A service recommendation method

Abstract

The application discloses a service recommendation method, which is used to improve the effectiveness of recommending service content to a user. The scheme provided by the application comprises the following steps: acquiring position information reported by a user terminal, wherein the user terminal is preinstalled with a super SIM card; if the position information indicates that the user terminal is located in a preset electronic fence, instructing the user terminal to display a super SIM message, wherein the super SIM message comprises information of optional service items corresponding to the preset electronic fence; performing identity verification through the super SIM card based on an interaction operation on the optional service items received by the user terminal; and instructing the user terminal to display service content matched with identity information of the super SIM card through a 5G message based on an identity verification result, wherein the service content is service content predicted by a pre-trained service content recommendation model based on the identity information.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication, in particular to a service recommendation method. BACKGROUND

[0002] In the field of communication services, the types of communication services are increasingly diversified. Users can see a large amount of information through user terminals, and it is difficult for service information matched with the users to effectively reach the users.

[0003] If the user actively searches for services, although the required service functions can be used, the user needs to perform active and tedious operations. If service recommendation is performed to the user, the user passively views service information, and there may be a case that the recommended service does not match the user. In this case, not only the user cannot conveniently use the required service, but also unnecessary information interference may be caused to the user.

[0004] How to improve the effectiveness of recommending service content to the user is a technical problem to be solved by the present application. SUMMARY

[0005] The purpose of the embodiments of the present application is to provide a service recommendation method to improve the effectiveness of recommending service content to the user.

[0006] In a first aspect, a service recommendation method is provided, comprising:

[0007] Obtaining position information reported by a user terminal, the user terminal being pre-installed with a super SIM card;

[0008] If the position information indicates that the user terminal is located within a preset electronic fence, instructing the user terminal to display a super SIM message, the super SIM message including information of selectable service items corresponding to the preset electronic fence;

[0009] Based on an interaction operation received by the user terminal on the selectable service items, performing identity verification through the super SIM card;

[0010] Based on an identity verification result, instructing the user terminal to display service content matched with identity information of the super SIM card through a 5G message, wherein the service content is service content predicted by a pre-trained service content recommendation model based on the identity information.

[0011] In a second aspect, a service recommendation device is provided, comprising:

[0012] An obtaining module, configured to obtain position information reported by a user terminal, the user terminal being pre-installed with a super SIM card;

[0013] The indication module indicates the user terminal to display a super-SIM message if the position information represents that the user terminal is located in a preset electronic fence, the super-SIM message including information of optional service items corresponding to the preset electronic fence;

[0014] The verification module performs identity verification through the super-SIM card based on the interaction operation of the user terminal on the optional service items;

[0015] The display module indicates the user terminal to display service content matching the identity information of the super-SIM card through a 5G message based on the identity verification result, wherein the service content is service content predicted by a pre-trained service content recommendation model based on the identity information.

[0016] In a third aspect, an electronic device is provided, which includes a processor, a memory, and a computer program stored in the memory and executable on the processor, and when the computer program is executed by the processor, the steps of the method of the first aspect are implemented.

[0017] In a fourth aspect, a computer readable storage medium is provided, which stores a computer program, and when the computer program is executed by a processor, the steps of the method of the first aspect are implemented.

[0018] In a fifth aspect, a computer program product is provided, which includes a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform some or all of the steps of the method of the first aspect.

[0019] In the embodiment of the present application, first, the location information reported by the user terminal is acquired, the user terminal is pre-installed with a super SIM card; if the location information indicates that the user terminal is located in a preset electronic fence, the user terminal is instructed to display a super SIM message, wherein the super SIM message includes information of optional service items corresponding to the preset electronic fence; then, based on the interaction operation on the optional service items received by the user terminal, identity verification is performed through the super SIM card; finally, based on the identity verification result, the user terminal is instructed to display service content matched with the identity information of the super SIM card through a 5G message, wherein the service content is service content predicted by a pre-trained service content recommendation model based on the identity information. Through the present scheme, the user is accurately provided with information of optional service items based on the location information reported by the user terminal, and then reliable and real identity verification is performed through the super SIM card. Subsequently, the display is performed through a 5G message, which can improve the richness of message content, realize multi-form display of service content matched with the identity information, and optimize the readability of service content. The service content is predicted by a pre-trained service content recommendation model based on the identity information, which can improve the matching degree of the displayed service content and the identity information of the super SIM card, and improve the attractiveness of the recommended service to the user. BRIEF DESCRIPTION OF DRAWINGS

[0020] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, and do not constitute improper limitations on the present application. In the drawings:

[0021] Figure 1a is one of the flow diagrams of a service recommendation method according to an embodiment of the present application;

[0022] Figure 1b is a system structure function diagram of a service platform for executing the service recommendation method according to an embodiment of the present application;

[0023] Figure 1c is a scene application diagram of a service platform for executing the service recommendation method according to an embodiment of the present application;

[0024] Figure 2 is the second flow diagram of a service recommendation method according to an embodiment of the present application;

[0025] Figure 3a is the third flow diagram of a service recommendation method according to an embodiment of the present application;

[0026] Figure 3b is a model training flow diagram of a service recommendation method according to an embodiment of the present application;

[0027] Figure 4Figure 4 is a flowchart of a service recommendation method according to an embodiment of the present application;

