on a keyword extraction model to obtain keyword tags for the information content, where the keyword extraction model is used to extract industry keyword information in the information title text;

According to the entity type tag and the keyword tag, the information topic tag of the information content is determined.

In some embodiments, determining the information topic tag of the information content based on the entity type tag and the keyword tag includes:

Generate a tag list according to the entity type tag and the keyword tag;

The first-ranked tag in the tag list is determined as the information topic tag of the information content.

In the embodiment of this application, an information topic tag API interface can be provided on the preset algorithm model side. According to the information title text transmitted from the back end, the entity extraction model (denoted as M1) and the keyword extraction model (denoted as M2 ), ultimately returning a list of tags. The first-ranked tag in the backend selection list is displayed as the information topic tag.

The entity extraction model and keyword extraction model of the information topic tag API interface share the same model structure from the model level, but they design different entity tags in the data collection stage, so the training data constructed are different, and the final extraction Ability to achieve different results.

For example, the entity extraction model mainly performs data annotation on the three most important entity types that are concerned in the financial field: companies, industries, and names of people.

For example, the keyword extraction model mainly annotates data for common and important words in financial information, such as: inflation, stock market, Hong Kong stocks, Federal Reserve, etc.

After collecting the corresponding annotation data, you can use the natural language processing pre-training model Bert and the global pointer module to build a named entity recognition model. The annotated data from different sources were used to fine-tune the models respectively, and finally two result models were obtained, namely the above-mentioned M1 and M2.

Among them, the Bert model is composed of an Embedding layer and 12 Transformer layers, with a total of 110 million parameters. The model parameters are very large. In this project, the Bert model serves as text_encoder to extract features from the input information title text. The input information title text will first go through the tokenizer (Tokenizer) in Bert to obtain a token sequence of length L. The token sequence further converts the token text into word id according to the mapping relationship in vocab, obtaining [1,L ], and then input it into the Bert model. Bert serves as the encoder. The tensor of [1, L] passes through the Embedding layer to obtain the tensor of [L, D] dimensions. This tensor is recorded as R (because it is Bert model, then D = 768), then, the tensor inputs the full pointer layer and outputs a [n_labels, L, L], n_labels is the total number of entity categories, for example: for the entity extraction model M1 that extracts companies, industries, and personal names at the same time, n_labels=3; the entity extraction model M2 that only extracts keywords has n_labels=1.

The function of the full pointer layer is to use text_encoder to extract rich semantic information of the entity, and indicate the head and tail of the entity through a pointer matrix at one time, so that the position of the entity in the original text can be quickly located and directly extracted. In this embodiment, a simplified version of the Multi-Head Attention module is used to implement this function. The Multi-Head Attention module uses three matrices Q (query), K (all keys), and V (values) to perform matrix calculations, and then performs Scaled Dot-Product Attention calculations. Here, the Q and K matrices (both matrices of [D, d]) are directly used, as well as the [L, D] dimension tensor obtained above (the tensor is recorded as R), to reduce the input [L, D] In the feature space with dimensions [L, d] (d=64 usually <<D), they are recorded as q and k. The relevant formulas are as follows:

q=R·Q;k=R·K;

Among them, S _α (i, j) represents the pointer square matrix of the αth type entity, and its shape is [L, L]. When there are n_labels classes, each entity category will calculate such a pointer square matrix, so the entire The output of the full pointer layer is a tensor of [n_labels,L,L]. It should be noted that the rows of S _α represent the head position of the entity, and the columns represent the tail position of the entity. Therefore, although S _α is a square matrix, only the upper triangular part has practical significance, and the output of the lower triangle is not considered directly.

The function of the classification output layer is to extract entities from the output pointer square matrix. Values greater than 0 in [n_labels, L, L] are considered to be the head and tail of activated entities. Therefore, this layer converts the logits output by the model into 0/1 In a binary square matrix, the head and tail of the activated entity are set to 1, and the rest are set to 0.

Among them, after the API interface receives the information title text, it first passes the text into the model M1. If the model does not return a result, that is, the entity extraction model M1 does not extract any company, industry, or person name information, it will continue to pass the title into the keywords. Extract model M2 and further extract keywords.

