Service push method and device, storage medium and electronic device

Abstract

The present application discloses a service push method and device, a storage medium and an electronic device, and relates to the field of artificial intelligence. The method includes: obtaining a target user portrait, wherein the target user portrait is used to characterize the characteristic information of the target user; constructing a target AI virtual robot based on the target user portrait, and placing the target AI virtual robot into a virtual scene; predicting the needs of the target user in the real scene based on the needs of the target AI virtual robot in the virtual scene; determining a first service and a target service associated with the first service based on the needs of the target user in the real scene, wherein the first service is a service predicted to be required by the target user; and pushing the target service to the target user according to a preset push method. Through this application, the problem of low accuracy of service push in related technologies, which affects user experience, is solved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and more specifically, to a service push method and device, a storage medium, and an electronic device. Background Art

[0002] Business recommendations are a crucial component of financial institutions' operations. A well-executed recommendation system not only enhances customer affinity for the institution but also ensures a positive customer experience. Furthermore, the number and quality of customers are crucial to the success of a financial institution's operations. Therefore, providing excellent service and ensuring customer satisfaction and enjoyment is paramount to the sustainable development of financial institutions.

[0003] In current related technologies, major financial institutions typically use the following methods to recommend services to users: manual lobby push notifications, door-to-door push notifications by sales representatives, and phone service push notifications; traditional direct sales models, which blindly recommend services approved by product managers online; popular services identified through big data analysis, which are pushed to users via apps, WeChat official accounts, websites, and other methods; and custom personalized push notification algorithms, which push services tailored to specific dates, such as holidays, birthdays, and anniversaries. However, these push notification methods are typically based on the experience and market analysis of service managers, and are based on methods that the service managers are personally familiar with. Therefore, these current push notification methods largely fail to fully consider the user's personal experience, resulting in service push notifications being a burden and harassment to users, and also causing a waste of resources and manpower. Furthermore, under the bombardment of these push notification methods, more and more users are actively blocking services from financial institutions, which inevitably leads to user annoyance, user churn, and resource waste.

[0004] Currently, no effective solution has been proposed to address the problem of low accuracy of service push in related technologies, which affects user experience. Summary of the Invention

[0005] The main purpose of this application is to provide a service push method and device, a storage medium and an electronic device to solve the problem in related technologies that the accuracy of service push is low and affects the user experience.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a service push method is provided. The method includes: obtaining a target user portrait, wherein the target user portrait is used to characterize the characteristic information of the target user; constructing a target AI virtual robot based on the target user portrait, and placing the target AI virtual robot into a virtual scene; predicting the needs of the target user in the real scene based on the needs of the target AI virtual robot in the virtual scene; determining a first service and a target service associated with the first service based on the needs of the target user in the real scene, wherein the first service is the service predicted to be required by the target user; and pushing the target service to the target user according to a preset push method.

[0007] Furthermore, the method for acquiring the target business includes one of the following: determining whether the first business is a business in a business library, wherein the business library is a collection of multiple businesses; if the first business is a business in the business library, extracting businesses related to the first business from the business library, and obtaining the target business from the businesses related to the first business; if the first business is not a business in the business library, generating a second business and making a decision on the second business, wherein the second business is a business related to the first business; and determining whether to use the second business as the target business based on the decision made on the second business.

[0008] Furthermore, obtaining a target user portrait includes: obtaining target information of the target user, wherein the target information is used to represent the attributes of the target user; analyzing the target information based on a deep learning network structure to obtain an analysis result; labeling the characteristic information of the target user based on the analysis result to obtain first target information, wherein the first target information is the information after labeling the characteristic information of the target user; constructing a user model of the target user based on the first target information; and obtaining the target user portrait based on the user model of the target user and three-dimensional modeling technology.

[0009] Furthermore, after determining the first business and the target business associated with the first business, the method also includes: using a decision tree algorithm through multiple AI virtual robots to vote on the target business to obtain voting results; and determining the degree of demand of the target user for the target business based on the voting results.

[0010] Furthermore, after determining the degree of demand of the target user for the target business based on the voting results, the method also includes: determining the cost data corresponding to the target business; calculating a first value based on the cost data corresponding to the target business and the benefits brought by the target AI virtual robot selecting the target business, wherein the first value is used to represent the estimated benefits brought by the target business.

[0011] Furthermore, after pushing the target service to the target user according to a preset push method, the method also includes: obtaining the target user's usage of the target service; if the target user does not use the target service, recalculating the first value until the target user uses the target service; and determining the push effect of the target service based on the recalculated value of the first value.

[0012] Furthermore, after determining the push effect of the target service, the method also includes: analyzing the target service based on the push effect of the target service; if it is analyzed that the first user has a demand for the target service and the revenue of the target service meets the preset requirements, pushing the target service to the first user in a target push manner, wherein the first user is a user other than the target user; if it is detected that the target user fails to use the target service, determining the reason for the target user's failure to use the target service, and optimizing the target service based on the reason.

[0013] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a service push device is provided. The device includes: a first acquisition unit, for acquiring a target user portrait, wherein the target user portrait is used to characterize the characteristic information of the target user; a first construction unit, for constructing a target AI virtual robot based on the target user portrait, and placing the target AI virtual robot into a virtual scene; a first prediction unit, for predicting the needs of the target user in a real scene based on the needs of the target AI virtual robot in the virtual scene; a first determination unit, for determining a first service and a target service associated with the first service based on the needs of the target user in the real scene, wherein the first service is the service predicted to be required by the target user; a first push unit, for pushing the target service to the target user in a preset push method.

