A rim building method and device, computer equipment and storage medium

Abstract

The application belongs to the technical field of artificial intelligence, and relates to a wheel circle building method, comprising the following steps: obtaining a brand logo based on obtained vehicle type information, and creating an initial vehicle brand circle according to the brand logo; identifying the brand logo in a user binding request, and adding the user to the corresponding initial vehicle brand circle according to the brand logo; generating a brand label according to brand content corresponding to the brand logo, and labeling the brand content with the brand label; associating the brand content to the corresponding initial vehicle brand circle based on the brand label, and updating the state of the initial vehicle brand circle to take effect; and sorting all brand contents under the effective vehicle brand circle to obtain a built vehicle brand circle. The application also provides a wheel circle building device, computer equipment and a storage medium. In addition, the application also relates to the blockchain technology, and the binding request can be stored in the blockchain. The application can improve the activity and user stickiness of the platform, release the operation manpower, and improve the efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a wheel circle building method and device, computer equipment and a storage medium. BACKGROUND

[0002] With the development and popularization of computer technology and the rapid rise of the Internet, people gradually withdraw from traditional communication forms and invest a lot of time and effort in circle communication forms. The circle is similar to a forum, which has many advantages, such as real-time and wide-ranging. It is these outstanding advantages that make people express their opinions, discuss hot issues, and exchange technology and experience in their respective interest circles.

[0003] However, the current wheel circle needs manual participation in creation and configuration, which is high in operation cost and time-consuming, low in creation efficiency and prone to errors. SUMMARY

[0004] The purpose of the embodiments of the present application is to provide a wheel circle building method and device, computer equipment and a storage medium to solve the technical problems of high cost and time-consuming, low creation efficiency and error-prone in the prior art.

[0005] To solve the above technical problems, the embodiments of the present application provide a wheel circle building method, which adopts the following technical solutions:

[0006] Obtain all vehicle model information in a vehicle model database, obtain a brand identifier based on the vehicle model information, and create an initial vehicle brand circle according to the brand identifier;

[0007] In response to a binding request of a user joining the vehicle brand circle, identify the brand identifier in the binding request, and add the user to the corresponding initial vehicle brand circle according to the brand identifier;

[0008] Obtain brand content corresponding to the brand identifier, generate a brand tag according to the brand content, and label the brand tag for the brand content;

[0009] Associate the brand content to the corresponding initial vehicle brand circle based on the brand tag, update the status of the initial vehicle brand circle to take effect, and obtain an effective vehicle brand circle;

[0010] Sort all brand content under the effective vehicle brand circle to obtain a built vehicle brand circle.

[0011] Further, the step of generating a brand tag according to the brand content comprises:

[0012] The brand content is segmented into words to obtain the segmentation results;

[0013] A keyword set is obtained based on the word segmentation results;

[0014] The keyword set is input into the trained tag generation model to generate brand tags corresponding to the brand content.

[0015] Furthermore, prior to the step of inputting the keyword set into the trained tag generation model, the method further includes:

[0016] Obtain sample data, wherein the sample data includes brand sample tags and brand sample content;

[0017] Obtain the keyword sample set corresponding to the brand sample content, input the keyword sample set into the pre-built initial tag generation model, and output the predicted tag;

[0018] Calculate the loss function value based on the predicted label and the brand sample label;

[0019] The model parameters of the initial label generation model are adjusted according to the loss function value, and training continues until the convergence condition is met, at which point the final trained label generation model is output.

[0020] Furthermore, the step of sorting all the brand content under the effective car brand circle includes:

[0021] Obtain basic data and user behavior data of all brand content under the effective car brand circle, wherein the basic data includes time dimension data and content dimension data of brand content;

[0022] Calculate the basic content score for each brand's content based on the aforementioned basic data;

[0023] Based on the user behavior data, a user score is calculated for each piece of brand content;

[0024] Calculate a quality score for each piece of brand content based on the aforementioned basic data and user behavior data;

[0025] The basic content score, the user score, and the quality score are combined to obtain the final score for each brand content, and all brand content is sorted from highest to lowest according to the final score.

[0026] Furthermore, the step of calculating the user score for each piece of brand content based on the user behavior data includes:

[0027] Based on the user behavior data, the number of users who have accessed each piece of brand content is obtained, and the set of users who have had positive interaction behavior in the brand content is deduplicated to obtain the number of users who have had positive interaction behavior in the brand content.

[0028] Based on the number of users with positive interactive behavior and the number of users with access behavior, the Wilson algorithm is used to calculate the user score for each piece of brand content.