As shown in the schematic diagram of the application scenario in Figure 2, during the training phase, the obtained information text data is manually annotated. For example, by performing data annotation on the three most important entity types in the information text data: company, industry, and person's name. Get the first labeled data. By performing data annotation on the industry keywords in the information text data, the second annotation data is obtained. The first annotated data is input into the preset algorithm model for model training. For example, the first annotated data and the Bert model are used to pre-train multiple transformers layers, global pointer layers and classification output layers, and the entity extraction model is obtained as M1). Input the second annotated data into the preset algorithm model for model training. For example, use the second annotated data and the Bert model to pre-train multiple transformers layers, global pointer layers and classification output layers to obtain the keyword extraction model M2).

In the application stage, the information title text is input into the entity extraction model M1 for the first time to extract the company, industry and person name information in the information title text, and output the first result set; then, determine whether the first result set is empty , if the first result set is not empty, then the entity type tags contained in the first result set are included in the final result set; if the first result set is empty, the second input keyword of the information title text is retrieved. In model M2, the industry keyword information in the information title text is extracted, and a second result set is output, and the keyword tags contained in the second result set are included in the final result set. Among them, in the final result set, a tag list can be constructed, and the back-end selects the first-ranked tag in the list to be displayed as the information topic tag. That is, when the algorithm provides multiple tags with the same frequency, the information side can select the first tag for image matching.

The above method obtains the entity extraction model M1 and the keyword extraction model M2 based on artificial intelligence training, and finally returns a tag list, and displays the tag ranked first in the back-end selection list as the information topic tag, which improves the determination of information topic tags. accuracy.

Step 130: Determine whether there is a target topic tag matching the information topic tag in the first image gallery, where the pictures in the first image gallery are annotated with the corresponding topic tag.

For example, after the information topic tag is generated through the preset algorithm model, a search and matching is performed in the first image gallery to find whether there is a target topic tag matching the information topic tag in the first image gallery.

For example, in the application scenario diagram shown in Figure 3, an information management platform interface is displayed on the background management device. In this information management platform interface, the first image gallery (the label image gallery as shown in Figure 3) can be maintained, in which the information When maintaining each picture in the first gallery on the management platform interface, the thumbnail, ID information, storage time, corresponding topic tag, number of hits and other information of the picture can be displayed.

Step 140: If there is a target topic tag matching the information topic tag in the first gallery, use the picture corresponding to the target topic tag as the cover image of the information content.

Among them, after a piece of information is crawled from the content source and automatically reviewed and added to the database, or it is manually created and added to the database, a cover image needs to be assigned to the information content. After the information is crawled into the database or manually created and saved, the intelligent matching can extract the information theme tags of the information content according to the preset algorithm model, and match the pictures in the gallery according to the information theme tags to complete the intelligent matching of the cover image of the information content. . If there is a target topic tag matching the information topic tag in the first image gallery, the picture corresponding to the target topic tag will be used as the cover image of the information content.

In some embodiments, if there is a target topic tag matching the information topic tag in the first gallery, then using the picture corresponding to the target topic tag as the cover image of the information content, including:

If there is a target topic tag matching the information topic tag in the first gallery, and there are multiple pictures corresponding to the target topic tag, then select one picture from the multiple pictures corresponding to the target topic tag. Said as the cover image of the information content.

For example, when multiple pictures corresponding to the target topic label are matched based on the information topic label, one picture can be randomly selected from the multiple pictures corresponding to the target topic label as the cover image of the information content. For example, according to the number of hits, you can select the picture with the most hits from the multiple pictures corresponding to the target topic tag as the cover image of the information content, so as to select the most commonly used pictures in the target topic tag for illustration, so as to fit the public as much as possible. aesthetic. For example, based on the number of hits, you can select the one with the least number of hits from the multiple pictures corresponding to the target topic tag as the cover image of the information content, so as to select the rarely used pictures in the target topic tag for illustration to increase the freshness of the illustration. .

In some embodiments, the method further includes:

If there is no target topic tag matching the information topic tag in the first image gallery, a random picture is obtained from the second image gallery as the cover image of the information content, wherein the picture in the first image gallery is Hashtags not tagged.

