Information recommendation method and device, electronic equipment, storage medium and program product

Abstract

The invention provides an information recommendation method and device, electronic equipment, a storage medium and a program product. The information recommendation method comprises the steps that N to-be-recommended objects are screened, target to-be-recommended objects meeting recommendation conditions are obtained, and N is a positive integer larger than 1; a target information recommendation mode corresponding to the target to-be-recommended object is determined from M information recommendation modes, and M is a positive integer larger than 1; and according to the target information recommendation mode, performing information recommendation on the target to-be-recommended object. Through the method, the information recommendation effect can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of information processing technology, and in particular to an information recommendation method, device, electronic device, storage medium and program product. Background Art

[0002] With the rapid development of the Internet, people's lives have been greatly facilitated, but also a huge amount of data information has been brought in. In the recommendation scenario, the recommendation system can provide users with interesting content based on the information recommendation model, and recommend online resources to those users who may be interested in it. Summary of the invention

[0003] The embodiments of the present application provide an information recommendation method, device, electronic device, computer-readable storage medium, and computer program product, which can effectively improve the information recommendation effect.

[0004] The technical solution of the embodiment of the present application is implemented as follows:

[0005] The present application provides an information recommendation method, including:

[0006] Screening N objects to be recommended to obtain target objects to be recommended that meet the recommendation conditions, where N is a positive integer greater than 1;

[0007] Determine a target information recommendation method corresponding to the target object to be recommended from M information recommendation methods, where M is a positive integer greater than 1;

[0008] According to the target information recommendation method, information is recommended to the target object to be recommended.

[0009] The present application provides an information recommendation device, including:

[0010] A screening module is used to screen N objects to be recommended to obtain target objects to be recommended that meet the recommendation conditions, where N is a positive integer greater than 1;

[0011] A determination module, used to determine a target information recommendation method corresponding to the target object to be recommended from M information recommendation methods, where M is a positive integer greater than 1;

[0012] The recommendation module is used to recommend information to the target object to be recommended according to the target information recommendation method.

[0013] An embodiment of the present application provides an electronic device, including:

[0014] A memory for storing computer executable instructions or computer programs;

[0015] The processor is used to implement the information recommendation method provided in the embodiment of the present application when executing the computer executable instructions or computer programs stored in the memory.

[0016] An embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions for causing a processor to execute the instructions to implement the information recommendation method provided in the embodiment of the present application.

[0017] The embodiment of the present application provides a computer program product, which includes a computer program or a computer executable instruction, and the computer program or the computer executable instruction is stored in a computer-readable storage medium. The processor of the electronic device reads the computer executable instruction from the computer-readable storage medium, and the processor executes the computer executable instruction, so that the electronic device performs the information recommendation method described in the embodiment of the present application.

[0018] The embodiments of the present application have the following beneficial effects:

[0019] By screening N objects to be recommended, a target object to be recommended that meets the recommendation conditions is obtained, and a target information recommendation method corresponding to the target object to be recommended is determined from M information recommendation methods, and information recommendation is performed on the target object to be recommended according to the target information recommendation method. In this way, by determining the target information recommendation method corresponding to the target object to be recommended from M information recommendation methods, diversified recommendations are performed through M different information recommendation methods, and the information recommendation strength is effectively improved. By determining the target information recommendation method corresponding to the target object to be recommended from M information recommendation methods, a customized target information recommendation method is provided for the target object to be recommended. By screening N objects to be recommended, a target object to be recommended that meets the recommendation conditions is obtained, and the interference of redundant objects to be recommended on the information recommendation strength is effectively reduced, thereby effectively improving the information recommendation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the architecture of the information recommendation system provided in the embodiment of the present application;

[0021] Figure 2 is a schematic diagram of the structure of an electronic device for information recommendation provided in an embodiment of the present application;

[0022] Figure 3 This is a schematic diagram of the process of the information recommendation method provided in the embodiment of the present application. Figure 1 ;

[0023] Figure 4 It is a schematic diagram of the principle of the componentized marketing system provided in the embodiment of the present application;

[0024] Figure 5It is a schematic diagram of the principle of the component life cycle provided by the embodiment of the present application;

[0025] Figure 6 It is a schematic diagram of the principle of operation of a single component provided in an embodiment of the present application;

[0026] Figure 7 This is a schematic diagram of the process of the information recommendation method provided in the embodiment of the present application. Figure 2 . DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0028] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0029] In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present application described here can be implemented in an order other than that illustrated or described here.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0031] Before further describing the embodiments of the present application in detail, the nouns and terms involved in the embodiments of the present application are explained. The nouns and terms involved in the embodiments of the present application are subject to the following interpretations.

[0032] 1) Information recommendation: Information recommendation refers to a method of providing personalized information recommendations to users by analyzing personal information such as user preferences and search history, combined with real-time recommendation conditions. These recommended information can include products, content, services, social contacts, etc., aiming to help users find the content they are interested in and need in the complex sea of ​​information. Information recommendation is usually based on certain recommendation algorithms and models, and provides users with personalized information recommendation services through collaborative filtering, content-based, hybrid recommendation and other methods.

[0033] 2) Information recommendation method: refers to the method of providing personalized information recommendations to users by analyzing personal information such as preference settings and search history, combined with real-time recommendation conditions. These recommended information can include products, content, services, social contacts, etc., aiming to help users find the content they are interested in and need in the complex sea of ​​information.

[0034] Information recommendation methods include telephone recommendation, SMS recommendation, APP recommendation, email recommendation, social media recommendation, search engine recommendation, etc. Telephone recommendation is a recommendation method based on telephone communication, which recommends products, services or information to users through telephone calls. Telephone recommendation is usually completed by manual customer service or automatic telephone system, which can provide personalized recommendations based on users' personal information and needs. SMS recommendation refers to recommending relevant information to users by sending SMS. This method is usually used in marketing promotion, news information, event invitations and other scenarios. SMS recommendation can quickly convey recommended information to specific user groups, but users may resist such push, and attention should be paid to the quality and frequency of SMS content. Recommend relevant information to users through mobile applications. This method can use the personalized recommendation algorithm of APP to provide users with customized recommendation content based on users' historical behavior and preference settings. APP recommendation is a recommendation service provided in mobile applications. It can recommend other related applications or functions to users based on users' usage habits, interests and behavior data. APP recommendation can better meet users' personalized needs and improve user experience, but it also needs to balance the quality of recommended content and user privacy protection. Email recommendation is a recommendation method that recommends products, services or information to users through email communication. Email recommendations usually include personalized email promotions, newsletters or other relevant recommended content. Social media recommendations are a recommendation service provided on social media platforms that can recommend other relevant content, products or services to users based on their interests, interactions and behavior data. Search engine recommendations are search engines that recommend relevant search results, ads or other relevant information to users based on their search history, keywords and query intent.