[0029] Furthermore, after the step of sorting all the brand content under the effective car brand circle to obtain the completed car brand circle, the method further includes:

[0030] In response to the poster's content posting request in the car brand's community, retrieve the content to be posted based on the content posting request;

[0031] The published content is input into the trained content quality scoring model for quality scoring, and the scoring result is obtained.

[0032] Based on the rating results, determine whether to display the published content;

[0033] When the score is greater than or equal to the preset score, the published content will be displayed in the car brand community.

[0034] Furthermore, after the step of displaying the published content within the car brand community, the following is also included:

[0035] Obtain user behavior data of the published content and the identity identifier of the poster;

[0036] The recommendation type is determined based on the user behavior data of the published content and the identity identifier;

[0037] Based on the recommendation type, the published content will be pushed to the corresponding recommendation location.

[0038] To address the aforementioned technical problems, this application also provides a wheel rim assembly device, which employs the following technical solution:

[0039] The acquisition module is used to acquire all vehicle information in the vehicle database, obtain brand identifiers based on the vehicle information, and create an initial vehicle brand circle based on the brand identifiers.

[0040] The binding module is used to respond to a user's binding request to join the car brand circle, identify the brand identifier in the binding request, and add the user to the corresponding initial car brand circle according to the brand identifier;

[0041] The labeling module is used to obtain the brand content corresponding to the brand logo, generate brand tags based on the brand content, and label the brand content with the brand tags.

[0042] The association module is used to associate the brand content with the corresponding initial car brand circle based on the brand tag, and update the status of the initial car brand circle to be active, so as to obtain the active car brand circle.

[0043] The sorting module is used to sort all the brand content under the effective car brand circle to obtain the completed car brand circle.

[0044] To address the aforementioned technical problems, this application also provides a computer device that employs the following technical solution:

[0045] The computer device includes a memory and a processor. The memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the wheel rim assembly method described above.

[0046] To address the aforementioned technical problems, this application also provides a computer-readable storage medium, employing the technical solution described below:

[0047] The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the wheel rim assembly method described above.

[0048] Compared with the prior art, the embodiments of this application have the following main advantages:

[0049] This application obtains all vehicle model information from the vehicle model database, acquires brand identifiers based on the vehicle model information, and creates initial car brand circles based on the brand identifiers. Responding to user binding requests to join car brand circles, it identifies the brand identifiers in the binding requests and adds users to the corresponding initial car brand circles based on the brand identifiers. It then acquires the brand content corresponding to the brand identifiers, generates brand tags based on the brand content, and labels the brand content with brand tags. Based on the brand tags, it associates the brand content with the corresponding initial car brand circles and updates the status of the initial car brand circles to "active," resulting in active car brand circles. Finally, it sorts all brand content under the active car brand circles to obtain the completed car brand circles. This application can create corresponding car brand circles in batches based on the vehicle model database and bind them to users' car brands, thereby attracting users with similar interests and improving platform activity and user stickiness. Batch extraction of corresponding brand content based on brand identifiers can free up operational manpower, and intelligent tagging can improve the efficiency of content updates. Furthermore, sorting the brand content allows users to easily view brand content of interest within the car brand circles, increasing content penetration on the platform. Attached Figure Description

[0050] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is an exemplary system architecture diagram to which this application can be applied;

[0052] Figure 2 This is a flowchart of one embodiment of the wheel rim assembly method according to this application;

[0053] Figure 3 yes Figure 2 A flowchart of a specific implementation of step S203;

[0054] Figure 4 This is a schematic diagram of a structure of one embodiment of the wheel rim assembly device according to this application;

[0055] Figure 5 This is a schematic diagram of the structure of one embodiment of the computer device according to this application. Detailed Implementation

[0056] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0057] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0058] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0059] This application provides a method for constructing a vehicle wheel, which involves artificial intelligence and can be applied to, for example... Figure 1 In the system architecture 100 shown, the system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. The network 104 is used as a medium to provide a communication link between the terminal devices 101, 102, and 103 and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables, etc.

[0060] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 101, 102, and 103, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc.

[0061] Terminal devices 101, 102, and 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III), MP4 players (Moving Picture Experts Group Audio Layer IV), laptops, and desktop computers, etc.

[0062] Server 105 can be a server that provides various services, such as a backend server that supports the pages displayed on terminal devices 101, 102, and 103.