For example, in the application scenario diagram shown in Figure 4, an information management platform interface is displayed on the background management device. In this information management platform interface, a second image gallery (the ordinary image gallery as shown in Figure 3) can be maintained, in which the information When maintaining each picture in the second gallery on the management platform interface, the thumbnail and ID information of the picture can be displayed.

For example, for a small number of information topic tags that are not maintained in the first image gallery, abstract pictures without specific themes in the second image gallery can be used as the cover image of the information content.

In some embodiments, the method further includes:

When a cover image configuration request for the information content sent by the background management device is received, the cover image of the information content is determined according to the configuration information carried in the cover image configuration request.

For example, after information is reviewed and stored in the database, operators can still re-edit the relevant information corresponding to the information content. For example, if the cover image generated by automatic image matching still does not meet the requirements, the information content can also be manually intervened in the image matching. The information personnel can enter a cover image configuration request for the information content on the background management device. The cover image configuration request The configuration information carries the specified cover image. When the server receives the cover image configuration request for the information content sent by the background management device, it determines the cover image for the information content based on the configuration information carried in the cover image configuration request. This can greatly increase the number of images. flexibility and improve operational efficiency.

In some embodiments, the method further includes:

When receiving a cover image display request for the information content sent by the client, sending the cover image of the information content to the client to display the cover image of the information content on the information preview interface of the client picture.

In some embodiments, the method further includes:

When receiving the triggering operation for the cover image sent by the client, send the information flow of the information content to the client, so as to display the information of the information content on the information preview interface of the client. Stream, the information stream includes at least one of information text data and information multimedia data corresponding to the information content.

For example, when a user is interested in the information content corresponding to the cover image, he or she will trigger operations such as clicking on the cover image or long-pressing the cover image, causing the server to push an information flow of information content to the client. The information flow may include information text data, information multimedia data, etc., where the information multimedia data may include multimedia resources such as images, animations, audios, and videos. Among them, the information text data may include information title, ID, release time, information content details, information content release end information, etc.

In some embodiments, the method further includes:

Obtain image matching record data within a preset period, wherein each of the image matching records includes a pair of information theme tags and an artificial intelligence-based information matching method corresponding to the information theme tag;

According to the picture recording data, count the frequency of occurrence of each information topic tag within the preset period to obtain tag statistics;

According to the mapping record data, count the data proportion of each artificial intelligence-based information mapping method within the preset period to obtain mapping statistics;

The label statistical data and the accompanying picture statistical data are sent to a backend management device, so that the label statistical data and the accompanying picture statistical data are displayed on the backend management device.

Among them, in order to continue to enrich the gallery of tag images and optimize the accuracy of images, it is necessary to add statistical tag statistics and image statistics. Operators can continue to maintain the image gallery based on the statistical results to improve the accuracy of intelligent images. A separate background display is provided for the label statistics and picture statistics of the information content to facilitate locating review issues.

For example, every time a picture is matched, a picture record will be recorded, and label statistics and picture statistics will use this as the data source statistical result. That is, data statistics are performed based on the picture recording data recorded within a preset period of time to determine the label statistics and picture statistics. For example, the preset time period may be one of the time periods of the last week, the last month, the last three months, the last year, etc.

Among them, for label statistics, due to the large amount of data recorded with pictures, the amount of data generated every day may reach tens of thousands. Simply relying on the aggregation statistics of the database cannot meet the performance requirements, so it is necessary to calculate the daily statistics on a daily basis. Tag statistical data, as shown in the application scenario diagram in Figure 5, when operators query tag statistics, they can enter a preset period, such as the last month, through the information management platform interface displayed on the background management device, and add the preset period Set a time period and send it to the server. The server queries the single-day tag statistics within the preset time period according to the preset time period and aggregates them into tag statistics within the preset time period, and returns the tag statistics within the preset time period to the backend management. device to display. By displaying tag statistical data on the background management device, operators can perform targeted statistics on frequently occurring tags when maintaining the first image gallery (the tag image gallery shown in Figure 3), and cover high-frequency tags as much as possible. The information content of the information theme makes the gallery operation more efficient. For example, the operation staff recently discovered that "oil price" is a label that appears frequently, so they maintain multiple related cover images for the "oil price" label. In this way, when the information content of the same label appears again in the future, the intelligent image matching will take effect. As shown in Figure 5, when counting tags, the frequency of occurrence of information topic tags is also counted to facilitate targeted maintenance of images.