[0035] 3) Client: An application running in a terminal to provide various services, such as a video player client, a game client, etc.

[0036] 4) Application Programming Interface (API): Also known as the application programming interface, it is the agreement for connecting different components of the software system. As the scale of software has become increasingly large in recent years, it is often necessary to divide complex systems into small components. The design of programming interfaces is very important. In the practice of programming, the design of programming interfaces must first ensure that the responsibilities of the software system are reasonably divided. Good interface design can reduce the mutual dependence of various parts of the system, improve the cohesion of the components, and reduce the degree of coupling between the components, thereby improving the maintainability and scalability of the system.

[0037] 5) Component: Also known as software component, a software component is a basic unit in a software system. It is a reusable software module that usually has specific functions or responsibilities. Software components can interact with other software units through interfaces and can be integrated into larger software systems. The characteristics of software components include: Reusability: Software components can be reused in different software systems, thereby improving development efficiency and code reusability. Independence: Software components usually have independent responsibilities and functions, can be independently developed and tested, and are not affected by other components. Interface: Software components interact with other components through well-defined interfaces, which helps reduce the coupling between components and improve the flexibility and maintainability of the system. Configurability: Software components can usually configure their behaviors and properties to adapt to different application scenarios and requirements. Pluggability: Software components can be easily inserted or removed without affecting the functions of other components. Types of software components include: Class Library: Provides a series of classes and interfaces for developers to reuse. Service: Provides specific functions or business logic, usually called through the network. Library: A collection of pre-written code that provides specific functionality or tools. Framework: Provides an infrastructure and template to help developers quickly build applications. Microservice: An architectural style that develops and deploys applications as a collection of small services.

[0038] 6) Kafka: It is a high-performance, scalable, publish-subscribe based message queue system maintained by the Apache Software Foundation. It is mainly used to process large-scale data streams and supports high throughput and persistence.

[0039] During the implementation of the embodiments of the present application, the applicant discovered that the related technology has the following problems:

[0040] In the related art, information recommendation is usually performed through a single information recommendation method. Since the information recommendation strength of the single information recommendation method is weak, the information recommendation effect is low.

[0041] The embodiments of the present application provide an information recommendation method, device, electronic device, computer-readable storage medium, and computer program product, which can effectively improve the information recommendation effect. The following describes an exemplary application of the information recommendation system provided by the embodiments of the present application.

[0042] See also Figure 1 , Figure 1 It is a schematic diagram of the architecture of the information recommendation system 100 provided in an embodiment of the present application. The terminal (terminal 400 is shown as an example) is connected to the server 200 via the network 300. The network 300 can be a wide area network or a local area network, or a combination of the two.

[0043] The terminal 400 is used for the user to use the client 410 to display the target object to be recommended on the graphical interface 410-1 (graphic interface 410-1 is shown as an example). The terminal 400 and the server 200 are connected to each other via a wired or wireless network.

[0044] In some embodiments, the server 200 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The terminal 400 may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart TV, a smart watch, a car terminal, etc., but is not limited thereto. The electronic device provided in the embodiment of the present application may be implemented as a terminal or as a server. The terminal and the server may be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present application.

[0045] In some embodiments, the server 200 screens N objects to be recommended, obtains target objects to be recommended that meet the recommendation conditions, determines the target information recommendation method from M information recommendation methods, and sends the target information recommendation method to the terminal 400. The terminal 400 recommends information to the target objects to be recommended according to the target information recommendation method.

[0046] In other embodiments, the terminal 400 screens N objects to be recommended, obtains target objects to be recommended that meet the recommendation conditions, determines a target information recommendation method from M information recommendation methods, and sends the target information recommendation method to the server 200. The server 200 recommends information to the target objects to be recommended according to the target information recommendation method.

[0047] In other embodiments, the embodiments of the present application can be implemented with the aid of cloud technology. Cloud technology refers to a hosting technology that unifies a series of resources such as hardware, software, and networks within a wide area network or a local area network to achieve data computing, storage, processing, and sharing.

[0048] Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, and application technology based on the cloud computing business model. It can form a resource pool that can be used on demand and is flexible and convenient. Cloud computing technology will become an important support. The background services of the technical network system require a large amount of computing and storage resources.

[0049] See also Figure 2 , Figure 2 is a schematic diagram of the structure of an electronic device 500 for information recommendation provided in an embodiment of the present application, wherein: Figure 2 The electronic device 500 shown may be Figure 1 The server 200 or the terminal 400 in Figure 2 The electronic device 500 shown includes: at least one processor 410, a memory 450, and at least one network interface 420. The various components in the electronic device 500 are coupled together via a bus system 440. It is understood that the bus system 440 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 440 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, the bus system 440 is not described in detail. Figure 2 Various buses are labeled as bus system 440 .

[0050] Processor 410 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., where the general-purpose processor can be a microprocessor or any conventional processor, etc.

[0051] The memory 450 may be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state memory, hard disk drives, optical disk drives, etc. The memory 450 may optionally include one or more storage devices that are physically remote from the processor 410.

[0052] The memory 450 includes a volatile memory or a non-volatile memory, and may also include both volatile and non-volatile memories. The non-volatile memory may be a read-only memory (ROM), and the volatile memory may be a random access memory (RAM). The memory 450 described in the embodiments of the present application is intended to include any suitable type of memory.

[0053] In some embodiments, memory 450 can store data to support various operations, examples of which include programs, modules, and data structures, or a subset or superset thereof, as exemplarily described below.

[0054] Operating system 451, including system programs for processing various basic system services and performing hardware-related tasks, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and processing hardware-based tasks;

[0055] The network communication module 452 is used to reach other electronic devices via one or more (wired or wireless) network interfaces 420. Exemplary network interfaces 420 include: Bluetooth, Wireless Fidelity (WiFi), and Universal Serial Bus (USB).