[0063] It should be noted that the wheel rim assembly method provided in this application embodiment is generally executed by a server / terminal device, and correspondingly, the wheel rim assembly device is generally located in the server / terminal device.

[0064] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0065] Continue to refer to Figure 2 The flowchart illustrates an embodiment of the wheel assembly method according to this application, including the following steps:

[0066] Step S201: Obtain all vehicle information from the vehicle database, obtain brand identifiers based on the vehicle information, and create an initial vehicle brand circle based on the brand identifiers.

[0067] In this embodiment, the vehicle model database includes several existing vehicle models and their corresponding model information. The model information includes, but is not limited to, brand identifiers, model numbers, configurations, and colors. Brand identifiers are extracted from the model information. A brand identifier refers to the manufacturer of the vehicle, such as BMW, Mercedes-Benz, or Volkswagen; model numbers include, but are not limited to, sports cars, SUVs, family cars, and sedans.

[0068] The initial car brand circle for that brand is automatically created based on the brand logo.

[0069] Step S202: In response to the user's binding request to join the car brand circle, identify the brand identifier in the binding request, and add the user to the corresponding initial car brand circle according to the brand identifier.

[0070] In this embodiment, a binding request to join a car brand circle is received from a user. The binding request is parsed to obtain the user's brand identifier. The user is then added to the corresponding car brand circle based on the brand identifier. By associating the car brand circle with the user's car brand, the platform's activity and user stickiness can be improved.

[0071] It should be emphasized that, to further ensure the privacy and security of the binding request, the binding request can also be stored in a node of a blockchain.

[0072] The blockchain referred to in this application is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0073] Step S203: Obtain the brand content corresponding to the brand logo, generate brand tags based on the brand content, and affix brand tags to the brand content.

[0074] In this embodiment, the corresponding brand content is automatically retrieved based on the brand identifier. The brand content can be car brand content purchased from suppliers, and the types include, but are not limited to, articles and videos.

[0075] In some alternative implementations, see [link to relevant documentation]. Figure 3 As shown, the steps for generating brand tags based on brand content include:

[0076] Step S301: Perform word segmentation on the brand content to obtain the word segmentation results;

[0077] Step S302: Obtain the keyword set based on the word segmentation results;

[0078] Step S303: Input the keyword set into the trained tag generation model to generate brand tags corresponding to the brand content.

[0079] When the brand content is an article, the entire text of the article is extracted, and the text is segmented into words. A keyword set is obtained based on the segmentation results. This keyword set is then used as input to a tag generation model, which outputs the brand tags corresponding to the article. Brand tags represent different dimensions of the brand content, such as brand identity, content category, and content type.

[0080] The word segmentation process can be performed using a word segmenter, such as Jieba Chinese Word Segmenter, Hanlp Word Segmenter, Foolnltk Word Segmenter, and Pullword Online Word Segmentation Engine. After word segmentation, stop words in the segmentation results can be filtered out to obtain a set of candidate keywords. The word weight, word length, word frequency, and position information of the candidate keywords in the sentences containing them are obtained. Based on the word weight, word length, word frequency, and position information, the word importance of the candidate keywords is calculated. The candidate keywords are then sorted from highest to lowest word importance, and the top N candidate keywords (where N is a positive integer) are selected as the keywords.

[0081] In this embodiment, the word importance is calculated using the following formula:

[0082]

[0083] Where f is the word importance, α, β, and γ are hyperparameters, TFIDFi is the word weight, lw is the word length, minpos is the position information, and C is the word presence.

[0084] In this embodiment, the TF-IDF algorithm (term frequency–inverse document frequency) is used to calculate the word weights of candidate keywords.

[0085] When the brand content is a video and the video only has subtitles, feature extraction is performed on each frame of the target video to obtain a feature map. Text recognition is performed on the feature map to obtain the text content. The obtained text content is then processed through steps S301 to S303 to obtain the brand label.

[0086] When the brand content is a video and the video only contains audio, perform speech recognition on the video to obtain the speech recognition text, and then perform steps S301 to S303 on the speech recognition text to obtain the brand label.

[0087] If the brand content is video, and the video includes audio and subtitles, the brand tags can be obtained using any one or a combination of the above methods.

[0088] In this embodiment, the structure of the tag generation model can be an LSTM (Long Short-Term Memory) model, a Transformer model, a BERT model, etc. The tag generation model can quickly and accurately determine the corresponding tags based on the brand content, while saving manpower and resources.

[0089] Step S204: Associate the brand content with the corresponding initial car brand circle based on the brand tag, and update the status of the initial car brand circle to active, thus obtaining the active car brand circle.