[0028] Figure 5 Figure 5 is a flowchart of a service recommendation method according to an embodiment of the present application;

[0029] Figure 6 Figure 6 is a flowchart of a service recommendation method according to an embodiment of the present application;

[0030] Figure 7 Figure 7 is a flowchart of a service recommendation method according to an embodiment of the present application;

[0031] Figure 8a Figure 8 is a flowchart of a service recommendation method according to an embodiment of the present application;

[0032] Figure 8b Figure 9 is a flowchart of a service recommendation method according to an embodiment of the present application;

[0033] Figure 9a Figure 10 is a flowchart of a service recommendation method according to an embodiment of the present application;

[0034] Figure 9b Figure 11 is a flowchart of a service recommendation method according to an embodiment of the present application;

[0035] Figure 10 Figure 12 is a flowchart of a service recommendation method according to an embodiment of the present application;

[0036] Figure 11 Figure 13 is a structural diagram of a service recommendation device according to an embodiment of the present application. DETAILED DESCRIPTION

[0037] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without any creative work fall within the scope of protection of the present application. The figure numbers in the present application are only used to distinguish each step in the scheme, and are not used to limit the execution order of each step, and the specific execution order is subject to the description in the specification.

[0038] In a service recommendation application scenario, the related information of service recommendation can be sent to the user through an application (Application, APP), a Web (World Wide Web, global wide area network) page, and the like. Specifically, after the geographic location or IP (Internet Protocol, Internet Protocol) address permission of the user terminal is obtained based on an LBS (Location Based Services, LBS) recommendation system, the user can be reached through in-site messages, PUSH (application message) push, and end-in-stream media advertising.

[0039] The defects of these service recommendation methods are that, on the one hand, the push message can be limited by the message permission and cannot be effectively displayed to the user in time. On the other hand, it is difficult to implement targeted service recommendation for the user, such as obtaining the identity information of the user and then determining the user demand, which often requires the user to perform tedious operations. Moreover, the display effect of the application program background pop-up window, short message and the like is limited. As a result, the click conversion rate of actual service recommendation is low.

[0040] In order to solve the problems existing in the related art, an embodiment of the present application provides a service recommendation method. The present scheme can be executed by a service platform, such as Figure 1a as shown, comprising the following steps:

[0041] S11: Obtain the position information reported by the user terminal, wherein the user terminal is pre-installed with a super SIM card.

[0042] The above position information can be reported by the user terminal to the communication base station based on the authorized function of the user, and then provided to the service platform by the communication base station. The position information reported by the user terminal can be used to provide communication services for the user terminal through the communication base station, and the position information has real reliability, which can help to improve the service recommendation quality.

[0043] The above-mentioned super SIM (Subscriber Identity Module, SIM) card can be inserted into the user terminal through the pre-set slot of the user terminal.

[0044] The SIM card is a kind of smart card, which can be used to store user identity identification data, short message data and telephone number, etc. The super SIM card has more practical functions than the SIM card, such as realizing the card swiping function through near field communication (Near Field Communication, NFC), realizing the identity verification function through 5G electronic signature, etc. In addition, the storage space of the super SIM card is often larger than that of the SIM card, which can store more information during use.

[0045] In this application embodiment, one Super SIM card corresponds to the identity information of one communication user, and one communication user can hold one or more Super SIM cards.

[0046] S12: If the location information indicates that the user terminal is located within a preset electronic fence, then instruct the user terminal to display a Super SIM message, the Super SIM message including information on optional service items corresponding to the preset electronic fence.

[0047] In the field of mobile communication technology, electronic fences can be pre-set based on real geographical locations. By setting electronic fences, different geographical areas can be divided, thereby providing customized service functions for different locations.

[0048] In this embodiment, the preset electronic fence can be pre-set by the service provider. For example, a merchant can set up a circular electronic fence with a certain radius centered on the location of their physical store. When a user enters this area with a mobile terminal, the electronic fence can effectively identify the user terminal that can provide the service.

[0049] In practical applications, different service providers can set up different electronic fences according to their own needs. These electronic fences can also be customized based on time periods, such as setting up one type of electronic fence for weekdays and another type for weekends. By setting up electronic fences, service providers can flexibly offer a variety of services.

[0050] The service provider can set corresponding optional service items for the electronic fence. These optional service items can specifically include the services that the service provider can offer to users within the electronic fence. The information for the optional service items can include various forms of information such as text, images, and videos, which can display the available services to the user. Then, the user can select the desired service item through interactive operations according to their actual needs.

[0051] In this embodiment, if the location information indicates that the user terminal is within a preset electronic fence, the user terminal is instructed to display a Super SIM message. The Super SIM message can be based on 5G (5th Generation Mobile Communication Technology) and can also be called a Super SIM guaranteed delivery message. In this step, the service platform issues an instruction to the Super SIM card within the preset electronic fence to display the Super SIM message on the connected user terminal via the Super SIM card.

[0052] For example, a Super SIM message might include a description of a recommended optional service, offering the option to accept or decline the service. Alternatively, a Super SIM message could include multiple optional services, which users can select by checking a box. In practice, optional services can also be displayed in other ways.

[0053] S13: Based on the interactive operation received by the user terminal for the optional service item, perform identity verification through the super SIM card.