[0014] Furthermore, the method for acquiring the target business includes one of the following: a first judgment unit, used to judge whether the first business is a business in the business library, wherein the business library is a collection of multiple businesses; a first processing unit, used to extract businesses related to the first business from the business library if the first business is a business in the business library, and obtain the target business from the businesses related to the first business; a second processing unit, used to generate a second business if the first business is not a business in the business library, and make a decision on the second business, wherein the second business is a business related to the first business; a second determination unit, used to determine whether to use the second business as the target business based on the decision made on the second business.

[0015] Furthermore, the first acquisition unit includes: a first acquisition module, used to acquire target information of the target user, wherein the target information is used to represent the attributes of the target user; a first analysis module, used to analyze the target information based on a deep learning network structure to obtain an analysis result; a first processing module, used to label the characteristic information of the target user based on the analysis result to obtain first target information, wherein the first target information is the information after labeling the characteristic information of the target user; a first construction module, used to construct a user model of the target user based on the first target information; a first acquisition module, used to obtain a portrait of the target user based on the user model of the target user and three-dimensional modeling technology.

[0016] Furthermore, the device also includes: a first voting unit, which is used to, after determining the first business and the target business associated with the first business, vote on the target business using a decision tree algorithm through multiple AI virtual robots to obtain a voting result; and a third determination unit, which is used to determine the target user's demand degree for the target business based on the voting result.

[0017] Furthermore, the device also includes: a fourth determination unit, used to determine the cost data corresponding to the target business after determining the degree of demand of the target user for the target business based on the voting results; a first calculation unit, used to calculate a first value based on the cost data corresponding to the target business and the benefits brought by the target AI virtual robot selecting the target business, wherein the first value is used to represent the estimated benefits brought by the target business.

[0018] Furthermore, the device also includes: a second acquisition unit, used to obtain the target user's usage of the target service after pushing the target service to the target user in a preset push method; a second calculation unit, used to recalculate the first value if the target user does not use the target service until the target user uses the target service; and a fifth determination unit, used to determine the push effect of the target service based on the recalculated value of the first value.

[0019] Furthermore, the device also includes: a first analysis unit, for analyzing the target business according to the push effect of the target business after determining the push effect of the target business; a third processing unit, for pushing the target business to the first user in a target push manner if it is analyzed that the first user has a demand for the target business and the revenue of the target business meets the preset requirements, wherein the first user is a user other than the target user; a fourth processing unit, for determining the reason for the target user's failure to use the target business if it is detected that the target user fails to use the target business, and optimizing the target business based on the reason.

[0020] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a computer-readable storage medium is provided, wherein the storage medium includes a stored program, wherein the program executes any one of the above-mentioned service push methods.

[0021] In order to achieve the above-mentioned purpose, according to another aspect of the present application, an electronic device is provided, which includes one or more processors and a memory, and the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any one of the above-mentioned service push methods.

[0022] Through this application, the following steps are adopted: obtaining a target user portrait, wherein the target user portrait is used to characterize the characteristic information of the target user; constructing a target AI virtual robot based on the target user portrait, and placing the target AI virtual robot into a virtual scene; predicting the target user's needs in the real scene based on the needs of the target AI virtual robot in the virtual scene; determining a first business and a target business associated with the first business based on the needs of the target user in the real scene, wherein the first business is the business predicted to be required by the target user; pushing the target business to the target user according to a preset push method solves the problem of low accuracy of business push in related technologies, which affects the user experience. By predicting the target user's needs in the real scene based on the needs of the constructed target AI virtual robot in the virtual scene, and pushing the business to the target user according to a preset push method based on the needs of the target user, the accuracy of the business push is improved, thereby improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0024] Figure 1 This is a flowchart of a service push method provided according to an embodiment of the present application;

[0025] Figure 2 is a personalized portrait of user A in the embodiment of the present application;

[0026] Figure 3 is a personalized portrait of user B in the embodiment of the present application;

[0027] Figure 4 This is a schematic diagram of placing the AI ​​virtual robot corresponding to user A into a three-dimensional virtual scene in an embodiment of the present application;

[0028] Figure 5 This is a schematic diagram of placing the AI ​​virtual robot corresponding to user B into a three-dimensional virtual scene in an embodiment of the present application;

[0029] Figure 6 Schematic diagram of the logical thinking model of the AI ​​virtual robot in the embodiment of the present application;

[0030] Figure 7 is a schematic diagram of a deep learning network model in an embodiment of the present application;

[0031] Figure 8 is a schematic diagram of cost estimation in an embodiment of the present application;

[0032] Figure 9This is a schematic diagram of updating a service in an embodiment of the present application;

[0033] Figure 10 is a schematic diagram of a service push device provided according to an embodiment of the present application;

[0034] Figure 11 is a schematic diagram of a service push system provided according to an embodiment of the present application;

[0035] Figure 12 is a schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0036] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0037] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0038] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0039] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties.

[0040] The present invention will be described below in conjunction with preferred implementation steps. Figure 1 This is a flow chart of a service push method according to an embodiment of the present application. Figure 1 As shown, the method includes the following steps:

[0041] Step S101: Obtain a target user portrait, wherein the target user portrait is used to represent characteristic information of the target user.

[0042] For example, based on the user model of the user, using 3D modeling technology, a user portrait that meets the user's characteristics is constructed in the 3D virtual space, and the user portrait is obtained. For example, the personalized portrait of user A is as follows: Figure 2 As shown in Figure 2, the personalized portrait of user B is as follows: Figure 3 shown.

[0043] Step S102: construct a target AI virtual robot based on the target user portrait, and place the target AI virtual robot into a virtual scene.