[0090] In this embodiment, brand identifiers can be obtained based on brand tags, thereby associating the corresponding brand content with the initial car brand circle.

[0091] After the initial car brand circle is created, the system automatically retrieves car brand content and iterates through all created initial car brand circles. If an initial car brand circle has no content data, its status is set to invalid by default. When matching content data is found, the circle is automatically updated to an active status. In active car brand circles, users can participate in discussions and exchanges, such as posting and answering questions.

[0092] Step S205: Sort all brand content under the effective car brand circle to obtain the completed car brand circle.

[0093] Specifically, the process involves acquiring basic data and user behavior data for all brand content within the relevant car brand circles. The basic data includes both time-based and content-based data for the brand content. A basic content score is calculated for each brand's content based on the basic data. A user score is calculated for each brand's content based on the user behavior data. A quality score is calculated for each brand's content based on both the basic data and the user behavior data. The basic content score, user score, and quality score are then combined to obtain the final score for each brand's content. Finally, all brand content is sorted from highest to lowest based on the final score.

[0094] The brand content's time-dimensional data includes content creation time, content publication time, and last update time; the brand content's content-dimensional data includes content popularity (content views within the preset number of days) and content length. User behavior data includes, but is not limited to, page views, likes, comments, favorites, user activity, and basic vehicle information. Basic vehicle information includes vehicle brand, model, and price.

[0095] In this embodiment, the step of calculating the user score for each brand's content based on user behavior data includes:

[0096] Based on user behavior data, the number of users who have accessed each brand's content is obtained, and the set of users who have had positive interaction with the brand's content is deduplicated to obtain the number of users who have had positive interaction with the brand's content.

[0097] Based on the number of users with positive interactive behavior and the number of users with access behavior, the Wilson algorithm is used to calculate the user score for each brand's content.

[0098] Positive interaction behaviors include data such as browsing, liking, saving, commenting, or sharing. In this embodiment, users with no positive interaction behaviors among all users who have access behavior are considered as users with negative interaction behaviors. The specific formula for calculating user scores using the Wilson algorithm is as follows:

[0099]

[0100] In the formula, p represents the number of users with positive interactive behavior divided by the number of users with access behavior; z α is the quantile (parameter) of the normal distribution, usually taken as 1.96; n represents the number of users with access behavior.

[0101] In some implementations of this embodiment, basic data and user behavior data are input into a trained quality scoring model for each brand's content to obtain a quality score.

[0102] Specifically, the quality scoring model can employ either a logistic regression model or BERT (Bidirectional Encoder Representation from Transformers, a pre-trained language representation model). Taking the logistic regression model as an example, the quality scoring model is trained through the following steps:

[0103] Obtain a training set, which includes sample content labeled with quality scores; obtain the basic data and user behavior data corresponding to each sample content as sample features; input the sample features into a pre-built logistic regression model for calculation to obtain the predicted content score; calculate the loss value based on the predicted content score and the labeled quality score; adjust the model parameters based on the loss value and continue training until the model converges, and output the quality score model.

[0104] The calculated basic content score, user score, and quality score are combined to obtain the final score for each brand's content. The calculation formula is as follows:

[0105] s = (s1 * w1 + s2 * w2) × quality;

[0106] In the formula, s1 represents the basic content score; s2 represents the user score; w1 and w2 represent the weights, which can be preset according to the actual situation; and quality represents the quality score.

[0107] In this embodiment, by integrating the basic content score, user score, and quality score, the ranking of brand content with different quality scores can be effectively differentiated, making the ranking more accurate.

[0108] Once the car brand communities are set up, they can be showcased to users, who can then engage in discussions and exchanges within those communities.

[0109] This application creates corresponding car brand circles in batches based on a car model database, binding them to users' car brands to attract like-minded users, thereby increasing platform activity and user stickiness. Extracting relevant brand content in batches based on brand identifiers frees up operational manpower, and intelligent tagging improves content update efficiency. Furthermore, sorting brand content allows users to easily find content of interest within car brand circles, increasing content penetration on the platform.

[0110] In some optional implementations of this embodiment, the method further includes the following step before inputting the keyword set into the trained tag generation model:

[0111] Obtain sample data, which includes brand sample tags and brand sample content;

[0112] Obtain the keyword sample set corresponding to the brand sample content, input the keyword sample set into the pre-built initial tag generation model, and output the predicted tag;

[0113] The loss function value is calculated based on the predicted labels and brand sample labels;

[0114] Adjust the model parameters of the initial label generation model based on the loss function value, and continue training until the convergence condition is met, and output the final trained label generation model.