[0054] If a user interacts with any of the optional services listed in the Super SIM message, an identity verification process is performed on the Super SIM card. Specifically, the user can interact with the Super SIM application to exchange authentication messages. During this interaction, an encrypted data channel is used to transmit authentication-related messages, ensuring the security and effectiveness of the authentication process.

[0055] The interactive operation received by the user terminal can refer to an operation performed by the user, indicating that the user has selected an acceptable service item from the available service items. This step verifies the user's Super SIM card identity after the user selects a service item but before providing the corresponding service. This enables user identity authentication and facilitates the subsequent provision of customized services to fully meet the user's personalized service needs.

[0056] S14: Based on the identity verification result, instruct the user terminal to display service content matching the identity information of the Super SIM card via 5G messages, wherein the service content is the service content predicted by the pre-trained service content recommendation model based on the identity information.

[0057] The identity verification result indicates whether the Super SIM card's identity information grants the user the right to use the services selected by the user. The identity verification result may also include the necessary information to determine the matched service content.

[0058] For example, the aforementioned pre-trained service content recommendation model can specifically be a generative pre-trained foundation model (PFM) based on the Transformer framework.

[0059] Optionally, if the identity verification result indicates that the identity information is entitled to use the aforementioned services, the identity information of the Super SIM card can be input into a pre-trained service content recommendation model to obtain the service content predicted by the model.

[0060] Optionally, if the identity verification result also carries other user profile information corresponding to the identity information, then the identity information and user profile information, as well as other information with user characteristics, can be jointly input into the above-mentioned pre-trained service content recommendation model to provide the model with sufficient user characteristics, thereby obtaining service content that fits the user characteristics output by the model.

[0061] Among these methods, the pre-trained service content recommendation model predicts service content that aligns with the user's individual characteristics. For instance, in a product recommendation scenario, the pre-trained model can recommend stores and related promotional information that match the user's age group based on their identity information, thus presenting the user with attractive service content.

[0062] In this step, the user terminal is instructed to display service content matching the identity information of the Super SIM card via 5G messages. These 5G messages can be rich media messages, capable of fully displaying various forms of information such as text, images, voice, and video. 5G messages enrich the expressiveness of service content, allowing users to intuitively experience it.

[0063] In this embodiment, firstly, location information reported by a user terminal, which is pre-installed with a Super SIM card, is obtained. If the location information indicates that the user terminal is within a preset electronic fence, the user terminal is instructed to display a Super SIM message, which includes information about optional service items corresponding to the preset electronic fence. Then, based on the user terminal's received interaction with the optional service items, identity verification is performed through the Super SIM card. Finally, based on the identity verification result, the user terminal is instructed to display service content matching the identity information of the Super SIM card via 5G messages. The service content is predicted by a pre-trained service content recommendation model based on the identity information. This solution accurately provides users with information about optional service items based on the location information reported by the user terminal, and then performs reliable and authentic identity verification through the Super SIM card. Subsequently, the display via 5G messages enhances the richness of the message content, enables multi-form presentation of service content matching the identity information, and optimizes the readability of the service content. The fact that the service content is predicted by a pre-trained service content recommendation model based on the identity information improves the matching degree between the displayed service content and the identity information of the Super SIM card, optimizes the quality of the recommended service content, and enhances the attractiveness of the recommended service to users.

[0064] The following section will further illustrate this solution using an application scenario.

[0065] The service recommendation method provided in this application can be used to recommend merchant services to users, and the service platform that executes the service recommendation method may include a communication base station. Figure 1bA schematic diagram of the system architecture of a service platform for executing service recommendation methods is shown.

[0066] First, acquire the necessary data to provide a foundation for subsequent service recommendations. This data foundation can include base station data, merchant and customer behavioral preferences, merchant and customer static attributes, model-generated results from pre-trained models predicting customer behavior, and manually labeled results. Specifically, base station data can include base station geographic location data, which can be used to implement geofencing functionality. Merchant and customer static data can be obtained from a merchant and customer database. Merchant and customer behavioral preferences or other non-static attributes can be obtained from merchant and user behavioral data. For example, behavioral data can include data related to actions such as setting parameters, swiping / clicking, sending messages, ordering and paying, and providing feedback.

[0067] The service platform includes a card management platform, specifically a Super SIM card management platform. It enables identity authentication and chip-level security protection based on the Super SIM card. Through an application installed on the USIM card, it can execute commands and interact with the user terminal, thereby displaying a guaranteed message window on the user terminal and ensuring the effective delivery of Super SIM messages.

[0068] The service platform includes a 5G messaging platform, which specifically encompasses subscription account services and industry account services. For merchants, this platform allows for flexible service recommendation settings tailored to their specific needs. These settings can include selecting message templates, push strategies, and delivery times, enabling personalized configuration to meet individual merchant requirements. For users, 5G messaging provides an immersive service experience, and combined with the authentication capabilities of the Super SIM card, it enables one-click operations from browsing to payment.

[0069] The service platform includes an algorithm scheduling engine, which can achieve effective service recommendation prediction based on preset rules, recommendation algorithms, and large models based on the Transformer framework.

[0070] The service platform can predict matching service functions based on the user's terminal location and flexibly recommend various merchant services such as coupons, discount coupons, membership benefits, and limited-time discounts.