[0056] In some embodiments, the information recommendation device provided in the embodiments of the present application can be implemented in software. Figure 2 The information recommendation device 455 stored in the memory 450 is shown, which can be software in the form of a program and a plug-in, etc., and includes the following software modules: a screening module 4551, a determination module 4552, and a recommendation module 4553. These modules are logical, and therefore can be arbitrarily combined or further split according to the functions implemented. The functions of each module will be described below.

[0057] In other embodiments, the information recommendation device provided in the embodiments of the present application can be implemented in hardware. As an example, the information recommendation device provided in the embodiments of the present application can be a processor in the form of a hardware decoding processor, which is programmed to execute the information recommendation method provided in the embodiments of the present application. For example, the processor in the form of a hardware decoding processor can adopt one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field programmable gate arrays (FPGAs), or other electronic components.

[0058] In some embodiments, the terminal or server can implement the information recommendation method provided in the embodiments of the present application by running a computer program or a computer executable instruction. For example, the computer program can be a native program (e.g., a dedicated information recommendation program) or a software module in the operating system, for example, an information recommendation module that can be embedded in any program (such as an instant messaging client, a photo album program, an electronic map client, a navigation client); for example, it can be a native application (APP, Application), that is, a program that needs to be installed in the operating system to run. In short, the above-mentioned computer program can be any form of application, module or plug-in.

[0059] The information recommendation method provided in the embodiment of the present application will be described in combination with the exemplary application and implementation of the server or terminal provided in the embodiment of the present application.

[0060] See also Figure 3 , Figure 3 is a flowchart of the information recommendation method provided in the embodiment of the present application, which will be combined with Figure 3 Steps 101 to 103 are shown for illustration. The information recommendation method provided in the embodiment of the present application can be implemented by a server or a terminal alone, or by a server and a terminal in collaboration. The following description will be made by taking the implementation by a server alone as an example.

[0061] In step 101, N objects to be recommended are screened to obtain target objects to be recommended that meet the recommendation conditions.

[0062] In some embodiments, N is a positive integer greater than 1.

[0063] As an example, the N objects to be recommended include object A to be recommended, object B to be recommended and object C to be recommended, and the recommendation condition may be that the object to be recommended has a credit limit, wherein, if object A to be recommended has a credit limit, then object A to be recommended meets the recommendation condition, object B to be recommended does not have a credit limit, then object B to be recommended does not meet the recommendation condition, and object C to be recommended does not have a credit limit, then object C to be recommended does not meet the recommendation condition.

[0064] In some embodiments, the above step 101 can be implemented in the following manner: for each object to be recommended, the object information of the object to be recommended is compared with the recommendation condition to obtain the condition comparison result of the object to be recommended; if the condition comparison result indicates that the object information of the object to be recommended meets the recommendation condition, the object to be recommended is determined as the target object to be recommended.

[0065] In some embodiments, determining whether the above-mentioned object to be recommended is the target object to be recommended can be achieved through a decision table component. The decision table component determines whether the object to be recommended is the target object to be recommended, and then transmits it to the downstream component through kafka asynchronously for subsequent processing.

[0066] Continuing with the above example, for the recommended object A, the object information of the recommended object A indicates that the recommended object A has a credit limit, and the recommendation condition may be that the recommended object has a credit limit. Then the condition comparison result of the recommended object A indicates that the object information of the recommended object A satisfies the recommendation condition, and the recommended object A is determined as the target recommended object.

[0067] In this way, by comparing the information of the object to be recommended with the recommendation conditions, the target object that meets the user's needs or scenario requirements can be accurately screened out, improving the accuracy and relevance of the recommendation. When the information or products required by users can be accurately provided, the user's satisfaction and experience can be significantly improved, thereby enhancing the user's loyalty to the service, and the recommendation conditions can be efficiently processed and matched, which can optimize resource utilization and reduce computing costs. For shopping, recruitment, matching and other services, providing qualified target objects can help make decisions faster and improve the conversion rate of transactions. The recommendation conditions can be quickly adjusted according to market changes and changes in user needs, so that the recommendation results are always timely and adaptable.

[0068] In some embodiments, if the condition comparison result indicates that the object information of the object to be recommended does not meet the recommendation condition, the object to be recommended is determined as a reference object to be recommended, and a prompt message is sent to the reference object to be recommended.

[0069] In some embodiments, the above-mentioned prompt information is used to prompt the reference object to be recommended to update the object information of the reference object to be recommended to obtain updated object information that meets the recommendation condition.

[0070] In some embodiments, after obtaining updated object information from the reference object to be recommended, the next round of screening for the object to be recommended can be continued or participated in, so that the reference object to be recommended is re-determined as the target object to be recommended, so that the reference object to be recommended can re-participate in the information recommendation process.

[0071] Continuing with the above example, if the recommended object B does not have a credit limit, then the recommended object B does not meet the recommendation conditions. The recommended object B is determined as the reference recommended object, and a prompt message is sent to the reference recommended object (the recommended object B) to prompt the reference recommended object (the recommended object B) to update the object information of the reference recommended object to obtain updated object information that meets the recommendation conditions, for example, the recommended object B completes the credit limit assessment and obtains the information limit.

[0072] In this way, a feedback mechanism is established by sending prompt information to the recommended objects that do not meet the recommendation conditions. This gives the objects the opportunity to update their information according to the conditions of the recommendation system, thereby increasing the possibility of being selected in the future. After receiving the prompt information, the reference recommended objects can dynamically adjust their object information based on this information. This dynamic optimization helps the objects better adapt to the screening criteria of the recommendation system and improve their competitiveness in the recommendation list. The user participation link is increased. Users are no longer passively accepting recommendations, but can actively adjust their information to meet the recommendation conditions of the system, thereby obtaining more satisfactory recommendation results. Through the prompt information, the reference recommended objects can more accurately understand the requirements of the recommendation system, thereby providing more satisfactory object information and improving the overall accuracy and efficiency of the recommendation system. The prompt information can help the reference recommended objects better adapt to system changes and demand changes, making the recommendation system more adaptable and flexible.