Among them, for the picture statistics data, due to the large amount of picture record data, the required statistical results cannot be obtained simply by relying on the aggregation statistics of the database. Therefore, the picture statistics data within the previous 30 days need to be calculated daily and stored in the database. , when viewing the matching statistics data, you can query the corresponding matching statistics data in the database according to the query date and return it. As shown in the application scenario diagram shown in Figure 6, when operators query image matching statistics, they can enter the preset period through the information management platform interface displayed on the background management device, for example, push forward within 7 days from 2022-5-17. The preset time period is sent to the server. The server queries the single-day picture matching statistics within the preset time period according to the preset time period and aggregates the picture matching statistics within the preset time period, and aggregates the picture matching statistics within the preset time period. The graph statistical data is returned to the background management device for display. The actual data proportion (coverage) of various types of graphics online can be displayed through the graphics statistical data, so as to have a more objective data evaluation of the intelligent graphics mechanism and provide a truly quantifiable statistical library coverage effect. For example, the types of graphics can include label mapping, random mapping, and manual mapping. In the mapping statistics shown in Figure 6, the coverage rate of label mapping is 3.40%, and the coverage rate of random mapping is 96.40%. Manual illustration coverage is 0.20%.

In some embodiments, the method further includes:

When an update request for the first gallery sent by the background management device is received, at least one of pictures and topic tags in the first gallery is updated according to the update request.

For example, you can also manually maintain the first image gallery and maintain related cover images for common information topic tags. The operator can maintain the first image gallery through the information management platform interface displayed on the background management device, input an update request for the first image gallery through the information management platform interface, and send the update request to the server, so that the server updates the first image gallery according to the update request. At least one of the pictures and hashtags in . For example, if the update request is to update the first topic tag corresponding to the first picture, then the first topic tag corresponding to the first picture in the first gallery is updated according to the update request. For example, if the update request is to update the second picture corresponding to the second topic tag, then the second picture corresponding to the second topic tag in the first gallery is updated according to the update request. For example, if the update request is to add a third picture and its corresponding third topic tag, then the third picture and its corresponding third topic tag are added to the first gallery according to the update request.

In some embodiments, the method further includes:

If it is detected that the same topic tag as the tag to be updated carried in the update request already exists in the first image gallery, a second prompt message indicating a duplication of tags is generated, and the update of the first image gallery is refused.

For example, one image cannot create multiple identical tags. If it is detected that the same topic tag as the tag to be updated carried in the update request already exists in the first gallery, a second prompt message indicating duplicate tags will be generated, and the update of the first image gallery will be rejected. gallery, and sends the second prompt information to the background management device to display the second prompt information on the background management device to remind the operator that the same tag already exists.

In order to better illustrate the artificial intelligence-based information mapping method provided by the embodiment of the present application, please refer to Figure 7. The process of the artificial intelligence-based information mapping method provided by the embodiment of the present application can be summarized into the following steps:

S1. The server obtains information content;

S1.1. The server crawls the information content based on the information content source and stores the information content into the information content library;

S1.2. The server responds to the storage request for information content sent by the background management device, obtains the information content, and stores the information content into the information content library;

S1.3. The server audits the information content based on the preset audit rules. The preset audit rules at least include sensitive word matching and filtering rule verification;

S1.4. If the information content hits the sensitive word, and/or the information content hits the filtering rule, the server determines that the information review fails, generates the first prompt message that the review fails, and sends the information content and the first prompt message to the background. Manage equipment;

S1.5. The background management device displays the first prompt message;

S1.6. If the information content does not match the sensitive words and the information content does not match the filtering rules, the server determines that the information review has passed and further proceeds to step S2;

S2. The server determines the information topic tag of the information content based on the preset algorithm model;

S2.1. The server processes the information title text of the information content based on the entity extraction model to obtain the entity type label of the information content. The entity extraction model is used to extract the company, industry and person name information in the information title text;

S2.2. The server processes the information title text of the information content based on the keyword extraction model to obtain the keyword tags of the information content. The keyword extraction model is used to extract industry keyword information in the information title text;

S2.3. The server determines the information topic tag of the information content based on the entity type tag and keyword tag;