[0044] For example, based on the above-mentioned user portrait or user model, by accessing the UNIT (Understanding and Interaction Technology) technology, the virtualized intelligent AI - Alpha person, and placed in a virtualized scene built based on the real scene for cultivation and training, this step is the process of constructing intelligent AI. In addition, the scene is a simplified 3D model simulation scene created using Unity3D. In addition to the necessary basic facilities such as housing, cars, shopping malls, schools and other buildings, the scene model does not have rich, vivid and beautiful scenes. This scene mainly provides a platform for growth for many Alpha people. At the same time, the personality and self-awareness of Alpha can be continuously improved through incremental learning. For example, the schematic diagram of placing the virtual robot corresponding to user A in a three-dimensional virtual scene is as follows Figure 4 As shown, the schematic diagram of placing the virtual robot corresponding to user B into the three-dimensional virtual scene is as follows Figure 5 shown.

[0045] Step S103: predict the needs of the target user in the real scene based on the needs of the target AI virtual robot in the virtual scene.

[0046] For example, based on the actual needs of the above-mentioned Alpha people in the virtual scene, the real needs of users in the real world can be inferred. However, intelligent AI can simulate some attributes of real users to a certain extent, but it cannot predict various complex situations in real life. In order to make up for this deficiency, this application introduces a network-oriented database Neo4j, which can effectively store the interpersonal relationship map and transaction map of Alpha people. And by changing the connection relationship and weight in the map, the user's current state of focus and emphasis can be described. In addition, its model diagram is as follows Figure 6As shown in the figure, the real-time marking lines A and B are used to ensure the real-time interaction and simulate the sudden inspiration of real people. Multi-stream parallel scanning technology is used. The scanning speed and decision priority of multiple parallel scanning lines are obtained by real-time incremental learning and are uncertain. This allows for a more realistic description of the unexpected situations that users may encounter in real scenes. Therefore, based on Figure 4 It can be inferred that user A wants to buy a pair of high-end princess crystal shoes. Figure 5 It can be inferred that user B wants to buy a new house, so based on the needs of user A and user B respectively, the services required by user A and user B can be inferred.

[0047] In addition, the user's logical thinking is a key factor in determining the degree of user preference. In previous business push methods, service businesses are usually formulated based on the unilateral needs of financial institutions, and the description method in the above background is used to push the business to users. In this model, financial institutions are the providers of services and are the active party; users are the recipients of services and are the passive party. This will lead to a common phenomenon that the services provided by financial institutions are not what some users want, and financial institutions do not push the services that some users want. This application changes the thinking, regards users as the active party, establishes the user's logical thinking through user portraits, and transforms user portraits from user feature descriptions into intelligent AI that can think actively, and can request the business services it needs from the business push service system.

[0048] Step S104, determining a first service and a target service associated with the first service based on the needs of the target user in a real scenario, wherein the first service is the service predicted to be needed by the target user;

[0049] For example, after calculating the business services required by the user, the business related to the business required by the user will be further determined, and the determined business related to the business required by the user will be submitted to the business matching decision system for voting and decision-making, and finally a business that best meets the user's needs will be selected.

[0050] Step S105: Push the target service to the target user according to the preset push method.

[0051] For example, after identifying the service that best meets a user's needs, the service's push progress is submitted to a business manager who is familiar with the user for review and selection of an appropriate push method as a recommendation. Based on their experience, the business manager selects a push method that best suits the current situation and delivers it to the user with precision. This application integrates an intelligent push system to deliver targeted, quantitative push notifications of services that users urgently need. This approach, using the user's preferred push method, provides users with targeted services. This approach not only meets their actual needs but also reduces resource waste and eases the burden of the service they choose, allowing users to experience personalized services, thereby retaining more customers and generating greater profits for the financial institution. For example, after receiving the task, the business manager analyzes that user A prefers to receive service recommendations over the phone and assigns the pleasant and sweet-sounding customer service representative C to push the service to user A. Furthermore, after analyzing that user B prefers to receive service recommendations in person, the business manager assigns the pleasant and sweet-sounding customer service representative D to push the service to user B in person.

[0052] Through the above steps S101 to S105, the needs of the target user in the real scene are predicted based on the needs of the constructed target AI virtual robot in the virtual scene, and the service is pushed to the target user according to the preset push method based on the needs of the target user, thereby improving the accuracy of the service push and thus improving the user experience.

[0053] Optionally, in the service push method provided in the embodiment of the present application, the method for obtaining the target service includes one of the following: determining whether the first service is a service in the service library, wherein the service library is a collection of multiple services; if the first service is a service in the service library, extracting services related to the first service from the service library, and obtaining the target service from services related to the first service; if the first service is not a service in the service library, generating a second service and making a decision on the second service, wherein the second service is a service related to the first service; and determining whether to use the second service as the target service based on the decision made on the second service.

[0054] For example, after figuring out the business services required by the user, further searches and generation of businesses will be conducted. Specifically, first, the corresponding service terms are searched in the existing business service library, and assisted in the revision. If the business service is not available in the local business library, the corresponding business needs to be drafted, submitted upward, and wait for the decision of the manager and other relevant personnel. That is, if the business does not exist in the business system, a number of AI voting methods will be used to decide whether the business should be provided. If it needs to be provided, it will be decided by the product manager; if it does not need to be provided, it will be stored in the database and wait for relevant personnel to browse and query. If there is already a corresponding business in the local business service library, the corresponding business will be extracted and handed over to the business matching decision system for voting and decision-making. In addition, the business search technology used in this application is the distributed search method Elasticsearch, which is a distributed search engine based on Lucene. Moreover, this method is different from the usual serial search method. It is a parallel search method based on multiple processors, which can greatly speed up the data retrieval speed. In the business generation module, open-source Python crawler technology is used to first search the internet for relevant business services. These services are then analyzed for phrases and sentences, and a relevance questionnaire is created. The business is then generated based on the table content. For example, a database search revealed that no existing services related to "crystal slippers" exist. Therefore, by searching the internet for relevant materials, a plan suitable for the user is assembled, providing a cash reserve worth 20,000 yuan, which is directly charged to the user's credit card. The plan is then submitted to the business manager for evaluation. A database search revealed that existing services include home purchase loans in the city where User B resides. The push system selects the three best loan plans and sends them to the business manager, recommending a push method suitable for the user for selection.