[0073] In step 102, a target information recommendation method corresponding to the target object to be recommended is determined from M information recommendation methods.

[0074] In some embodiments, M is a positive integer greater than 1.

[0075] In some embodiments, the above-mentioned information recommendation method refers to a recommendation method that provides personalized information recommendation to users by analyzing personal information such as preference settings and search history, combined with real-time recommendation conditions. These recommended information may include commodities, content, services, social contacts, etc., aiming to help users find content that they are interested in and need in a complex sea of ​​information. Information recommendation methods include SMS recommendation, APP recommendation, email recommendation, social media recommendation, search engine recommendation, etc. SMS recommendation refers to recommending relevant information to users by sending SMS, which is usually used in marketing promotion, news information, event invitations and other scenarios. SMS recommendation can quickly convey recommended information to specific user groups, but users may resist such push and need to pay attention to the quality and frequency of SMS content. Recommend relevant information to users through mobile applications. This method can use the personalized recommendation algorithm of APP to provide users with customized recommended content based on the user's historical behavior and preference settings. APP recommendation can better meet the personalized needs of users and improve user experience, but it also needs to balance the quality of recommended content and user privacy protection.

[0076] In some embodiments, the above step 102 can be implemented as follows: if the total number of the target objects to be recommended is one, then the M types of information recommendation methods are determined as the target information recommendation methods corresponding to the target objects to be recommended; if the total number of the target objects to be recommended is multiple, then based on the number of the information recommendation methods and the total number of the target objects to be recommended, the target information recommendation method corresponding to each of the target objects to be recommended is determined.

[0077] In some embodiments, if the total number of target objects to be recommended is one, it means that there is only one target object to be recommended among the N objects to be recommended. Then, each information recommendation method can be determined as the target information recommendation method corresponding to the target object to be recommended, so that repeated recommendations for the target object to be recommended can be achieved through M information recommendation methods, thereby effectively improving the recommendation effect.

[0078] As an example, the M types of information recommendation methods include SMS recommendation method, APP recommendation method, email recommendation method, social media recommendation method and search engine recommendation method. Then the SMS recommendation method, APP recommendation method, email recommendation method, social media recommendation method and search engine recommendation method can be determined as the target information recommendation method corresponding to the target object to be recommended, so that information recommendation can be made to the target object to be recommended through SMS recommendation method, APP recommendation method, email recommendation method, social media recommendation method and search engine recommendation method respectively.

[0079] In this way, if the total number of target objects to be recommended is one, then each information recommendation method can be determined as the target information recommendation method corresponding to the target object to be recommended, so that repeated recommendations for the target object to be recommended can be achieved through M information recommendation methods, thereby effectively improving the recommendation effect.

[0080] In some embodiments, if the total number of the target objects to be recommended is multiple, the target information recommendation method corresponding to each target object to be recommended is determined based on the number of the information recommendation methods and the total number of the target objects to be recommended.

[0081] In some embodiments, the number of information recommendation methods is M, where M is a positive integer greater than 1, the total number of target objects to be recommended is at least one, and the target objects to be recommended correspond to at least one target information recommendation method. The target information recommendation methods corresponding to different target objects to be recommended may be the same or different.

[0082] As an example, the target information recommendation method corresponding to the target object to be recommended A is SMS recommendation, the target information recommendation method corresponding to the target object to be recommended B is SMS recommendation, and the target information recommendation method corresponding to the target object to be recommended C is email recommendation. Then the target information recommendation methods corresponding to the target object to be recommended A and the target object to be recommended B are the same, the target information recommendation methods corresponding to the target object to be recommended A and the target object to be recommended C are different, and the target information recommendation methods corresponding to the target object to be recommended B and the target object to be recommended C are the same.

[0083] In some embodiments, the above-mentioned determination of the target information recommendation method corresponding to each target object to be recommended based on the number of information recommendation methods and the total number of target objects to be recommended can be achieved in the following manner: comparing the number of the information recommendation methods and the total number of the target objects to be recommended to obtain a comparison result; if the comparison result indicates that the number of the information recommendation methods is greater than or equal to the total number of the target objects to be recommended, then for each target object to be recommended, at least one of the information recommendation methods is determined as the target information recommendation method corresponding to the target object to be recommended; if the comparison result indicates that the number of the information recommendation methods is less than the total number of the target objects to be recommended, then the target objects to be recommended are divided to obtain object sets corresponding to each of the information recommendation methods; for each object set, the information recommendation method corresponding to the object set is determined as the target information recommendation method corresponding to the target object to be recommended in the object set.

[0084] In some embodiments, the object set includes at least one target object to be recommended.

[0085] In some embodiments, if the comparison result indicates that the number of information recommendation methods is greater than or equal to the total number of target objects to be recommended, it means that the same target object to be recommended needs to reuse different information recommendation methods to achieve effective use of the full amount of information recommendation methods, so that each target object to be recommended can correspond to at least one information recommendation method, and different target objects to be recommended can correspond to different information recommendation methods. Then, for each target object to be recommended, at least one information recommendation method can be determined as the target information recommendation method corresponding to the target object to be recommended. For example, one information recommendation method is determined as the target information recommendation method corresponding to the target object to be recommended. Then, since the number of information recommendation methods is greater than or equal to the total number of target objects to be recommended, each target object to be recommended corresponds to one information recommendation method, and different target objects to be recommended correspond to different information recommendation methods. At this time, there is an information recommendation method that does not perform information recommendation, that is, the effective use of the full amount of information recommendation methods is not achieved, but the normal recommendation of each target object to be recommended is guaranteed.

[0086] As an example, for the target object to be recommended A, the SMS recommendation method is determined as the target information recommendation method corresponding to the target object to be recommended A; for the target object to be recommended B, the APP recommendation method is determined as the target information recommendation method corresponding to the target object to be recommended B; for the target object to be recommended B, the email recommendation method is determined as the target information recommendation method corresponding to the target object to be recommended C.

[0087] In some embodiments, if the comparison result indicates that the number of information recommendation methods is less than the total number of target objects to be recommended, it means that the same target objects to be recommended do not need to reuse different information recommendation methods to achieve effective utilization of all information recommendation methods. At this time, the target objects to be recommended can be divided to obtain object sets corresponding to each information recommendation method, so that the same target objects to be recommended do not need to reuse different information recommendation methods, that is, the same target object to be recommended only corresponds to one information recommendation method, and effective utilization of all information recommendation methods can be achieved.