S3. The server determines whether there is a target topic tag matching the information topic tag in the first gallery, and the pictures in the first gallery are annotated with the corresponding topic tag;

S4. If there is a target topic tag matching the information topic tag in the first gallery, the server will use the picture corresponding to the target topic tag as the cover image of the information content;

S5. If there is no target topic tag matching the information topic tag in the first image gallery, the server obtains a random picture from the second image gallery as the cover image of the information content. The pictures in the first image gallery are not labeled with topic tags;

S6. When the server receives the cover image configuration request for the information content sent by the background management device, it determines the cover image of the information content based on the configuration information carried in the cover image configuration request;

S7. When the server receives the cover image display request for the information content sent by the client, it sends the cover image of the information content to the client;

S8. Display the cover image of the information content on the information preview interface of the client;

S9. When the server receives the trigger operation for the cover image sent by the client, it sends the information flow of the information content to the client;

S10. Display the information flow of the information content on the client's information preview interface. The information flow includes at least one of information text data and information multimedia data corresponding to the information content;

S11. The server obtains the picture record data within the preset period. Each picture record contains a pair of information topic tags and a method for determining the cover image corresponding to the information topic tag;

S12. The server records the data based on the picture and counts the frequency of occurrence of each information topic tag within the preset period to obtain tag statistics;

S13. The server records the data of the accompanying images and counts the data proportion of the determination method of each cover image within the preset period to obtain the statistical data of the accompanying images;

S14. The server sends label statistics and picture statistics to the background management device;

S15. The background management device displays label statistics and picture statistics;

S16. When the server receives an update request for the first gallery sent by the background management device, it updates at least one of the pictures and topic tags in the first gallery according to the update request;

S17. If it is detected that the same topic tag as the tag to be updated carried in the update request already exists in the first gallery, the server generates a second prompt message indicating that the tag is repeated, and refuses to update the first gallery; in step S17, also The second prompt information can be sent to the background management device.

S18. The background management device displays the second prompt information.

All the above technical solutions can be combined in any way to form optional embodiments of the present application, and will not be described again one by one.

The artificial intelligence-based information mapping method provided by the embodiment of the present application obtains information content; determines the information topic tag of the information content based on a preset algorithm model; and determines whether there is a target topic tag matching the information topic tag in the first image gallery. , where the pictures in the first gallery are annotated with corresponding topic tags; if there is a target topic tag matching the information topic tag in the first gallery, the picture corresponding to the target topic tag is used as the cover image of the information content. In the embodiment of this application, information topic tags are identified through an algorithm, tag retrieval and matching is performed in the first gallery, and the picture corresponding to the target topic tag that matches the information topic tag in the first gallery is used as the cover image of the information content, which can be achieved Intelligent configuration of cover images reduces labor costs, enhances the fit between information content and accompanying images, and improves the efficiency and effect of information accompanying images.

In order to facilitate better implementation of the artificial intelligence-based information mapping method in the embodiment of the present application, the embodiment of the present application also provides a client. Please refer to Figure 8, which is a schematic structural diagram of an artificial intelligence-based information mapping device provided by an embodiment of the present application. Among them, the information mapping device 200 based on artificial intelligence may include:

Obtaining unit 210 is used to obtain information content;

The determining unit 220 is configured to determine the information topic tag of the information content based on a preset algorithm model;

The judging unit 230 is used to judge whether there is a target theme tag matching the information theme tag in the first gallery, wherein the pictures in the first gallery are marked with corresponding theme tags;

The processing unit 240 is configured to, if there is a target topic tag matching the information topic tag in the first gallery, use the picture corresponding to the target topic tag as the cover image of the information content.

In some embodiments, the determining unit 220 is specifically configured to: obtain the information title text corresponding to the information text data in the information content, and process the information title text based on the preset algorithm model to determine The information topic tag of the information content.

In some embodiments, the preset algorithm model includes an entity extraction model and a keyword extraction model, and the determination unit 220 processes the information title text based on the preset algorithm model to determine the information content. When the information topic label is used, it is specifically used to: process the information title text based on an entity extraction model to obtain the entity type label of the information content, wherein the entity extraction model is used to extract the information title text. Company, industry and person name information; the information title text is processed based on a keyword extraction model to obtain the keyword tags of the information content, wherein the keyword extraction model is used to extract the industry in the information title text Keyword information: determine the information topic tag of the information content according to the entity type tag and the keyword tag.