[0071] Figure 1cThe diagram illustrates a scenario application. First, merchants can set the location range of an electronic fence within the system. Once a user's terminal enters the electronic fence's range, a Super SIM message is sent to the user's terminal via a combination of the Super SIM card and 5G messaging. This allows the user to interact with the Super SIM message, achieving strong user reach and one-click security services. Then, based on the user's interaction, Super SIM card identity verification is performed. Subsequently, based on the merchant's pre-configured algorithm engine, predicted service content is pushed to the user via 5G messaging. After service recommendation, the system automatically collects merchant and customer behavior data involved in the recommendation process and distributes it to technical personnel. This information can be used for tagging, calibration, and fine-tuning, using a "machine + human" approach to optimize the system's basic model and achieve model quality optimization and updates.

[0072] For example, based on Figure 1c The scenario is illustrated. Merchants pre-configure the necessary information in the system, including the electronic fence distance, Super SIM card parameters, 5G message templates, and algorithm models. By using base station data to detect customers who have entered the fence, the mobile terminal sends a request command to the Super SIM card. The Super SIM card then displays a Super SIM guaranteed delivery message on the customer's terminal based on pre-set data and strategies. If the user interacts with the Super SIM message and accepts the request, identity verification is further performed through the Super SIM card. After successful verification, the 5G messaging application is launched, displaying recommended service content so the user can continue to enjoy one-click, no-follow-interaction services such as browsing, recommendations, and ordering. During the service recommendation process, all data is collected for modeling and analysis. Furthermore, if the user rejects the request, a follow-up message to retrieve the service method can be sent, preventing further service recommendations and avoiding impact on the customer's other communication experiences.

[0073] In the solution provided in this application embodiment, the service content displayed via 5G messages is predicted by a pre-trained service content recommendation model. This model application method can balance commercial stability and leapfrogging, and can also provide service differentiation, allowing merchants to flexibly configure it according to actual needs. Specifically, the prediction of service content based on the pre-trained model and its display via 5G messages can be implemented based on a database and service interface, possessing stability, service function differentiation, and practical application economy. Compared with preset service content recommendation methods such as keyword triggering, this solution, through the pre-trained model, can pay attention to the details of users' actual application preferences, effectively improving the matching degree between service content recommendations and users' own preferences.

[0074] Based on the solutions provided in the above embodiments, optionally, such as Figure 2As shown, before step S14 above, that is, before instructing the user terminal to display service content matching the identity information of the Super SIM card via 5G messages based on the identity verification result, the following steps are also included:

[0075] S21: Obtain service operation information of the sample user, the service operation information including the identity information of the sample user and the interaction operation information of the sample user on the optional service items.

[0076] In this step, service operation information of sample users is pre-acquired in batches. This service operation information can be obtained from a database and includes the sample users' identity information. Specifically, this identity information can be the user's own information, the user's super SIM card identity information, or other information that can be used to identify the user. The interaction operation information included in the service operation information refers to the information on the sample users' interactive operations on optional service items. This interaction operation information indicates that the sample users have selected the service item they need from at least one optional service item, that is, it indicates that the sample users have chosen to accept the service information.

[0077] S22: Construct training samples based on the service operation information and generate training labels corresponding to the training samples. The training labels represent whether the sample user of the corresponding training sample selects the target service item through interactive operation.

[0078] In this step, training samples and corresponding training labels are constructed based on the aforementioned service operation information to provide a foundation of sample data for subsequent model training. The training samples are constructed from the service operation information and possess the identity characteristics and interaction operation characteristics of the sample users. Specifically, the interaction operation characteristics can be represented as the features of the optional service items provided by the user. The training labels corresponding to the training samples indicate whether the sample user has selected a target service item. These labels indicate whether there is a match between the user's identity and the optional service item in the corresponding training sample, i.e., whether the optional service item is one that the user is willing to choose.

[0079] S23: Train a transformation neural network based on the training samples and corresponding training labels to obtain a service content recommendation model.

[0080] In this step, the Transformer-based neural network is trained based on the training samples and corresponding training labels. This enables the model to learn the correspondence between user identity features and interactive operations, that is, to learn which service items the user's identity features match. As a result, the trained model can predict the service items that the user is willing to accept based on the user's identity features and other information.

[0081] The solution provided by the embodiments of this application can train a transform neural network based on the user's identity information and interaction operation information, enabling the model to learn the matching relationship between the user's identity information and the service items the user is willing to accept. This allows the service content recommendation model to accurately predict which service the user actually needs based on the user's identity information, thereby improving the effectiveness of service recommendation.

[0082] Based on the solutions provided in the above embodiments, optionally, such as Figure 3a As shown, in step S23 above, a transformation neural network is trained based on the training samples and corresponding training labels to obtain a service content recommendation model, including:

[0083] S31: Input the service operation information into the embedding layer to obtain the service operation vector.

[0084] Figure 3b A schematic diagram of an optional model training process is shown. In this step, service operation information is used as input to the embedding layer, thereby converting the service operation information into vector form to obtain the service operation vector.

[0085] S32: Input the service operation vector into the transformation neural network, and sequentially pass it through the multi-head attention layer, the first residual normalization layer, the feedforward network layer and the second residual normalization layer of the transformation neural network. Determine the model loss through the loss function, and iteratively train the transformation neural network to obtain the service content recommendation model.