[0088] As an example, the target objects to be recommended include target object A, target object B and target object C, the information recommendation method includes SMS recommendation method and APP recommendation method, and the target objects to be recommended are divided to obtain the object set corresponding to the SMS recommendation method {target object A, target object B}, and the object set corresponding to the APP recommendation method {target object C}.

[0089] Continuing with the above example, for the object set {target object to be recommended A, target object to be recommended B} corresponding to the SMS recommendation method, the SMS recommendation method corresponding to the object set is determined as the target information recommendation method corresponding to the target object to be recommended A and the target object to be recommended B in the object set. For the object set corresponding to the APP recommendation method, the APP recommendation method corresponding to the object set is determined as the target information recommendation method corresponding to the target object to be recommended C in the object set.

[0090] As an example, the target objects to be recommended include target object A, target object B and target object C, the information recommendation method includes SMS recommendation method and APP recommendation method, the target objects to be recommended are divided to obtain the object set corresponding to the SMS recommendation method {target object C}, and the object set corresponding to the APP recommendation method {target object A, target object B}.

[0091] In this way, by comparing the number of information recommendation methods with the total number of target objects to be recommended, the complexity of the recommendation process can be effectively controlled. If the number of information recommendation methods is greater than or equal to the total number of target objects to be recommended, it means that each target object can correspond to at least one information recommendation method, which can ensure that each object can be appropriately recommended. If the number of information recommendation methods is less than the total number of target objects to be recommended, the system will divide the target objects to be recommended to ensure that each information recommendation method has a corresponding set of objects. This can avoid too many objects waiting without enough recommendation methods, thereby optimizing resource allocation. Ensuring that each target object to be recommended has a corresponding target information recommendation method can improve the accuracy and coverage of recommendations. Each object has the opportunity to be discovered and selected by the user through at least one recommendation method. When the number of information recommendation methods is not enough to cover all target objects to be recommended, by dividing the objects, it is ensured that each recommendation method can be fully utilized, which helps to provide effective recommendations under limited resources.

[0092] In some embodiments, the above-mentioned division of the target objects to be recommended to obtain the object sets corresponding to each information recommendation method can be achieved in the following ways: based on the weight of each information recommendation method and the total number of the target objects to be recommended, determine the target number of the target objects to be recommended that need to be recommended by each information recommendation method; divide the target objects to be recommended according to the target number corresponding to each information recommendation method to obtain the object sets corresponding to each information recommendation method.

[0093] In some embodiments, the number of target objects to be recommended in the object set is equal to the corresponding target number, the object set includes at least one target object to be recommended, each information recommendation method corresponds to an object set, and the object set corresponds to the information recommendation method one by one.

[0094] In some embodiments, the sum of the weights of each information recommendation method is equal to 1. For example, the weight of the information recommendation method is positively correlated with the importance of the information recommendation method, and the weight of the information recommendation method is positively correlated with the number of objects to be recommended that the information recommendation method needs to undertake.

[0095] As an example, the weight of the SMS recommendation method is 0.1, the weight of the APP recommendation method is 0.9, and the total number of target objects to be recommended is 100. Then, the target number of target objects to be recommended by the SMS recommendation method is 10, and the target number of target objects to be recommended by the APP recommendation method is 90. According to the target numbers corresponding to the SMS recommendation method and the APP recommendation method, the target objects to be recommended are divided to obtain the object set corresponding to the SMS recommendation method {target object to be recommended 1, target object to be recommended 2 ... target object to be recommended 10}, and the object set corresponding to the APP recommendation method {target object to be recommended 11, target object to be recommended 12 ... target object to be recommended 100}.

[0096] In this way, by determining the number of targets that each recommendation method needs to process based on the weight of the information recommendation method and the total number of target objects to be recommended, load balancing can be achieved, and recommendation tasks can be evenly distributed to different recommendation methods to avoid the situation where some recommendation methods are overloaded while other recommendation methods are idle. Determining the target number of each recommendation method helps to optimize the efficiency of the recommendation process. The system can allocate tasks according to the processing capacity and efficiency of each recommendation method to ensure that the recommendation task can be completed with high efficiency. By dividing the target objects to be recommended, it is ensured that each recommendation method can make full use of resources and avoid resource waste, which helps to provide the best recommendation service under limited resources. Reasonable diversion and division can improve the accuracy of recommendations. Each recommendation method has the opportunity to provide users with more personalized recommendations in the field it excels in, thereby improving the overall recommendation quality. Users can receive more personalized and accurate recommendation information, thereby improving user experience. Improved user satisfaction helps to increase users' trust and reliance on the recommendation system. According to different information recommendation methods and target numbers, the system can flexibly adjust the recommendation strategy to adapt to different user needs and scenarios.

[0097] In some embodiments, the above-mentioned determination of the target number of target objects to be recommended required to be recommended by each of the information recommendation methods based on the weight of each of the information recommendation methods and the total number of target objects to be recommended can be achieved in the following way: the following processing is performed for each of the information recommendation methods: the weight of the information recommendation method is multiplied by the total number to obtain a candidate number; if the candidate number is an integer, the candidate number is determined as the target number of the information recommendation method; if the candidate number is not an integer, the candidate number is rounded up to obtain the target number of the information recommendation method.

[0098] In some embodiments, if the candidate number is an integer, since the target number of the target objects to be recommended required by the information recommendation method needs to be an integer, the candidate number can be directly determined as the target number of the information recommendation method.

[0099] In some embodiments, if the number of candidates is not an integer, the target number of target objects to be recommended that the information recommendation method needs to perform recommendation needs to be an integer, and at the same time, it is ensured that each target object to be recommended can be assigned to the corresponding information recommendation method. If necessary, the same information recommendation method can be reused for different objects to be recommended. If the number of candidates is rounded down, it cannot be guaranteed that each target object to be recommended can be assigned to the corresponding information recommendation method. In this case, the number of candidates can be rounded up to obtain the target number of the information recommendation method. The sum of the target numbers of each information recommendation method is greater than or equal to the total number of target objects to be recommended. When the sum of the target numbers of each information recommendation method is greater than the total number of target objects to be recommended, it means that different objects to be recommended have reused the same information recommendation method. When the sum of the target numbers of each information recommendation method is equal to the total number of target objects to be recommended, it means that different objects to be recommended do not reuse the same information recommendation method.