[0115] Among them, the methods for judging convergence include: (1) calculating the loss function value in the two iterations before and after. If the loss function value is still changing, continue iterative training; if the loss function value does not change significantly, the model can be considered to have converged; (2) setting the number of iterations in advance, performing iterative training according to the preset number of iterations, and considering the model to have converged when the preset number of iterations is reached.

[0116] For example, the initial label generation model has the specific structure of a BERT model. The BERT model includes an input layer, a feature extraction layer, and a softmax layer. Specifically, the keyword sample set is transformed into a vector through the input layer to obtain the feature vector data corresponding to the keyword sample set; the feature vector data is input into the feature extraction layer for feature extraction to obtain the keyword features; the keyword features are input into the softmax layer for calculation to obtain the predicted probability of each label corresponding to the sample data, and the predicted label is output based on the predicted probability.

[0117] The feature extraction layer consists of a stacked 12-layer network structure. Each layer uses the encoder part of a Transformer. Each encoder layer includes a multi-head attention mechanism and a position-wise-feed-forward network.

[0118] This application utilizes a tag generation model training method to effectively train a tag generation model that can quickly and accurately determine the corresponding brand tags based on brand content.

[0119] In some optional implementations, after sorting all brand content under the aforementioned car brand circle to obtain the completed car brand circle, the following steps are also included:

[0120] Responding to users' content publishing requests in the car brand community, retrieve the content to be published based on the content publishing request;

[0121] The published content is input into the trained content quality scoring model to perform a quality score, and the score result is obtained.

[0122] Whether to display the published content is determined based on the scoring results;

[0123] When the score is greater than or equal to the preset score, the published content will be displayed in the car brand's circle.

[0124] The content published includes articles, videos, Q&A, and posts. When determining whether a user-published content is a post, its quality needs to be scored. This scoring can be achieved using a content quality scoring model or through manual rule-based judgment.

[0125] In this embodiment, the content quality scoring model can be trained by obtaining 1,000 posts of different qualities provided by the operator as a training set, outputting post scores, and determining the quality level of the posts based on different score levels.

[0126] Quality levels include: Excellent posts, with a score of 8 or higher (out of 10); Medium posts, with a score of 5 or higher but less than 8; and Poor posts, with a score of less than 5.

[0127] Posts rated as high quality or medium quality can be displayed on the front end. High-quality posts can also be automatically recommended to communities and featured topics. Poor quality posts are not displayed on the front end and can only be viewed by the user themselves.

[0128] By rating published content and determining whether to display it on the front end based on the rating results, we can ensure content quality and maintain a positive environment.

[0129] In some alternative implementations, following the steps of showcasing the published content within the car brand community, the following is also included:

[0130] Obtain user behavior data and poster identification for published content;

[0131] Recommendation types are determined based on user behavior data and identity identifiers of the published content;

[0132] Based on the recommendation type, the published content will be pushed to the corresponding recommendation location.

[0133] User behavior data includes the number of reads, likes, and comments; user identities include circle owners, experts, and ordinary users. Recommendation types include "Highly Recommended," "Recommended," and "Generally Recommended," and recommendation placements include featured topics, featured circles, and the circle square. "Highly Recommended" corresponds to featured topics, "Recommended" to featured circles, and "Generally Recommended" to the circle square.

[0134] In this embodiment, the recommendation criteria that the published content meets are determined based on user behavior data and identity identifiers, and the recommendation type is determined based on the recommendation criteria.

[0135] For example, the recommended conditions are shown in Table 1 below.

[0136] Table 1

[0137]

[0138]

[0139] In this embodiment, by recommending published content, users can directly access brand content of interest within the brand community, which can increase the penetration rate of content on the platform and improve user experience.

[0140] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0141] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware with computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0142] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0143] Further reference Figure 4 As a response to the above Figure 2 The present application provides an embodiment of a wheel rim assembly device to implement the method shown. This embodiment of the device is similar to... Figure 2 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0144] like Figure 4As shown, the wheel rim assembly device 400 described in this embodiment includes: an acquisition module 401, a binding module 402, an annotation module 403, an association module 404, and a sorting module 405. Wherein:

[0145] The acquisition module 401 is used to acquire all vehicle information in the vehicle database, obtain brand identifiers based on the vehicle information, and create an initial vehicle brand circle based on the brand identifiers;

[0146] The binding module 402 is used to respond to a user's binding request to join the car brand circle, identify the brand identifier in the binding request, and add the user to the corresponding initial car brand circle according to the brand identifier;

[0147] The labeling module 403 is used to obtain the brand content corresponding to the brand logo, generate brand tags based on the brand content, and label the brand content with the brand tags;

[0148] The association module 404 is used to associate the brand content with the corresponding initial car brand circle based on the brand tag, and update the status of the initial car brand circle to be effective, so as to obtain the effective car brand circle.