[0086] Then, the service operation vector is input into the Transformer neural network, which sequentially passes through a multi-head attention layer, a first residual normalization layer (ResNets 101 Norm, connected to the multi-head attention), a feedforward network layer, and a second residual normalization layer (ResNets 101 Norm, connected to the feedforward). The model loss is then determined by a loss function; in this example, cross-entropy loss can be used. The Transformer neural network is then iteratively trained to obtain the service content recommendation model.

[0087] The service content recommendation model provided in this application uses a training and fine-tuning method for iterative processing. Since 5G messages are rich media communication messages, the core modules in this solution all employ multimodal tasks, utilizing encoders to process the input data sequentially. Specifically, the input data can include various information such as user location, text, images, and interactive actions related to service operations.

[0088] Specifically, multimodal tasks can involve M-ICL (Multimodal In-Context Learning) and M-CoT (Multimodal Chain of Thought), which are used to ensure that when the model processes the position of words in a sequence, it can calculate weights based on different positional information, thereby effectively capturing the contextual relationships between words, sentences, and paragraphs and triggering the association mechanism of thought chains.

[0089] Optionally, for the input data, the model can generate multiple different and reasonable responses based on contextual information, thereby simulating the non-linear, multi-dimensional, and highly flexible thinking characteristics of humans when dealing with problems.

[0090] The model's understanding of generalized input content and recommendation accuracy can be fine-tuned using the M-IT (Multimodal Instruction Tuning) module. Specifically, instruction tuning can be achieved through reinforcement learning, reward / penalty evaluation, and other fine-tuning methods.

[0091] The optimization algorithm employs backpropagation to update model parameters, using the gradient changes of the cross-entropy loss function to gradually adjust the parameters. During training, monitoring can be used to control the gradient changes appropriately, preventing overfitting and avoiding gradient explosion or vanishing.

[0092] In the solution provided in this application embodiment, the model input layer can convert information into different vector block representations according to the rich media type, for example, This represents 5G message information. Feature dimensionality reduction and extraction are performed using embedding technology, followed by z-score normalization and feature concatenation, resulting in: .in, This indicates the embedding result. This represents a message vector.

[0093] Because Transformer uses global information, PE (Positioning Object) can be used to encode positions in practical applications. For example, it can be represented as:

[0094]

[0095]

[0096] Where d represents the dimension, and 2i and 2i+1 represent parity.

[0097] Similarly, the recommended service items and feedback vector representations are calculated accordingly, and ResNets101 is selected as the network for service information extraction.

[0098] The solution provided in this application can achieve information vectorization through the embedding layer, fully extract service operation information features, which is conducive to the model learning the feature association between user identity and selected service items, thereby improving the accuracy of the prediction results of the service content recommendation model.

[0099] Based on the solutions provided in the above embodiments, optionally, such as Figure 4 As shown, before step S32 above, that is, before inputting the service operation vector into the transformation neural network, the following is also included:

[0100] S41: Construct a graph neural network based on service operation information, wherein the graph neural network is used to represent the association between the identity information of sample users and the optional service items.

[0101] See Figure 3b Before inputting the vector into the Transformer for encoding, a non-Euclidean graph network (GNN) can be used. In the GNN, matching nodes are constructed based on the user's identity information and available services. The nodes store the user's identity information and the information of available services, as well as the association elements between identity nodes and service node, fully reflecting the relationship between the user's identity information and the services.

[0102] S42: The parameters of the multi-head attention layer are determined based on the graph neural network.

[0103] In this embodiment, the service operation vectors output by the aforementioned embedding layer are used to train the graph neural network, and then the parameters of Multi-HeadAttention are generated based on the training results of the graph neural network. By transforming the multi-head attention layer in the neural network, multiple independent self-attention layers are used for parallel computation, and then the results are merged to capture information from different levels.

[0104] This parameter can be represented as ,in, It is the output of a single-layer self-attention mechanism. It is a matrix concatenation transformation function, and Z is the final result.

[0105] in, The calculation formula is: ,in, For the input vector The result is obtained using matrix transformation. yes The number of columns in a vector can be used to prevent the inner product from becoming too large.

[0106] In practical applications, the parameters of the multi-head attention layer The data is fed into the residual feedforward device, where the optimized parameter quality of the GNN can accelerate network convergence. Here, Q, K, and V are obtained by linear transformation of the input vector x, which improves the model's fitting ability. Optionally, Q can refer to the information to be queried, K can refer to the queried vector, and V can refer to the retrieved value.

[0107] Based on the solutions provided in the above embodiments, optionally, such as Figure 5 As shown, after step S32 above, that is, after determining the model loss through the loss function, the following steps are also included:

[0108] S51: The transformation neural network is iteratively trained based on the classification evaluation index and the model loss to obtain a service content recommendation model, wherein the classification evaluation index is determined based on the weighted harmonic mean of accuracy and recall.

[0109] See Figure 3b In this embodiment, content comprehension can be evaluated using the F1 module. This evaluation combines human assessment dimensions with penalty scores, employing a hybrid "machine + human" approach to reward and punish the output. This reward and punishment evaluation method is highly compatible with the characteristics of the task scenario, effectively ensuring the rationality and quality of the model's output.

[0110] Based on the solutions provided in the above embodiments, optionally, such as Figure 6 As shown, before step S51 above, that is, before iteratively training the transform neural network based on the classification evaluation index and the model loss to obtain the service content recommendation model, the following steps are also included:

[0111] S61: Based on BLEU evaluation index parameters ROUGE evaluation index parameters and preset parameters Determine the classification evaluation indicators for: .