[0100] In some embodiments, rounding down, mathematically called a floor function or floor function, is a mathematical operation that means rounding a real number x to the largest integer less than or equal to x. In other words, the floor function returns the largest integer less than or equal to x.

[0101] In some embodiments, rounding up, mathematically called a ceiling function or a ceiling function, is a mathematical operation that means rounding a real number x to the smallest integer greater than or equal to x. In other words, the ceiling function returns the smallest integer greater than or equal to x.

[0102] In this way, by multiplying the weight of the information recommendation method by the total number, it can be ensured that the recommendation method with a higher weight receives more attention. This method can assign different priorities to the recommendation methods based on different business needs and goals. If the candidate number is an integer, it can be directly determined as the target number of the information recommendation method, which can ensure that each recommendation method can get an accurate target number, which helps to improve the accuracy and efficiency of the recommendation. If the candidate number is not an integer, the target number of the information recommendation method is obtained by rounding up, which can ensure that each recommendation method can get at least one target object. This processing method avoids the problem of insufficient target number allocation caused by non-integer results. According to the real-time performance of the information recommendation method and business needs, the target number of each recommendation method can be dynamically adjusted by adjusting the weight, thereby realizing a flexible recommendation strategy. By accurately allocating the target number of information recommendation methods, the resource utilization of the recommendation system can be improved, which can avoid waste of resources and ensure that each recommendation method can be fully utilized.

[0103] In some embodiments, the above-mentioned dividing the target objects to be recommended according to the target numbers corresponding to each of the information recommendation methods to obtain the object sets corresponding to each of the information recommendation methods can be achieved as follows: generating random values ​​for each of the target objects to be recommended; determining the target number corresponding to the first information recommendation method as the first target number, and aggregating the target objects to be recommended whose random values ​​are less than the first target number to obtain the object set corresponding to the first information recommendation method; summing the target numbers corresponding to the first i information recommendation methods to obtain the ith target number, and aggregating the target objects to be recommended whose random values ​​are less than the ith target number and not less than the i-1th target number to obtain the object set corresponding to the ith information recommendation method, where i is a positive integer greater than 1; traversing i to perform the above-mentioned processing until the object set corresponding to the Mth information recommendation method is obtained.

[0104] In some embodiments, the above-mentioned generation of random values ​​for each target object to be recommended can be achieved by the following method: sequentially generate random values ​​for each target object to be recommended, wherein the random values ​​are not less than 1 and not greater than the total number of the target objects to be recommended.

[0105] As an example, the target objects to be recommended include target object to be recommended A, target object to be recommended B, target object to be recommended C, target object to be recommended D, target object to be recommended E, target object to be recommended F and target object to be recommended G, and random values ​​are generated for each of the target objects to be recommended in turn. For example, the random value of target object to be recommended A is 1, the random value of target object to be recommended B is 2, the random value of target object to be recommended C is 3, the random value of target object to be recommended D is 4, the random value of target object to be recommended E is 5, the random value of target object to be recommended F is 6, and the random value of target object to be recommended G is 7.

[0106] Continuing with the above example, the target number corresponding to the first information recommendation method is 2, the target number corresponding to the second information recommendation method is 3, and the target number corresponding to the third information recommendation method is 2. The target objects to be recommended whose random values ​​are less than the target number (2) corresponding to the first information recommendation method are aggregated to obtain the object set corresponding to the first information recommendation method {target object to be recommended A, target object to be recommended B}.

[0107] Continuing with the above example, the target numbers corresponding to the first two information recommendation methods are summed to obtain the second target number (2+3=5), and the target objects to be recommended whose random values ​​are less than the second target number (5) and not less than the first target number (2) are aggregated to obtain the object set corresponding to the second information recommendation method {target object to be recommended C, target object to be recommended D, target object to be recommended E}.

[0108] Continuing with the above example, the target numbers corresponding to the first three information recommendation methods are summed to obtain the third target number (2+3+2=7), and the target objects to be recommended whose random values ​​are less than the third target number (7) and not less than the second target number (5) are aggregated to obtain the object set corresponding to the third information recommendation method (target object to be recommended F, target object to be recommended G).

[0109] In this way, by generating random values ​​to determine the allocation of target objects to be recommended, it is ensured that the chances of each object being recommended in the recommendation process are random and balanced, reducing manual intervention and unfairness. The random allocation method can flexibly respond to different recommendation needs and scenarios, and at the same time has a certain degree of scalability and can adapt to changes in the number of recommended objects. By aggregating target objects to be recommended whose random values ​​are less than the corresponding target number, it is possible to achieve optimal allocation of resources and ensure that each recommendation method can obtain a set of objects that matches its weight. By generating random values ​​and aggregating them according to the target number of information recommendation methods, it is possible to achieve optimal allocation of resources, improve the accuracy of recommendations and user experience, and support dynamic adjustment and stability.

[0110] In other embodiments, the above-mentioned division of the target objects to be recommended to obtain the object sets corresponding to each information recommendation method can be achieved as follows: the following processing is performed for each information recommendation method: the carrying quantity of the target objects to be recommended that can be recommended by the information recommendation method is obtained; if the carrying quantity is greater than or equal to the total quantity, the target objects to be recommended are aggregated to obtain the object set corresponding to the information recommendation method; if the carrying quantity is less than the total quantity, the target objects to be recommended with the carrying quantity are aggregated to obtain the object set corresponding to the information recommendation method.

[0111] In some embodiments, the carrying capacity of the target objects to be recommended that can be recommended by the above-mentioned information recommendation method is used to indicate the load of the information recommendation method on the target objects to be recommended. The carrying capacity of the target objects to be recommended that can be recommended by the information recommendation method usually refers to the carrying capacity of the target objects to be recommended that can be processed and recommended by each information recommendation method within a certain recommendation period. The processing capacity and storage resources of the information recommendation method determine the maximum number of objects that can be recommended.