[0149] The sorting module 405 is used to sort all the brand content under the effective car brand circle to obtain the completed car brand circle.

[0150] It should be emphasized that, to further ensure the privacy and security of the binding request, the binding request can also be stored in a node of a blockchain.

[0151] Based on the aforementioned car brand community building mechanism, corresponding car brand communities can be created in batches according to the car model database and linked to users' car brands, thereby identifying users with similar interests and improving platform activity and user stickiness. Batch extraction of relevant brand content based on brand identifiers can free up operational manpower, and intelligent tagging applications can improve the efficiency of content updates. Furthermore, sorting brand content allows users to easily find content of interest within car brand communities, increasing content penetration on the platform.

[0152] In this embodiment, the annotation module 403 includes a word segmentation submodule, a keyword submodule, and a generation submodule, wherein:

[0153] The word segmentation submodule is used to segment the brand content into words and obtain the segmentation results;

[0154] The keyword submodule is used to obtain a set of keywords based on the word segmentation results;

[0155] The generation submodule is used to input the keyword set into the trained tag generation model to generate brand tags corresponding to the brand content.

[0156] By using a tag generation model, the corresponding tags can be quickly and accurately determined based on brand content, while saving manpower and resources.

[0157] In some optional implementations, the generation submodule includes an acquisition unit, a generation unit, a calculation unit, and an adjustment unit, wherein:

[0158] The acquisition unit is used to acquire sample data, wherein the sample data includes brand sample tags and brand sample content;

[0159] The generation unit is used to obtain the keyword sample set corresponding to the brand sample content, input the keyword sample set into the pre-built initial tag generation model, and output the predicted tag;

[0160] The calculation unit is used to calculate the loss function value based on the predicted label and the brand sample label;

[0161] The adjustment unit is used to adjust the model parameters of the initial label generation model according to the loss function value, and continue training until the convergence condition is met, and output the finally trained label generation model.

[0162] By using the training method of the tag generation model, a tag generation model can be effectively trained to quickly and accurately determine the corresponding brand tags based on the brand content.

[0163] In this embodiment, the sorting module 405 includes an acquisition submodule, a content rating submodule, a user rating submodule, a quality rating submodule, and a sorting submodule, wherein:

[0164] The acquisition submodule is used to acquire basic data and user behavior data of all brand content under the effective car brand circle, wherein the basic data includes time dimension data and content dimension data of brand content;

[0165] The content scoring submodule is used to calculate the basic content score for each brand's content based on the basic data;

[0166] The user rating submodule is used to calculate a user score for each piece of brand content based on the user behavior data.

[0167] The quality scoring submodule is used to calculate the quality score of each piece of brand content based on the basic data and the user behavior data;

[0168] The sorting submodule is used to integrate the basic content score, the user score, and the quality score to obtain the final score of each brand content, and sort all the brand content from high to low according to the final score.

[0169] By integrating the basic content score, user score, and quality score, the ranking of brand content with different quality scores can be effectively differentiated, making the ranking more accurate.

[0170] In some optional implementations of this embodiment, the user rating submodule is further used for:

[0171] Based on the user behavior data, the number of users who have accessed each piece of brand content is obtained, and the set of users who have had positive interaction behavior in the brand content is deduplicated to obtain the number of users who have had positive interaction behavior in the brand content.

[0172] Based on the number of users with positive interactive behavior and the number of users with access behavior, the Wilson algorithm is used to calculate the user score for each piece of brand content.

[0173] In some alternative implementations, the wheel assembly device 400 also includes a content publishing module for performing the following steps:

[0174] In response to the poster's content posting request in the car brand's community, retrieve the content to be posted based on the content posting request;

[0175] The published content is input into the trained content quality scoring model for quality scoring, and the scoring result is obtained.

[0176] Based on the rating results, determine whether to display the published content;

[0177] When the score is greater than or equal to the preset score, the published content will be displayed in the car brand community.

[0178] By rating published content and determining whether to display it on the front end based on the rating results, we can ensure content quality and maintain a positive environment.