[0112] Regarding information content comprehension, the solution provided in this application embodiment can use the BLEU metric for evaluation. Considering the characteristics of message length, parameters can be set. In Furthermore, considering that this metric only focuses on accuracy and may cause missed recalls, it can be combined with ROUGE to improve recall.

[0113] In this embodiment of the application, the BLEU evaluation index is combined with the ROUGE evaluation index, see [link to relevant documentation]. Figure 3b ,when At that time , and These represent the results of BLEU and ROUGE, respectively.

[0114] In addition, the entire Transformer network uses a fully connected approach for fusion computation, which can effectively enhance attention to message content, user profiles, and service category characteristics, thereby improving the rationality of the matching degree.

[0115] Based on the solutions provided in the above embodiments, optionally, such as Figure 7 As shown, in step S23 above, a transformation neural network is trained based on the training samples and corresponding training labels to obtain a service content recommendation model, including:

[0116] S71: Based on the training samples and corresponding training labels, a transform neural network is trained using the cross-entropy loss function to obtain a service content recommendation model, wherein the cross-entropy loss function is:

[0117]

[0118] Where N represents the total number of training samples, i represents the identifier of the training sample, j represents the classification index of the data category in the training sample, M represents the total number of data categories in the training sample, p represents the output predicted value, and y represents the training label. The training label is, for example, 0 or 1, where 1 indicates relevance, i.e., the user selected the corresponding service item, and 0 indicates no relevance, i.e., the user rejected the service item.

[0119] In the solution provided in this application embodiment, since the data categories involved in the model are diverse, the cross-entropy function is used as the evaluation function, which can effectively improve the training effectiveness for multi-category information.

[0120] Based on the solutions provided in the above embodiments, optionally, such as Figure 8a As shown, before step S11 above, that is, before obtaining the location information reported by the user terminal, the process also includes:

[0121] S81: Receive configuration information corresponding to optional service items, wherein the configuration information includes at least one of the following: preset area, super SIM card parameters, and identifier of service content recommendation model.

[0122] See Figure 8b In practical applications, merchants providing services can pre-configure the relevant information for optional service items. Merchants can first register and log in to the service platform. During registration, they can choose to activate either a card management platform account or a subscription account, depending on their needs. Registered merchants can then log in using their platform accounts.

[0123] After successfully logging in, configure the card management platform parameters or subscription account platform parameters through the service platform. Configurable parameters include, for example, the user's electronic fence range, the frequency of Super SIM messages, 5G message templates, bottom menu bar visualization settings, ChatBot interaction settings, and algorithm scheduling strategies.

[0124] The configuration results can be displayed on a visual page so that merchants can intuitively see the configuration effect. If they are not satisfied with the configuration effect, they can modify the parameters of the card management platform or the subscription account platform until they are satisfied with the configuration effect. Then they can save the configuration and log out of the service platform account.

[0125] Based on the solutions provided in the above embodiments, optionally, such as Figure 9a As shown, in step S13 above, based on the interactive operation received by the user terminal for the optional service item, identity verification is performed through the Super SIM card, including:

[0126] S91: If the interaction operation indicates that the target user selects a target service item from the Super SIM message, then the target user's identity is verified through the Super SIM card.

[0127] In this step, the interaction of the target user is parsed. If the target user selects a target service item in the operation performed on the Super SIM message, the target user's identity is further verified through the Super SIM card in order to provide the target user with relevant service recommendations for the target service item in subsequent steps.

[0128] The following is combined with Figure 9b The flowchart shown further illustrates this solution.

[0129] Merchants providing services pre-configure parameters on the card management platform and the 5G message subscription platform. The service platform provides the execution basis for subsequent steps through data collection and algorithm engine.

[0130] The user terminal reports its location information based on its own location. Once the user terminal enters the merchant's electronic fence range, the service platform can activate the card platform's STK forced pop-up capability and send a Super SIM guaranteed delivery message to the user terminal through the SIM card channel.

[0131] After receiving and displaying the Super SIM Guaranteed Delivery message on their user terminal, users can choose to accept or reject service items based on their actual needs. If the user chooses to accept the target service item, the terminal's 5G messaging application is invoked based on SIM authentication and identity verification. The recommended service content is then displayed via 5G messaging, which is predicted by an algorithm scheduling engine using a pre-trained service content recommendation model. During the service recommendation process, users can enjoy services within 5G messaging without registering or following the service provider, allowing them to immerse themselves in the service experience or browsing.

[0132] If a user chooses to reject the service, they will be unable to access subsequent services, and a reminder to re-acquire the service will be sent to the user's terminal. In practical applications, if a user accidentally triggers the rejection, they can re-acquire the service through this reminder.

[0133] In practical applications, user interactions can indicate whether or not the user has opted into the local life information push service. If the user clicks on the Super SIM guaranteed message to indicate their willingness to subscribe, a confirmation message can be sent to the user's terminal confirming the successful subscription. Upon successful subscription, subscription, configuration, account, and other related information can be synchronized to both the Super SIM platform and the 5G message subscription platform to ensure consistency of service-related information.