[0112] In some embodiments, if the carrying quantity is greater than or equal to the total quantity, it means that the information recommendation method is capable of realizing information recommendation for all target objects to be recommended. Then, the target objects to be recommended can be aggregated to obtain an object set corresponding to the information recommendation method. At this time, the object set corresponding to the information recommendation method includes all target objects to be recommended, thereby realizing information recommendation for all target objects to be recommended through the information recommendation method.

[0113] In some embodiments, if the carrying quantity is less than the total quantity, it means that the information recommendation method cannot achieve information recommendation for all target objects to be recommended, and can only achieve information recommendation for the carrying quantity of target objects to be recommended. Then, the carrying quantity of target objects to be recommended can be aggregated to obtain an object set corresponding to the information recommendation method.

[0114] In this way, obtaining the number of target objects to be recommended by the information recommendation method can ensure the optimal configuration under limited resources and avoid resource waste. When the number of targets is greater than or equal to the total number, all target objects to be recommended can be directly aggregated into an object set, which can improve the processing efficiency of the recommendation and reduce repeated calculations and resource consumption. When the number of targets is less than the total number, by aggregating the number of recommended objects, the load of different information recommendation methods can be balanced to ensure that each recommendation method can efficiently handle the recommendation tasks within its scope.

[0115] In some embodiments, the above-mentioned determination of at least one of the information recommendation methods as the target information recommendation method corresponding to the target object to be recommended can be achieved in the following way: obtaining the feedback degree of the target object to be recommended for each of the information recommendation methods; and determining the information recommendation method whose feedback degree is greater than a feedback degree threshold as the target information recommendation method corresponding to the target object to be recommended.

[0116] In some embodiments, a feedback degree greater than a feedback degree threshold is a positive feedback degree, and a feedback degree less than or equal to the feedback degree threshold is a negative feedback degree. The feedback degree is used to indicate the feedback degree of the target object to be recommended when information is recommended to the target object to be recommended according to the information recommendation method. The greater the feedback degree, the better the recommendation effect of the information recommendation method for the target object to be recommended, and the smaller the feedback degree, the worse the recommendation effect of the information recommendation method for the target object to be recommended.

[0117] In other embodiments, the above-mentioned determination of at least one of the information recommendation methods as the target information recommendation method corresponding to the target object to be recommended can be achieved by the following method: determining M types of information recommendation methods as the target information recommendation methods corresponding to the target object to be recommended.

[0118] In this way, by screening out recommendations with high feedback, we can better adapt to users' actual preferences, thereby improving the relevance of recommended content. When users find that the information provided by the recommendation system meets their interests and needs, they are more likely to continue using the service, which helps increase user stickiness to the platform. Effective recommendation methods can help reduce resource waste, such as avoiding recommending content that users are not interested in, thereby focusing resources on effective recommendations. For e-commerce platforms, highly relevant recommendations can increase user conversion rates.

[0119] In step 103, information is recommended to the target object to be recommended according to the target information recommendation method.

[0120] In some embodiments, the target information recommendation method corresponding to the above-mentioned target object to be recommended is at least one, and the target information recommendation methods corresponding to different targets to be recommended may be the same or different.

[0121] As an example, the above-mentioned target information recommendation method can be APP recommendation, SMS recommendation and email recommendation, and the target objects to be recommended include target object to be recommended A, target object to be recommended B, target object to be recommended C and target object to be recommended D; for target object to be recommended A, information recommendation is made to target object to be recommended A according to the APP recommendation and SMS recommendation corresponding to target object to be recommended A; for target object to be recommended B, information recommendation is made to target object to be recommended B according to the APP recommendation and email recommendation corresponding to target object to be recommended B; for target object to be recommended C, information recommendation is made to target object to be recommended according to the SMS recommendation and email recommendation corresponding to target object to be recommended C; for target object D, information recommendation is made to target object D according to the SMS recommendation corresponding to target object D.

[0122] In some embodiments, the above step 103 can be implemented as follows: if the total number of the target objects to be recommended is one, the information to be recommended is sent to the target objects to be recommended respectively according to each target information recommendation method; if the total number of the target objects to be recommended is multiple, for each target object to be recommended, the information to be recommended is sent to the target object to be recommended according to the target information recommendation method corresponding to the target object to be recommended.

[0123] In some embodiments, if the total number of the target objects to be recommended is one, the target information recommendation methods are all target information recommendation methods corresponding to the target objects to be recommended, and the recommended information can be sent to the target objects to be recommended respectively according to each target information recommendation method.

[0124] As an example, the total number of target objects to be recommended is one, for example, target object to be recommended A, and the target information recommendation method corresponding to target object to be recommended A can be SMS recommendation and APP recommendation. Then, through SMS recommendation, a SMS including the information to be recommended can be sent to the target object to be recommended A, and through APP recommendation, an APP message including the information to be recommended can be sent to the APP account corresponding to the target object to be recommended A.

[0125] In some embodiments, if the total number of the target objects to be recommended is multiple, it means that the target information recommendation method may correspond to different target objects to be recommended, and the target information recommendation methods corresponding to different target objects to be recommended may be the same or different. Then, for each target object to be recommended, the information to be recommended is sent to the target object to be recommended according to the target information recommendation method corresponding to the target object to be recommended.

[0126] Continuing with the above example, for the target object to be recommended A, according to the APP recommendation and SMS recommendation corresponding to the target object to be recommended A, an APP message carrying the recommendation information is sent to the target object to be recommended A, and a SMS carrying the recommendation information is sent to the target object to be recommended A; for the target object to be recommended B, according to the APP recommendation and email recommendation corresponding to the target object to be recommended B, an APP message carrying the recommendation information is sent to the APP account of the target object to be recommended B, and an email carrying the recommendation information is sent to the email address of the target object to be recommended.

[0127] In this way, by screening N objects to be recommended, the target object to be recommended that meets the recommendation conditions is obtained, and the target information recommendation method corresponding to the target object to be recommended is determined from the M information recommendation methods, and information recommendation is performed on the target object to be recommended according to the target information recommendation method. In this way, by determining the target information recommendation method corresponding to the target object to be recommended from the M information recommendation methods, diversified recommendations are performed through M different information recommendation methods, and the information recommendation strength is effectively improved. By determining the target information recommendation method corresponding to the target object to be recommended from the M information recommendation methods, a customized target information recommendation method is provided for the target object to be recommended. By screening N objects to be recommended, the target object to be recommended that meets the recommendation conditions is obtained, and the interference of redundant objects to be recommended on the information recommendation strength is effectively reduced, thereby effectively improving the information recommendation effect.