[0055] Through the above solution, we can not only save human resources, but also quickly learn from others' strengths and improve work efficiency.

[0056] Optionally, in the service push method provided in the embodiment of the present application, obtaining the target user portrait includes: obtaining target information of the target user, wherein the target information is used to represent the attributes of the target user; analyzing the target information based on the deep learning network structure to obtain an analysis result; labeling the characteristic information of the target user based on the analysis result to obtain first target information, wherein the first target information is the information after the characteristic information of the target user is labeled; constructing a user model of the target user based on the first target information; and obtaining a target user portrait based on the user model of the target user and three-dimensional modeling technology.

[0057] For example, first, user information needs to be collected. Based on the user's history of bank-based counter and online transactions, basic personal information and account status can be extracted. This basic data is then stored in a relational model database, serving as the basis for describing the user. With the user's permission, OpenCV graphics and image processing technology is then used to identify and classify user images and photos. Basic user attribute information is constructed based on similar facial features and personality traits. Web crawler technology is then used to search the internet for publicly available content and descriptions related to the user, extracting key data and constructing a relationship and personality profile. RSA encryption is employed throughout the process to protect the user's basic information from being leaked. For example, a portion of the user information obtained is shown in Table 1.

[0058] Table 1. Feature information of similar users

[0059] Feature 1 Feature 2 … Feature 20 Label User 1 x1 y1 z1 0 User 2 x2 y2 z2 1 … … … … … User 100 x100 y100 … z100 0

[0060] Using the user data collected in the table above, the data is packaged and divided into three groups, one of which is used for training the model, one is used for verifying the model, and one is used for testing the model. The three groups of data A, B, and C are randomly selected according to the distribution method of 7:2:1, and multiple random adjustments are used to detect whether the deep learning network model is close to the optimal solution. In this training, the following Figure 7 The deep learning network structure shown in the figure is CIOU Loss function:

[0061]

[0062] Among them, IOU represents the logarithmic ratio of user attribute relevance and attribute union. The larger the ratio, the more similar the attributes between users are, and the smaller the ratio, the opposite is true. 2 represents the Euclidean distance between sample data, c represents the difference between the predicted model and the hypothesized model, b, b gt This represents the central descriptive point of the prediction model. For more efficient data processing, data is combined into batches for training. This method labels the user's preferences, habits, personality traits, and beliefs, and stores them in the database to generate a user model. Based on this user model, 3D modeling technology is used to construct a user profile in 3D virtual space that matches the user's characteristics. By establishing this user model, a one-to-one mapping between reality and virtual reality is achieved, effectively characterizing the user profile.

[0063] In addition, user portraits are databases that describe the basic characteristics of users. In the traditional model, the description of users basically stays on the user's identity information and the impression of the lobby manager who receives customers. Although these basic information can describe user characteristics to a certain extent, they can also serve as the fundamental basis for business recommendations. However, the success or failure of business recommendations is completely unpredictable. This application uses a deep learning algorithm to use every move of the user as a data source to characterize user characteristics. For example, the user's consumption habits, financial management habits, transfer habits, travel habits, dressing habits, weather preferences, eating habits, fitness and beauty, and the user's preference for business push, etc., in order to establish a complete user portrait.

[0064] The above solution effectively avoids storing user data from a single perspective, thereby achieving a multi-dimensional understanding of users and inferring their needs. It also achieves the effect of understanding users in real life and analyzing them in virtual scenarios. Furthermore, the constructed intelligent AI robot's consciousness can lead the characters in the scene to engage in autonomous communication and simulate interactions, thereby emulating real-world user needs for services. Comprehensive user profiling can also be used to infer user personality traits and preferences, providing the most robust and comprehensive data foundation for service recommendation systems.

[0065] Optionally, in the service push method provided in the embodiment of the present application, after determining the first service and the target service associated with the first service, the method further includes: using a decision tree algorithm through multiple AI virtual robots to vote on the target service to obtain voting results; and determining the degree of demand of the target user for the target service based on the voting results.

[0066] For example, when determining the user's demand for a service, the Alpha individuals generated in the above steps are used to test the service that best meets the user's needs, ultimately selecting the service that will be pushed to the user. Multiple batches of random selection tests are then conducted, and ultimately multiple Alpha individuals are asked to vote using a decision tree algorithm to select the maximum value from all optimal values. The algorithm calculation formula is as follows:

[0067]

[0068] In addition, in order to avoid over-voting and over-abandoning during the voting process. (The so-called over-voting means that during the voting process, due to some factors, all Alpha people vote for yes or no at the same time; over-abandoning means that under the current conditions, they are unable to vote, creating the illusion that all votes have been abstained.) Pre-pruning algorithms and post-pruning methods will be used in the voting process to ensure the validity of the vote. For example, when the above-mentioned solution corresponding to user A was put into the virtual scene, it was found that the virtual robot corresponding to user A immediately adopted the solution and bought a pair of crystal shoes worth 19,888 yuan; when the above-mentioned solution corresponding to user B was put into the virtual scene, it was found that the virtual robot corresponding to user B hesitated for a long time before adopting the solution, bought a new house with a loan, and lived a life of eating dirt.

[0069] The above solution not only effectively avoids the subjectivity of manual judgment, but also effectively meets user needs.