[0134] Furthermore, users can perform actions on subscribed local lifestyle information to gain a deeper experience of local information services. For example, by opening, browsing, and selecting pages related to local lifestyle information, users can be directed to the 5G Messaging Subscription Account. Within the 5G Messaging Subscription Account, users can experience the service content in depth by clicking the bottom menu bar, clicking the floating menu, and browsing rich media messages.

[0135] Based on the solutions provided in the above embodiments, optionally, such as Figure 10 As shown, in step S14 above, based on the identity verification result, the user terminal is instructed to display service content matching the identity information of the Super SIM card via 5G messages, including:

[0136] S101: If the identity verification result indicates that the target user's identity verification is successful, and the service content matched by the target user includes transaction service content, then instruct the Super SIM card to provide identity authentication function to the user terminal through a digital certificate.

[0137] See Figure 9bThe process shown illustrates that if the target user's matched service content includes transaction services, such as sending red envelopes or transferring funds, then authentication and payment functions are invoked based on the Super SIM card, utilizing the Super SIM card's secure transaction capabilities through a digital certificate. Specifically, the SIM Shield application can provide effective identity security for mobile payment scenarios.

[0138] Furthermore, in the process provided in this application embodiment, if an execution step fails due to an error or other reason, it can be retried a preset number of times, such as three times. If the problem persists after the preset number of retries, the customer service center can be notified, and the customer can be manually assisted in troubleshooting the cause of the operation failure, such as the reason for subscription failure or SIM capability call failure.

[0139] In addition, during the steps that require users to wait, such as authentication, function calls, and data retrieval, the service platform can display the current progress to users and guide them to wait or perform the necessary authentication operations through pop-up prompts, thereby optimizing the user's visual experience.

[0140] The solution provided in this application combines the advantages of multiple technologies such as Super SIM card, 5G messaging, and large model. It achieves strong user reach and service interaction through Super SIM card and 5G messaging, and continuously evolves service recommendation capabilities through large model iteration.

[0141] Based on the Super SIM card's guaranteed messaging function, it can effectively receive user feedback receipts for Super SIM messages, instantly knowing whether the user has accepted or rejected services. For the user-selected target service, the Super SIM card interoperates with the 5G messaging gateway, combining large-scale network static data to achieve a unique mapping between terminal information, mobile phone number, service system, and identifier, effectively verifying the user's login identity. In transaction scenarios involving sensitive user identity information, the Super SIM card's SIM Shield service uses digital certificates to reliably verify user identity, combining multiple information verifications such as the operator's three key elements, transaction party interface, and message service identifier to ensure payment security.

[0142] Based on the capabilities of large-scale models, a service content recommendation model is trained using the Transformer framework. This is extended with multimodal modules such as M-ICL, M-CoT, and M-IT, and the model parameters are fine-tuned by mimicking human reinforcement learning and reward / punishment strategies. The output content can be validated manually or using professional technical data, enabling the application of large-scale industry models that integrate machine and human expertise.

[0143] For merchants providing services, the solution provided in this application uses Super SIM messages and 5G messages to achieve service recommendation, ensuring message delivery and avoiding being overwhelmed by a large amount of invalid message content. Furthermore, Super SIM messages can effectively obtain user feedback, allowing merchants to adjust their service delivery strategies based on actual user actions. In addition, the pre-trained service content recommendation model can effectively predict the service content needed by users, reducing the cost of service recommendation for merchants and improving service conversion rates.

[0144] This solution allows users to easily and effectively view service content through Super SIM messages and 5G messages, without requiring users to follow official accounts or log in with their identity information. This enables users to conveniently and quickly experience services and reduces the complexity of service experience operations.

[0145] In addition, this solution utilizes location information reported by the user terminal that the user has authorized to use, and can perform service recommendations based on the actual valid location information of the user terminal.

[0146] To address the problems existing in related technologies, this application also provides a service recommendation device 110, such as... Figure 11 As shown, it includes:

[0147] The acquisition module 111 acquires the location information reported by the user terminal, wherein the user terminal is pre-installed with a super SIM card;

[0148] The instruction module 112, if the location information indicates that the user terminal is located within a preset electronic fence, instructs the user terminal to display a Super SIM message, the Super SIM message including information on optional service items corresponding to the preset electronic fence;

[0149] The verification module 113 performs identity verification through the super SIM card based on the interactive operation received by the user terminal for the optional service item.

[0150] The display module 114 instructs the user terminal to display service content matching the identity information of the super SIM card via 5G messages based on the identity verification result. The service content is the service content predicted by the pre-trained service content recommendation model based on the identity information.

[0151] The apparatus provided in this application first obtains location information reported by a user terminal, which has a pre-installed Super SIM card. If the location information indicates that the user terminal is within a preset electronic fence, the user terminal is instructed to display a Super SIM message, which includes information about optional service items corresponding to the preset electronic fence. Then, based on the user terminal's received interaction with the optional service items, identity verification is performed through the Super SIM card. Finally, based on the identity verification result, the user terminal is instructed to display service content matching the identity information of the Super SIM card via 5G messages. The service content is predicted by a pre-trained service content recommendation model based on the identity information. This solution accurately provides users with information about optional service items based on the location information reported by the user terminal, and then performs reliable and authentic identity verification through the Super SIM card. Subsequently, the display via 5G messages enhances the richness of the message content, enables multi-form presentation of service content matching the identity information, and optimizes the readability of the service content. The service content, predicted by a pre-trained service content recommendation model based on the identity information, improves the matching degree between the displayed service content and the identity information of the Super SIM card, thus increasing the attractiveness of the recommended services to users.