[0179] In some optional implementations, the wheel assembly 400 also includes a recommended module for:

[0180] Obtain user behavior data of the published content and the identity identifier of the poster;

[0181] The recommendation type is determined based on the user behavior data of the published content and the identity identifier;

[0182] Based on the recommendation type, the published content will be pushed to the corresponding recommendation location.

[0183] By recommending published content, users can directly access brand content of interest within the brand community, which can increase the penetration rate of content on the platform and improve user experience.

[0184] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 5 , Figure 5 This is a basic structural block diagram of the computer device in this embodiment.

[0185] The computer device 5 includes a memory 51, a processor 52, and a network interface 53 that are interconnected via a system bus. It should be noted that only the computer device 5 with components 51-53 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0186] The computer device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The computer device can interact with the user via a keyboard, mouse, remote control, touchpad, or voice control.

[0187] The memory 51 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 51 may be an internal storage unit of the computer device 5, such as the hard disk or memory of the computer device 5. In other embodiments, the memory 51 may also be an external storage device of the computer device 5, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 5. Of course, the memory 51 may include both the internal storage unit and its external storage device of the computer device 5. In this embodiment, the memory 51 is typically used to store the operating system and various application software installed on the computer device 5, such as computer-readable instructions for wheel rim construction methods. In addition, the memory 51 can also be used to temporarily store various types of data that have been output or will be output.

[0188] In some embodiments, the processor 52 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 52 is typically used to control the overall operation of the computer device 5. In this embodiment, the processor 52 is used to execute computer-readable instructions stored in the memory 51 or to process data, for example, to execute computer-readable instructions for the wheel assembly method.

[0189] The network interface 53 may include a wireless network interface or a wired network interface, which is typically used to establish communication connections between the computer device 5 and other electronic devices.

[0190] This embodiment implements the steps of the car circle building method described above by executing computer-readable instructions stored in the memory through the processor. By creating corresponding car brand circles in batches according to the car model database and binding them with users' car brands, users with similar interests can be identified, thereby improving platform activity and user stickiness. Batch extraction of corresponding brand content based on brand identifiers can free up operational manpower, and intelligent tagging applications can improve the efficiency of content updates. In addition, sorting brand content can make it convenient for users to directly view brand content of interest in car brand circles, thereby increasing the penetration rate of content on the platform.

[0191] This application also provides another implementation method, namely, providing a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to perform the steps of the car circle building method described above. By creating corresponding car brand circles in batches based on a car model database and binding them to users' car brands, users with similar interests can be identified, thereby improving platform activity and user stickiness. Batch extraction of corresponding brand content based on brand identifiers can free up operational manpower, and intelligent tagging applications can improve the efficiency of content updates. In addition, sorting brand content can make it convenient for users to directly view brand content of interest in car brand circles, thereby increasing the penetration rate of content on the platform.

[0192] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0193] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A method for assembling a bicycle wheel, characterized in that, Includes the following steps: Obtain all vehicle information from the vehicle database, obtain brand identifiers based on the vehicle information, and create an initial vehicle brand circle based on the brand identifiers; In response to a user's binding request to join the car brand circle, the system identifies the brand identifier in the binding request and adds the user to the corresponding initial car brand circle based on the brand identifier. Obtain the brand content corresponding to the brand logo, generate brand tags based on the brand content, and label the brand content with the brand tags; The step of generating brand tags based on the brand content includes: performing word segmentation on the brand content to obtain word segmentation results; obtaining a keyword set based on the word segmentation results; inputting the keyword set into a trained tag generation model to generate brand tags corresponding to the brand content; wherein, the tag generation model includes an input layer, a feature extraction layer, and a softmax layer, wherein the feature extraction layer includes a stacked 12-layer network structure, each layer including a multi-head attention mechanism and a feedforward neural network; the step of inputting the keyword set into the trained tag generation model to generate brand tags corresponding to the brand content includes: performing vector transformation on the keyword set through the input layer to obtain feature vector data corresponding to the keyword set; inputting the feature vector data into the feature extraction layer for feature extraction to obtain keyword features; inputting the keyword features into the softmax layer for calculation to obtain the predicted probability of each tag corresponding to the keyword set, and outputting brand tags based on the predicted probabilities; Based on the brand tag, the brand content is associated with the corresponding initial car brand circle, and the status of the initial car brand circle is updated to active, thus obtaining an active car brand circle; Sort all the brand content under the effective car brand circle to obtain the completed car brand circle.