In some embodiments, when determining the information topic tag of the information content based on the entity type tag and the keyword tag, the determining unit 220 is specifically configured to: based on the entity type tag and the key word tag The word tag generates a tag list; the tag ranked first in the tag list is determined as the information topic tag of the information content.

In some embodiments, the acquisition unit 210 is specifically configured to: crawl information content based on the information content source and store the information content into the information content library; or respond to the information sent by the background management device for the information content. The information content warehousing request obtains the information content and stores the information content into the information content library.

In some embodiments, after the information mapping device 200 based on artificial intelligence is used to obtain information content, it is also used to: conduct information review of the information content based on preset review rules, wherein the preset review The rules at least include sensitive word matching and filtering rule verification; if the information content does not match the sensitive word, and the information content does not match the filtering rule, then perform the step of determining the information topic tag of the information content based on the preset algorithm model .

In some embodiments, the artificial intelligence-based information mapping device 200 is also configured to: if the information content hits a sensitive word, and/or the information content hits a filtering rule, generate a first review failed Prompt information, and send the information content and the first prompt information to the background management device.

In some embodiments, the processing unit 240 is also configured to: if there is no target topic tag matching the information topic tag in the first image gallery, obtain a random picture from the second image gallery as the The cover picture of the information content, wherein the pictures in the first gallery are not labeled with topic tags.

In some embodiments, the processing unit 240 is further configured to: when receiving a cover image configuration request for the information content sent by the background management device, determine the cover image configuration request based on the configuration information carried in the cover image configuration request. Cover image of the information content.

In some embodiments, the processing unit 240 is further configured to: when receiving a cover image display request for the information content sent by the client, send the cover image of the information content to the client, so as to Display the cover image of the information content on the information preview interface of the client.

In some embodiments, the processing unit 240 is further configured to: when receiving a triggering operation for the cover image sent by the client, send the information stream of the information content to the client, so as to The information flow of the information content is displayed on the information preview interface of the client. The information flow includes at least one of information text data and information multimedia data corresponding to the information content.

In some embodiments, the processing unit 240 is also configured to: obtain image record data within a preset period, wherein each image record includes a pair of information topic tags corresponding to the information topic tag. How to determine the cover image; according to the accompanying image record data, count the frequency of occurrence of each information theme tag within the preset period to obtain tag statistical data; according to the accompanying image record data, count the The data proportion of the determined method of each cover image within the preset period is used to obtain the accompanying image statistical data; the label statistical data and the accompanying image statistical data are sent to the background management device for management in the background The label statistical data and the accompanying picture statistical data are displayed on the device.

In some embodiments, the processing unit 240 is further configured to: when receiving an update request for the first image gallery sent by the background management device, update the information in the first image gallery according to the update request. At least one of pictures and hashtags.

In some embodiments, the processing unit 240 is further configured to: if it is detected that the same topic tag as the tag to be updated carried in the update request already exists in the first gallery, generate a third tag indicating that the tag is repeated. The second prompt message and refusal to update the first gallery.

In some embodiments, the processing unit 240 is configured to: if there is a target topic tag matching the information topic tag in the first gallery, and there are multiple pictures corresponding to the target topic tag, then from Select one picture from the plurality of pictures corresponding to the target topic tag as the cover picture of the information content.

It should be understood that the artificial intelligence-based information mapping device embodiments and method embodiments may correspond to each other, and similar descriptions may refer to the method embodiments. To avoid repetition, they will not be repeated here. Specifically, the artificial intelligence-based information mapping device shown in Figure 8 can execute the above-mentioned artificial intelligence-based information mapping method embodiment, and the aforementioned and other operations of each unit in the artificial intelligence-based information mapping device and/or or functions respectively implement the corresponding processes of the above method embodiments. For the sake of simplicity, they will not be described again here.

Optionally, this application also provides a computer device, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, it implements the steps in the above method embodiments.