[0070] Optionally, in the service push method provided in the embodiment of the present application, after determining the degree of demand of the target user for the target service based on the voting results, the method also includes: determining the cost data corresponding to the target service; calculating a first value based on the cost data corresponding to the target service and the benefits brought by the target AI virtual robot selecting the target service, wherein the first value is used to represent the estimated benefits brought by the target service.

[0071] For example, after determining the user's demand for the service, it is necessary to estimate the cost and benefit. That is, it is necessary to calculate the corresponding cost P. The cost estimation flow chart is as follows: Figure 8 After establishing the corresponding business service, it is necessary to evaluate the service's profitability. The loss function for the evaluation uses the Sigmoid function. When the evaluation value is greater than 0.5, it is judged that there is a high probability of obtaining positive profits. Otherwise, it is considered that there is a risk of loss. The function expression is as follows:

[0072]

[0073] For example, by analyzing the business push costs and the revenue brought by the virtual robot corresponding to user A, the revenue valuation ratio is calculated to be 0.8, and thus it is judged to be a high probability positive revenue; by analyzing the business push costs and the revenue brought by the virtual robot corresponding to user B, the revenue valuation ratio is calculated to be 0.3, and thus it is judged to be a high probability negative revenue.

[0074] In summary, risk assessment can effectively avoid unnecessary business pushes, thereby saving costs and providing users with more valuable services.

[0075] Optionally, in the service push method provided in the embodiment of the present application, after pushing the target service to the target user according to the preset push method, the method also includes: obtaining the target user's usage of the target service; if the target user does not use the target service, recalculating the first value until the target user uses the target service; and determining the push effect of the target service based on the recalculated value of the first value.

[0076] For example, consider push effectiveness evaluation: After a service is pushed to a user, real-time and long-term tracking of user usage is required, recorded in a report document. Using the observed data captured in the report document, key data is further extracted and fed into the user information collection step for incremental learning, further enriching user attributes. The next step then involves recalculating and predicting the service push performance. This is continuously refined using real-world data. Ultimately, after the user has used the service, an evaluation of the service push effectiveness, β, is generated, used to assess the effectiveness of the service push. For example, the observed report shows that user A accepted the service push and used it two days later to purchase their desired pair of crystal slippers. Therefore, the push effectiveness evaluation result is β = 0.93, which is in line with expectations. User B hesitated for a long time before accepting the service, and after six months of discussion, used it to purchase a new home. However, due to policy constraints, the purchase fell through. Therefore, the push effectiveness evaluation result is β = 0.23, which is in line with expectations.

[0077] In summary, the combination of virtual online and real offline environments allows for precise service delivery, avoiding user discomfort and resistance. Furthermore, it effectively guides the AI ​​in virtualized scenarios to continuously approach real users, thereby providing data support for the next service delivery.

[0078] Optionally, in the service push method provided in the embodiment of the present application, after determining the push effect of the target service, the method further includes: analyzing the target service based on the push effect of the target service; if it is analyzed that the first user has a demand for the target service and the revenue of the target service meets the preset requirements, then pushing the target service to the first user in a target push manner, wherein the first user is a user other than the target user; if it is detected that the target user fails to use the target service, then determining the reason for the target user's failure to use the target service, and optimizing the target service based on the reason.

[0079] For example, after accurately pushing a service, it is necessary to determine the consequences of the user accepting or rejecting the service. That is, further feedback analysis is conducted on the above service from generation to push and evaluation plan, so as to complete the service push. And the flowchart of updating the service in the system is as follows: Figure 9 For example, by analyzing user A's case, we can find that users A1, A2, and A3 have the same needs. In this case, the recommendation system will adopt different push methods and plans based on their respective user profiles to promote the business. By analyzing user B's case, we can find that real-world policies also need to be an important factor affecting business acceptance.

[0080] Through the above solution, the system for pushing services can be iteratively updated, thereby more accurately expanding the scope of service push.

[0081] In summary, this application can make full use of the user's basic information and personalized attribute descriptions, fully consider the user's unique needs from the user's perspective, and tailor services to meet the user's needs based on these needs. Therefore, this application not only allows users to use it with confidence, but also fundamentally reduces cost waste and achieves good economic benefits. In addition, a statistical comparative analysis with the existing push methods can produce a push effect comparison table, as shown in Table 3. As can be seen from the table, the precise push method is far superior to the traditional push mode. It can not only shorten the push cycle and improve the push effect, but also further expand the benefits, making users more worry-free and the business benefits more stable.

[0082] Table 3 Push effect comparison table

[0083]

[0084] In summary, the service push method provided by the embodiment of the present application obtains a target user portrait, wherein the target user portrait is used to characterize the characteristic information of the target user; constructs a target AI virtual robot based on the target user portrait, and places the target AI virtual robot into a virtual scene; predicts the target user's needs in the real scene based on the needs of the target AI virtual robot in the virtual scene; determines a first service and a target service associated with the first service based on the needs of the target user in the real scene, wherein the first service is the service predicted to be required by the target user; pushes the target service to the target user according to a preset push method, thereby solving the problem of low accuracy of service push in related technologies, which affects the user experience. By predicting the target user's needs in the real scene based on the needs of the constructed target AI virtual robot in the virtual scene, and pushing services to the target user according to a preset push method based on the needs of the target user, the accuracy of service push is improved, thereby improving the user experience.

[0085] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0086] The embodiment of the present application further provides a service push device. It should be noted that the service push device of the embodiment of the present application can be used to execute the service push method provided in the embodiment of the present application. The service push device provided in the embodiment of the present application is introduced below.

[0087] Figure 10 Schematic diagram of a service push device according to an embodiment of the present application. Figure 10 As shown, the apparatus includes: a first acquiring unit 1001 , a first constructing unit 1002 , a first predicting unit 1003 , a first determining unit 1004 and a first pushing unit 1005 .