[0152] In this application, the modules in the apparatus provided can also implement the method steps provided in the method embodiments. Alternatively, the apparatus provided in this application may include other modules besides those described above to implement the method steps provided in the method embodiments. Furthermore, the apparatus provided in this application can achieve the technical effects achievable by the method embodiments.

[0153] Preferably, this application embodiment also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described service recommendation method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0154] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the above-described service recommendation method embodiment and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0155] This application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program. The computer program is operable to cause a computer to perform some or all of the steps of the above-described service recommendation method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0156] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0157] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0158] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0159] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0160] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0161] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0162] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0163] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0164] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0165] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A service recommendation method, characterized in that, include: Obtain location information reported by a user terminal, wherein the user terminal is pre-installed with a super SIM card; If the location information indicates that the user terminal is located within a preset electronic fence, then the user terminal is instructed to display a Super SIM message, which includes information about optional service items corresponding to the preset electronic fence. Based on the interactive operations received by the user terminal for the optional service items, identity verification is performed through the super SIM card; Based on the identity verification result, the user terminal is instructed to display service content matching the identity information of the Super SIM card via 5G messages. The service content is the service content predicted by a pre-trained service content recommendation model based on the identity information.

2. The method as described in claim 1, characterized in that, Before instructing the user terminal to display service content matching the identity information of the Super SIM card via 5G messages based on the identity verification result, the process also includes: Obtain service operation information of sample users, including the identity information of the sample users and the interaction operation information of the sample users on optional service items; Training samples are constructed based on the service operation information, and training labels corresponding to the training samples are generated. The training labels represent whether the sample user of the corresponding training sample selects the target service item through interactive operation. Based on the training samples and corresponding training labels, a transformation neural network is trained to obtain a service content recommendation model.

3. The method as described in claim 2, characterized in that, Based on the training samples and corresponding training labels, a transformation neural network is trained to obtain a service content recommendation model, including: The service operation information is input into the embedding layer to obtain the service operation vector; The service operation vector is input into the transformation neural network and sequentially passed through the multi-head attention layer, the first residual normalization layer, the feedforward network layer, and the second residual normalization layer of the transformation neural network. The model loss is determined by the loss function, and the transformation neural network is iteratively trained to obtain the service content recommendation model.

4. The method as described in claim 3, characterized in that, Before inputting the service operation vector into the transform neural network, the process further includes: A graph neural network is constructed based on service operation information, wherein the graph neural network is used to represent the association between the identity information of sample users and the available service items; The graph neural network is used to determine the parameters of the multi-head attention layer.

5. The method as described in claim 3, characterized in that, After determining the model loss using the loss function, the following steps are also included: The transformation neural network is iteratively trained based on the classification evaluation index and the model loss to obtain a service content recommendation model, wherein the classification evaluation index is determined based on the weighted harmonic mean of precision and recall.

6. The method as described in claim 5, characterized in that, Before iteratively training the transform neural network based on the classification evaluation index and the model loss to obtain the service content recommendation model, the process further includes: Based on BLEU evaluation index parameters ROUGE evaluation index parameters and preset parameters Determine the classification evaluation indicators for: .

7. The method as described in claim 5, characterized in that, Based on the training samples and corresponding training labels, a transformation neural network is trained to obtain a service content recommendation model, including: Based on the training samples and corresponding training labels, a transform neural network is trained using the cross-entropy loss function to obtain a service content recommendation model, wherein the cross-entropy loss function is: Where N represents the total number of training samples, i represents the identifier of the training sample, j represents the classification index of the data category in the training sample, M represents the total number of data categories in the training sample, p represents the output predicted value, and y represents the training label.

8. The method according to any one of claims 1 to 7, characterized in that, Before obtaining the location information reported by the user terminal, the process also includes: Receive configuration information corresponding to optional service items, wherein the configuration information includes at least one of the following: preset region, super SIM card parameters, and identifier of service content recommendation model.

9. The method according to any one of claims 1 to 7, characterized in that, Based on the interactive operations received by the user terminal for the optional service items, identity verification is performed through the Super SIM card, including: If the interactive operation indicates that the target user has selected a target service item from the Super SIM message, then the target user's identity is verified through the Super SIM card.

10. The method according to any one of claims 1 to 7, characterized in that, Based on the identity verification result, the user terminal is instructed to display service content matching the identity information of the Super SIM card via 5G messages, including: If the identity verification result indicates that the target user's identity verification is successful, and the service content matched by the target user includes transaction service content, then the Super SIM card is instructed to provide identity authentication function to the user terminal through a digital certificate.

11. A service recommendation device, characterized in that, include: The acquisition module acquires location information reported by the user terminal, which is pre-installed with a super SIM card; The instruction module, if the location information indicates that the user terminal is located within a preset electronic fence, instructs the user terminal to display a Super SIM message, the Super SIM message including information on optional service items corresponding to the preset electronic fence; The verification module performs identity verification through the Super SIM card based on the interactive operations received by the user terminal for the optional service items. The display module instructs the user terminal, based on the identity verification result, to display service content matching the identity information of the Super SIM card via 5G messages. The service content is the service content predicted by a pre-trained service content recommendation model based on the identity information.

12. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 10.

Images