Figure 9 is a schematic structural diagram of a computer device provided by an embodiment of the present application. The computer device may be a terminal or a server. As shown in FIG. 9 , the computer device 300 may include: a communication interface 301 , a memory 302 , a processor 303 and a communication bus 304 . The communication interface 301, the memory 302, and the processor 303 realize communication with each other through the communication bus 304. The communication interface 301 is used for data communication between the computer device 300 and external devices. The memory 302 can be used to store software programs and modules, and the processor 303 runs the software programs and modules stored in the memory 302, such as the software programs for corresponding operations in the foregoing method embodiments.

Optionally, the processor 303 can call software programs and modules stored in the memory 302 to perform the following operations: obtain information content; determine the information topic tag of the information content; determine whether the information topic tag exists in the first gallery Matching target topic tags, wherein the pictures in the first gallery are marked with corresponding topic tags; if there is a target topic tag matching the information topic tag in the first gallery, then the target The picture corresponding to the topic tag serves as the cover image of the information content.

Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions, or by controlling relevant hardware through instructions. The instructions can be stored in a computer-readable storage medium, and loaded and executed by the processor.

To this end, embodiments of the present application provide a computer-readable storage medium in which multiple computer programs are stored. The computer programs can be loaded by the processor to execute any of the artificial intelligence-based methods provided by the embodiments of the present application. Steps in the information mapping method. For the specific implementation of each of the above operations, please refer to the previous embodiments and will not be described again here.

Among them, the storage medium may include: read-only memory (Read Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk, etc.

Since the computer program stored in the storage medium can execute the steps in any of the artificial intelligence-based information mapping methods provided by the embodiments of the present application, it is possible to implement any of the artificial intelligence-based information mapping methods provided by the embodiments of the present application. The beneficial effects that the artificial intelligence information mapping method can achieve are detailed in the previous embodiments and will not be described again here.

Embodiments of the present application also provide a computer program product. The computer program product includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to execute the corresponding process in any of the artificial intelligence-based information mapping methods in the embodiments of the present application, For the sake of brevity, no further details will be given here.

An embodiment of the present application also provides a computer program. The computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to execute the corresponding process in any of the artificial intelligence-based information mapping methods in the embodiments of the present application, For the sake of brevity, no further details will be given here.

The above is a detailed introduction to an artificial intelligence-based information mapping method, client, server, equity incentive system and storage medium provided by the embodiments of this application. This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and the core idea of the present application; at the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the ideas of the present application. In summary, the content of this specification should not be construed as a limitation on this application.

Claims

An information mapping method based on artificial intelligence, characterized in that the method includes:

Obtain information content;

Determine the information topic tag of the information content based on a preset algorithm model;

Determine whether there is a target topic tag matching the information topic tag in the first image gallery, wherein the pictures in the first image gallery are marked with corresponding topic tags;

If there is a target topic tag matching the information topic tag in the first image gallery, the picture corresponding to the target topic tag is used as the cover image of the information content.
The information mapping method based on artificial intelligence according to claim 1, wherein the information topic tags of the information content are determined based on a preset algorithm model, including:

Obtain the information title text corresponding to the information text data in the information content, and process the information title text based on the preset algorithm model to determine the information topic tag of the information content.
The information mapping method based on artificial intelligence according to claim 2, characterized in that the preset algorithm model includes an entity extraction model and a keyword extraction model, and the information title is extracted based on the preset algorithm model. The text is processed to determine the information topic tag of the information content, including:

The information title text is processed based on an entity extraction model to obtain the entity type label of the information content, wherein the entity extraction model is used to extract company, industry and person name information in the information title text;

Process the information title text based on a keyword extraction model to obtain keyword tags for the information content, where the keyword extraction model is used to extract industry keyword information in the information title text;

According to the entity type tag and the keyword tag, the information topic tag of the information content is determined.
The artificial intelligence-based information mapping method according to claim 3, wherein determining the information topic tag of the information content based on the entity type tag and the keyword tag includes:

Generate a tag list according to the entity type tag and the keyword tag;

The first-ranked tag in the tag list is determined as the information topic tag of the information content.
The information mapping method based on artificial intelligence according to claim 1, characterized in that said obtaining information content includes:

Crawl the information content based on the information content source and store the information content into the information content library; or