[0088] Specifically, the first acquiring unit 1001 is configured to acquire a target user portrait, wherein the target user portrait is used to represent characteristic information of the target user;

[0089] The first construction unit 1002 is configured to construct a target AI virtual robot based on the target user portrait and place the target AI virtual robot into a virtual scene;

[0090] The first prediction unit 1003 is used to predict the needs of the target user in the real scene based on the needs of the target AI virtual robot in the virtual scene;

[0091] A first determining unit 1004 is configured to determine a first service and a target service associated with the first service based on the needs of the target user in a real scenario, wherein the first service is a service predicted to be required by the target user;

[0092] The first pushing unit 1005 is configured to push a target service to a target user according to a preset pushing method.

[0093] In summary, the service push device provided in the embodiment of the present application obtains the target user portrait through the first acquisition unit 1001, wherein the target user portrait is used to characterize the characteristic information of the target user; the first construction unit 1002 constructs the target AI virtual robot based on the target user portrait, and places the target AI virtual robot into the virtual scene; the first prediction unit 1003 predicts the needs of the target user in the real scene based on the needs of the target AI virtual robot in the virtual scene; the first determination unit 1004 determines the first service and the target service associated with the first service based on the needs of the target user in the real scene, wherein the first service is the service predicted to be required by the target user; the first push unit 1005 pushes the target service to the target user according to the preset push method, which solves the problem of low accuracy of service push in related technologies, which affects user experience, and predicts the needs of the target user in the real scene based on the needs of the constructed target AI virtual robot in the virtual scene, and pushes the service to the target user according to the preset push method according to the needs of the target user, thereby improving the accuracy of service push and thus improving user experience.

[0094] Optionally, in the service push device provided in the embodiment of the present application, the method for acquiring the target service includes one of the following: a first judgment unit, used to judge whether the first service is a service in the service library, wherein the service library is a collection of multiple services; a first processing unit, used to extract services related to the first service from the service library if the first service is a service in the service library, and obtain the target service from the services related to the first service; a second processing unit, used to generate a second service if the first service is not a service in the service library, and make a decision on the second service, wherein the second service is a service related to the first service; a second determination unit, used to determine whether to use the second service as the target service based on the decision made on the second service.

[0095] Optionally, in the service push device provided in the embodiment of the present application, the first acquisition unit includes: a first acquisition module, used to acquire target information of the target user, wherein the target information is used to represent the attributes of the target user; a first analysis module, used to analyze the target information based on the deep learning network structure to obtain an analysis result; a first processing module, used to label the characteristic information of the target user based on the analysis result to obtain first target information, wherein the first target information is the information after the characteristic information of the target user is labeled; a first construction module, used to construct a user model of the target user based on the first target information; a first acquisition module, used to obtain a portrait of the target user based on the user model of the target user and three-dimensional modeling technology.

[0096] Optionally, in the service push device provided in the embodiment of the present application, the device also includes: a first voting unit, used to, after determining the first service and the target service associated with the first service, vote on the target service by using a decision tree algorithm through multiple AI virtual robots to obtain a voting result; a third determination unit, used to determine the degree of demand of the target user for the target service based on the voting result.

[0097] Optionally, in the service push device provided in the embodiment of the present application, the device also includes: a fourth determination unit, used to determine the cost data corresponding to the target service after determining the degree of demand of the target user for the target service based on the voting results; a first calculation unit, used to calculate a first value based on the cost data corresponding to the target service and the benefits brought by the target AI virtual robot selecting the target service, wherein the first value is used to represent the estimated benefits brought by the target service.

[0098] Optionally, in the service push device provided in the embodiment of the present application, the device also includes: a second acquisition unit, used to obtain the target user's usage of the target service after pushing the target service to the target user according to a preset push method; a second calculation unit, used to recalculate the first value if the target user does not use the target service until the target user uses the target service; and a fifth determination unit, used to determine the push effect of the target service based on the value recalculated from the first value.

[0099] Optionally, in the service push device provided in the embodiment of the present application, the device also includes: a first analysis unit, for analyzing the target service based on the push effect of the target service after determining the push effect of the target service; a third processing unit, for pushing the target service to the first user in a target push manner if it is analyzed that the first user has a demand for the target service and the revenue of the target service meets the preset requirements, wherein the first user is a user other than the target user; a fourth processing unit, for determining the reason for the target user's failure to use the target service if it is detected that the target user fails to use the target service, and optimizing the target service based on the reason.

[0100] The service push device includes a processor and a memory. The above-mentioned first acquisition unit 1001, first construction unit 1002, first prediction unit 1003, first determination unit 1004 and first push unit 1005 are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0101] like Figure 11As shown, the schematic diagram of the service push system includes: a user management module 101, a service management module 102, a service generation and decision module 103, an intelligent service push module 104, a cost budget and estimated revenue module 105, and a web control terminal and an app control terminal 106. Specifically, the user management module 101 is used to collect user information, establish a user model based on the collected user information, characterize the user portrait, and then construct an intelligent AI based on the user model or user portrait, that is, to build an intelligent AI robot; the service management module 102 is used to infer the services required by users; the service generation and decision module 103 is used to search for and generate services, and determine the user's demand for the services, estimate costs, and estimate revenue; the intelligent service push module 104 is used to accurately push services; the cost budget and estimated revenue module 105 is used to evaluate the push effect; and the web control terminal and the app control terminal 106 are used to complete service push.