2. The method for assembling a bicycle wheel according to claim 1, characterized in that, The method further includes the following steps before inputting the keyword set into the trained tag generation model: Obtain sample data, wherein the sample data includes brand sample tags and brand sample content; Obtain the keyword sample set corresponding to the brand sample content, input the keyword sample set into the pre-built initial tag generation model, and output the predicted tag; Calculate the loss function value based on the predicted label and the brand sample label; The model parameters of the initial label generation model are adjusted according to the loss function value, and training continues until the convergence condition is met, at which point the final trained label generation model is output.

3. The method for assembling a bicycle wheel according to claim 1, characterized in that, The step of sorting all the brand content under the effective car brand circle includes: Obtain basic data and user behavior data of all brand content under the effective car brand circle, wherein the basic data includes time dimension data and content dimension data of brand content; Calculate the basic content score for each brand's content based on the aforementioned basic data; Based on the user behavior data, a user score is calculated for each piece of brand content; Calculate a quality score for each piece of brand content based on the aforementioned basic data and user behavior data; The basic content score, the user score, and the quality score are combined to obtain the final score for each brand content, and all brand content is sorted from highest to lowest according to the final score.

4. The method for assembling a bicycle wheel according to claim 3, characterized in that, The step of calculating the user score for each piece of brand content based on the user behavior data includes: Based on the user behavior data, the number of users who have accessed each piece of brand content is obtained, and the set of users who have had positive interaction behavior in the brand content is deduplicated to obtain the number of users who have had positive interaction behavior in the brand content. Based on the number of users with positive interactive behavior and the number of users with access behavior, the Wilson algorithm is used to calculate the user score for each piece of brand content.

5. The method for assembling a bicycle wheel according to claim 1, characterized in that, After the step of sorting all the brand content under the effective car brand circle to obtain the completed car brand circle, the method further includes: In response to the poster's content posting request in the car brand's community, retrieve the content to be posted based on the content posting request; The published content is input into the trained content quality scoring model for quality scoring, and the scoring result is obtained. Based on the rating results, determine whether to display the published content; When the score is greater than or equal to the preset score, the published content will be displayed in the car brand community.

6. The method for assembling a bicycle wheel according to claim 5, characterized in that, Following the step of displaying the published content within the car brand community, the following is also included: Obtain user behavior data of the published content and the identity identifier of the poster; The recommendation type is determined based on user behavior data of the published content and the identity identifier; Based on the recommendation type, the published content will be pushed to the corresponding recommendation location.

7. A wheel rim assembly device, characterized in that, include: The acquisition module is used to acquire all vehicle information in the vehicle database, obtain brand identifiers based on the vehicle information, and create an initial vehicle brand circle based on the brand identifiers. The binding module is used to respond to a user's binding request to join the car brand circle, identify the brand identifier in the binding request, and add the user to the corresponding initial car brand circle according to the brand identifier; The labeling module is used to obtain the brand content corresponding to the brand logo, generate brand tags based on the brand content, and label the brand content with the brand tags. The association module is used to associate the brand content with the corresponding initial car brand circle based on the brand tag, and update the status of the initial car brand circle to be active, so as to obtain the active car brand circle. The sorting module is used to sort all the brand content under the effective car brand circle to obtain the completed car brand circle. The annotation module includes a word segmentation submodule, a keyword submodule, and a generation submodule, wherein: The word segmentation submodule is used to segment the brand content into words and obtain the segmentation results; The keyword submodule is used to obtain a set of keywords based on the word segmentation results; The generation submodule is used to input the keyword set into a trained tag generation model to generate brand tags corresponding to the brand content. The tag generation model includes an input layer, a feature extraction layer, and a softmax layer. The feature extraction layer comprises a stacked 12-layer network structure, with each layer including a multi-head attention mechanism and a feedforward neural network. The step of inputting the keyword set into the trained tag generation model to generate brand tags corresponding to the brand content includes: performing vector transformation on the keyword set through the input layer to obtain feature vector data corresponding to the keyword set; inputting the feature vector data into the feature extraction layer for feature extraction to obtain keyword features; inputting the keyword features into the softmax layer for calculation to obtain the predicted probability of each tag corresponding to the keyword set; and outputting brand tags based on the predicted probabilities.

8. A computer device comprising a memory and a processor, the memory storing computer-readable instructions, wherein the processor, when executing the computer-readable instructions, implements the steps of the wheel rim assembly method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the wheel rim assembly method as described in any one of claims 1 to 6.

Images