In response to the storage request for the information content sent by the background management device, the information content is obtained, and the information content is stored in the information content library.
The information mapping method based on artificial intelligence according to claim 5, characterized in that after obtaining the information content, it also includes:

Conduct information review on the information content based on preset review rules, where the preset review rules at least include sensitive word matching and filtering rule verification;

If the information content does not match the sensitive word and the information content does not match the filtering rule, then the step of determining the information topic tag of the information content based on the preset algorithm model is performed.
The information mapping method based on artificial intelligence according to claim 6, characterized in that the method further includes:

If the information content hits a sensitive word, and/or the information content hits a filtering rule, a first prompt message indicating that the review fails is generated, and the information content and the first prompt message are sent to the backend management equipment.
The information mapping method based on artificial intelligence as claimed in claim 1, characterized in that the method further includes:

If there is no target topic tag matching the information topic tag in the first image gallery, a random picture is obtained from the second image gallery as the cover image of the information content, wherein the picture in the first image gallery is Hashtags not tagged.
The information mapping method based on artificial intelligence as claimed in claim 1, characterized in that the method also includes:

When a cover image configuration request for the information content sent by the background management device is received, the cover image of the information content is determined according to the configuration information carried in the cover image configuration request.
The information mapping method based on artificial intelligence as claimed in claim 1, characterized in that the method further includes:

When receiving a cover image display request for the information content sent by the client, sending the cover image of the information content to the client to display the cover image of the information content on the information preview interface of the client picture.
The information mapping method based on artificial intelligence according to claim 10, characterized in that the method further includes:

When receiving the triggering operation for the cover image sent by the client, send the information flow of the information content to the client, so as to display the information of the information content on the information preview interface of the client. Stream, the information stream includes at least one of information text data and information multimedia data corresponding to the information content.
The information mapping method based on artificial intelligence as claimed in claim 1, characterized in that the method further includes:

Acquire the picture record data within a preset period, wherein each picture record contains a pair of information topic tags and a method for determining the cover image corresponding to the information topic tag;

According to the picture recording data, count the frequency of occurrence of each information topic tag within the preset period to obtain tag statistics;

According to the picture matching record data, count the data proportion of the determination method of each cover image within the preset period to obtain picture matching statistics;

The label statistical data and the accompanying picture statistical data are sent to a backend management device, so that the label statistical data and the accompanying picture statistical data are displayed on the backend management device.
The information mapping method based on artificial intelligence according to claim 12, characterized in that the method further includes:

When an update request for the first gallery sent by the background management device is received, at least one of pictures and topic tags in the first gallery is updated according to the update request.
The information mapping method based on artificial intelligence according to claim 13, characterized in that the method further includes:

If it is detected that the same topic tag as the tag to be updated carried in the update request already exists in the first image gallery, a second prompt message indicating a duplication of tags is generated, and the update of the first image gallery is refused.
The information mapping method based on artificial intelligence according to claim 1, characterized in that if there is a target topic tag matching the information topic tag in the first gallery, the target topic tag will be The corresponding picture serves as the cover image of the information content, including:

If there is a target topic tag matching the information topic tag in the first gallery, and there are multiple pictures corresponding to the target topic tag, then select one picture from the multiple pictures corresponding to the target topic tag. Said as the cover image of the information content.
An information mapping device based on artificial intelligence, characterized in that the device includes:

Acquisition unit, used to obtain information content;

A determining unit, configured to determine the information topic tag of the information content based on a preset algorithm model;

A judging unit configured to judge whether there is a target topic tag matching the information topic tag in the first gallery, wherein the pictures in the first gallery are marked with corresponding topic tags;

A processing unit configured to, if there is a target topic tag matching the information topic tag in the first gallery, use the picture corresponding to the target topic tag as a cover image of the information content.
A computer device, characterized in that the computer device includes a processor and a memory, a computer program is stored in the memory, and the processor is used to execute the claims by calling the computer program stored in the memory. The artificial intelligence-based information mapping method described in any one of 1-15.
A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and the computer program is suitable for loading by a processor to execute the method based on any one of claims 1-15. Artificial intelligence information mapping method.
A computer program product, including computer instructions, characterized in that when the computer instructions are executed by a processor, the information mapping method based on artificial intelligence as described in any one of claims 1-15 is implemented.