[0102] In summary, by leveraging the advantages of deep learning algorithms, we overcome the inability to accurately deliver targeted services to customers when using traditional service recommendation methods. We provide an intelligent service push system based on user profiles. This system fully understands user attributes, simulates user thinking based on their unique personality, and deduces their actual needs and preferences. This system then provides targeted services to users, effectively reducing operating costs for financial institutions, conserving human resources, and enhancing user-to-financial institution relationships. Furthermore, this "deep learning-based targeted service push system" primarily includes three functions: First, leveraging the intelligent learning capabilities of deep learning, it deeply mines collected user behavior data to create a virtualized model of the user's personality, generating a user profile that closely reflects their behavior. This profile describes the user's personal habits, preferences, style, beliefs, and values, thereby enabling a multi-dimensional understanding and service of the user. Second, based on the established user virtual profile model, an intelligent AI is constructed that can simulate the user's logical thinking. As more user behavior data is collected, this AI virtual robot will gradually grow. Ultimately, through continuous incremental learning, virtual deduction, and practical feedback, it will gradually approximate the user's actual behavior patterns, thereby characterizing the user's actual business needs in their current state, helping financial institutions expand more business services that meet customer needs. Third, based on the user's actual needs, intelligently push business services that meet their psychological expectations to achieve the goal of precise customer service. In addition, based on the user profile established above, it is easy to understand the unique business needs of each user. The push system then uses a decision tree judgment algorithm to determine whether to provide the user with the corresponding business push. Therefore, this push method can effectively avoid user resistance, thereby reducing push costs and achieving the expected benefits.

[0103] The processor contains a kernel, which retrieves the corresponding program unit from the memory. You can set one or more kernels, and adjust the kernel parameters to improve the accuracy of service push.

[0104] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0105] An embodiment of the present invention provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the service push method is implemented.

[0106] An embodiment of the present invention provides a processor, which is used to run a program, wherein the service push method is executed when the program is running.

[0107] like Figure 12 As shown, an embodiment of the present invention provides an electronic device, which includes a processor, a memory, and a program stored in the memory and runnable on the processor. When the processor executes the program, the following steps are implemented: obtaining a target user portrait, wherein the target user portrait is used to characterize the characteristic information of the target user; constructing a target AI virtual robot based on the target user portrait, and placing the target AI virtual robot into a virtual scene; predicting the needs of the target user in a real scene based on the needs of the target AI virtual robot in the virtual scene; determining a first business and a target business associated with the first business based on the needs of the target user in the real scene, wherein the first business is the business predicted to be required by the target user; and pushing the target business to the target user according to a preset push method.

[0108] When the processor executes the program, the following steps are also implemented: the target business is obtained in one of the following ways: determining whether the first business is a business in a business library, wherein the business library is a collection of multiple businesses; if the first business is a business in the business library, extracting businesses related to the first business from the business library, and obtaining the target business from businesses related to the first business; if the first business is not a business in the business library, generating a second business and making a decision on the second business, wherein the second business is a business related to the first business; and determining whether to use the second business as the target business based on the decision made on the second business.

[0109] When the processor executes the program, the following steps are also implemented: obtaining a target user portrait includes: obtaining target information of the target user, wherein the target information is used to represent the attributes of the target user; analyzing the target information based on a deep learning network structure to obtain an analysis result; labeling the characteristic information of the target user based on the analysis result to obtain first target information, wherein the first target information is the information after the characteristic information of the target user is labeled; constructing a user model of the target user based on the first target information; obtaining the target user portrait based on the user model of the target user and three-dimensional modeling technology.

[0110] When the processor executes the program, the following steps are also implemented: after determining the first business and the target business associated with the first business, the method also includes: using a decision tree algorithm through multiple AI virtual robots to vote on the target business to obtain a voting result; based on the voting result, determine the target user's demand for the target business.

[0111] When the processor executes the program, the following steps are also implemented: after determining the degree of demand of the target user for the target business based on the voting results, the method also includes: determining the cost data corresponding to the target business; calculating a first value based on the cost data corresponding to the target business and the benefits brought by the target AI virtual robot selecting the target business, wherein the first value is used to represent the estimated benefits brought by the target business.

[0112] When the processor executes the program, the following steps are also implemented: after pushing the target service to the target user according to a preset push method, the method also includes: obtaining the target user's usage of the target service; if the target user does not use the target service, recalculating the first value until the target user uses the target service; and determining the push effect of the target service based on the recalculated value of the first value.

[0113] When the processor executes the program, the following steps are also implemented: after determining the push effect of the target service, the method further includes: analyzing the target service based on the push effect of the target service; if it is analyzed that the first user has a demand for the target service and the revenue of the target service meets the preset requirements, then pushing the target service to the first user in a target push manner, wherein the first user is a user other than the target user; if it is detected that the target user fails to use the target service, then determining the reason for the target user's failure to use the target service, and optimizing the target service based on the reason. The device in this article can be a server, PC, PAD, mobile phone, etc.

[0114] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program initialized with the following method steps: obtaining a target user portrait, wherein the target user portrait is used to characterize the characteristic information of the target user; constructing a target AI virtual robot based on the target user portrait, and placing the target AI virtual robot into a virtual scene; predicting the needs of the target user in a real scene based on the needs of the target AI virtual robot in the virtual scene; determining a first business and a target business associated with the first business based on the needs of the target user in the real scene, wherein the first business is the business predicted to be required by the target user; and pushing the target business to the target user according to a preset push method.

[0115] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: the target business is obtained in one of the following ways: judging whether the first business is a business in a business library, wherein the business library is a collection of multiple businesses; if the first business is a business in the business library, extracting businesses related to the first business from the business library, and obtaining the target business from businesses related to the first business; if the first business is not a business in the business library, generating a second business and making a decision on the second business, wherein the second business is a business related to the first business; and determining whether to use the second business as the target business based on the decision made on the second business.

[0116] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: obtaining a target user portrait includes: obtaining target information of the target user, wherein the target information is used to represent the attributes of the target user; analyzing the target information based on a deep learning network structure to obtain an analysis result; labeling the characteristic information of the target user based on the analysis result to obtain first target information, wherein the first target information is the information after the characteristic information of the target user is labeled; constructing a user model of the target user based on the first target information; obtaining the target user portrait based on the user model of the target user and three-dimensional modeling technology.

[0117] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: after determining a first business and a target business associated with the first business, the method also includes: using a decision tree algorithm through multiple AI virtual robots to vote on the target business to obtain a voting result; based on the voting result, determining the target user's demand for the target business.

[0118] When executed on a data processing device, it is also suitable for executing an initialized program having the following method steps: after determining the degree of demand of the target user for the target business based on the voting results, the method also includes: determining the cost data corresponding to the target business; calculating a first numerical value based on the cost data corresponding to the target business and the benefits brought by the target AI virtual robot selecting the target business, wherein the first numerical value is used to represent the estimated benefits brought by the target business.

[0119] When executed on a data processing device, it is also suitable for executing a program that is initialized with the following method steps: after pushing the target service to the target user in a preset push method, the method also includes: obtaining the target user's usage of the target service; if the target user does not use the target service, recalculating the first value until the target user uses the target service; and determining the push effect of the target service based on the recalculated value of the first value.

[0120] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: after determining the push effect of the target service, the method also includes: analyzing the target service based on the push effect of the target service; if it is analyzed that the first user has a demand for the target service and the revenue of the target service meets the preset requirements, then pushing the target service to the first user in a target push manner, wherein the first user is a user other than the target user; if it is detected that the target user fails to use the target service, then determining the reason for the target user's failure to use the target service, and optimizing the target service based on the reason.

[0121] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0122] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0123] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0124] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0125] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0126] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0127] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules 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 technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0128] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0129] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0130] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A service push method, characterized in that: include: Obtaining a target user portrait, wherein the target user portrait is used to represent characteristic information of the target user; Building a target AI virtual robot based on the target user portrait, and placing the target AI virtual robot into a virtual scene; Predicting the needs of the target user in the real scene based on the needs of the target AI virtual robot in the virtual scene; Determining a first service and a target service associated with the first service according to the needs of the target user in a real scenario, wherein the first service is a service predicted to be required by the target user; Push the target service to the target user according to a preset push method; Methods for acquiring the target business include: Determining whether the first service is a service in a service library, wherein the service library is a collection of multiple services; If the first service is a service in the service library, services related to the first service are extracted from the service library, and the target service is obtained from the services related to the first service.

2. The method according to claim 1, characterized in that The target business acquisition method further includes: If the first service is not a service in the service library, generating a second service and making a decision on the second service, wherein the second service is a service related to the first service; According to the decision made on the second service, it is determined whether to use the second service as the target service.

3. The method according to claim 1, characterized in that Obtaining target user portraits includes: Acquire target information of a target user, wherein the target information is used to represent attributes of the target user; Analyzing the target information according to the deep learning network structure to obtain an analysis result; Labeling the characteristic information of the target user according to the analysis result to obtain first target information, wherein the first target information is information obtained by labeling the characteristic information of the target user; Constructing a user model of the target user based on the first target information; A portrait of the target user is obtained based on the user model of the target user and three-dimensional modeling technology.

4. The method according to claim 1, wherein After determining the first service and the target service associated with the first service, the method further includes: Using a decision tree algorithm, multiple AI virtual robots vote on the target business to obtain voting results; The degree of demand of the target user for the target service is determined based on the voting result.

5. The method according to claim 4, characterized in that After determining the target user's demand for the target service based on the voting result, the method further includes: Determining cost data corresponding to the target business; A first value is calculated based on the cost data corresponding to the target business and the revenue brought by the target AI virtual robot selecting the target business, wherein the first value is used to represent the estimated revenue brought by the target business.

6. The method according to claim 5, characterized in that After pushing the target service to the target user in a preset push manner, the method further includes: Obtaining usage of the target service by the target user; If the target user does not use the target service, recalculating the first value until the target user uses the target service; The push effect of the target service is determined according to a value obtained by recalculating the first value.

7. The method according to claim 6, characterized in that After determining the push effect of the target service, the method further includes: Analyze the target business based on the push effect of the target business; If it is analyzed that the first user has a demand for the target service and the revenue of the target service meets the preset requirements, the target service is pushed to the first user in a targeted push manner, wherein the first user is a user other than the target user; If it is detected that the target user fails to use the target service, the reason why the target user fails to use the target service is determined, and the target service is optimized according to the reason.

8. A service push device, characterized in that: include: A first acquiring unit is configured to acquire a target user portrait, wherein the target user portrait is used to represent characteristic information of the target user; A first construction unit is configured to construct a target AI virtual robot based on the target user portrait and place the target AI virtual robot into a virtual scene; A first prediction unit is configured to predict the needs of the target user in the real scene based on the needs of the target AI virtual robot in the virtual scene; A first determining unit is configured to determine a first service and a target service associated with the first service according to the needs of the target user in a real scenario, wherein the first service is a service predicted to be required by the target user; A first pushing unit, configured to push the target service to the target user according to a preset pushing method; Methods for acquiring the target business include: a first determining unit, configured to determine whether the first service is a service in a service library, wherein the service library is a collection of multiple services; The first processing unit is configured to extract services related to the first service from the service library if the first service is a service in the service library, and obtain the target service from the services related to the first service.

9. A computer-readable storage medium, characterized in that The storage medium includes a stored program, wherein the program executes the service push method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: It includes one or more processors and a memory, the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the service push method described in any one of claims 1 to 7.

Images