[0128] Next, an exemplary application of the embodiment of the present application in an actual application scenario of advertisement recommendation will be described.

[0129] Ad recommendation is a personalized recommendation method based on user behavior, interests and preferences, which aims to provide users with relevant and attractive advertising content. By analyzing users' historical data, browsing history, search behavior, purchase preferences and other information, the ad recommendation system can recommend ads that users may be interested in, thereby increasing the click-through rate and conversion rate of ads, while optimizing the advertiser's effectiveness and return on investment.

[0130] In some embodiments, see Figure 4 , Figure 4It is a schematic diagram of the principle of the componentized marketing system provided by the embodiment of the present application. With the rapid development of information technology, marketing methods are also constantly updated and iterated. Traditional marketing methods have been difficult to meet the real-time, accurate and efficient marketing needs of modern enterprises. Therefore, developing a marketing system that can be flexibly configured, responded in real time and iterated quickly has become an urgent need for the development of the industry. The componentized marketing system provided by the embodiment of the present application includes a configuration module, a component module and a component communication (asynchronous data flow mode when the component is running): kafka / mq, and the configuration module includes component configuration: in the canvas engine, by selecting component module components, dragging and dragging, connecting and other methods, marketing frequency and cycle, crowd package: configure crowd package labels, and salvage marketing users. The component module includes a data processing component, which is used to obtain the labels configured by the crowd package configured by the planned configuration, and salvage marketing users. The decision-making component includes a decision tree component and a decision table component. The proportional diversion component includes a random diversion component, a proportional diversion component and a user tail number diversion component, a label processing component and a basic marketing component, wherein the basic marketing component includes a text message contact component, an APP contact component and a telephone marketing component.

[0131] In some embodiments, see Figure 4 , Configuration module: responsible for the visualization configuration storage of marketing strategies, using various components by dragging and dropping lines on the page, and storing the configuration results. Component module: various components in the component module, and data interaction is carried out through the component communication module when the component is running. Contains 5 major types of components: data processing components, decision-making components, proportional diversion components, label processing components, and basic marketing channel components. Data processing components: mainly based on the generation of primary labels of various dimensions for users, and aggregating and classifying the labels to form marketing population packages. Decision-making components: optional components, segmenting marketing groups according to marketing strategies. Proportional diversion components: optional components, splitting the population package by proportion, and conducting group test marketing. Label processing type components: output processed labels after the population package is split for marketing. Basic marketing channel components: marketing contact methods, including SMS, APP, telephone sales, etc.

[0132] In some embodiments, see Figure 5 , Figure 5It is a schematic diagram of the principle of the component life cycle provided by the embodiment of the present application. The component life cycle includes component startup, component operation and component destruction. The component operation includes the operating environment (component operation parameters and component execution relationship), input and output parameters (input parameter conversion, standard parameters and output parameter conversion), component business logic includes business code and context, component exception handling mechanism includes recording / retry, delay mechanism, component attributes include public attributes: deduplication / filtering, private attributes: for example, the proportion of proportional diversion, decision table of decision component, etc. The component life cycle includes: component startup, component operation, component destruction. Component startup, when the component is created, the necessary configuration and initialization work is performed, such as setting the properties of the component, dependency injection, and the system for allocating component operation. Component operation, when the component receives data or triggers an event, the corresponding business logic is executed. At this stage, the component can perform operations such as data processing, user diversion and marketing strategy execution according to the marketing plan. Component destruction, when the component is no longer needed, clean up and destroy it to release related resources.

[0133] In some embodiments, see Figure 6 , Figure 6 It is a schematic diagram of the principle of the operation of a single component provided by the embodiment of the present application, the component operation mechanism, including the operation environment, input and output parameters, the business logic of each component, the component operation exception handling mechanism, the delay mechanism, and the properties of the component. Through the management and design of the component life cycle and mechanism, more abundant components can be flexibly expanded to reduce the coupling of the system. Component operation environment, maintain component execution order and operation parameters, reduce system complexity, and realize the rapid iteration of new components and the offline of unnecessary components. Input and output parameter management: configure and define component data input and output parameters to realize the flexible operation of different components and perform more flexible component configuration. Component business logic: reserve the customized logic for extended components. Different components realize different business functions. You can customize the business logic required for development and extension, and quickly implement new components. Component exception retry mechanism: define and maintain component exception handling strategies to improve the stability of component operation. Delay mechanism: define delayed marketing, unified logic of delayed reach, and enrich marketing strategies. Component properties, define and extend the properties of different components.

[0134] In some embodiments, see Figure 6 , data enters, loads parameter conversion, inputs parameter conversion, and parameter conversion; executes component logic and reads component properties; outputs data and enters the next component for processing, and monitors the output data for abnormalities to improve stability.

[0135] In some embodiments, the operating mechanism of a component determines how the component works with other components and how it processes data and events. In view of the needs of the real-time marketing system, the following operating mechanisms are designed:

[0136] Component communication mechanism: driven by event mechanism, through message queues such as kafka / mq, the message flow between components is realized, data stream forwarding can be carried out quickly, and the performance advantages of the system are brought into play. Dynamic configuration mechanism: allows dynamic adjustment of component configuration at runtime. In this way, the embodiment of the present application can quickly adjust the marketing strategy and component configuration according to real-time market changes and user feedback, and improve the flexibility and response speed of the system. By carefully designing the life cycle and operation mechanism of the components, the embodiment of the present application can build an efficient, flexible and maintainable real-time marketing system. Such a system can quickly adapt to market changes and user needs, realize accurate marketing strategy formulation and execution, and improve marketing effects and user experience.

[0137] In some embodiments, the componentized marketing system provided by the embodiments of the present application has the advantages of flexible configuration, real-time response, and precision marketing, which can meet the real-time, precise, and efficient marketing needs of modern enterprises, and improve marketing effects and user satisfaction. Configuration module: Configure marketing plans through a visual canvas to achieve personalized marketing strategies, reduce marketing configuration costs, reduce the threshold for operators to use, and improve operational efficiency. Data processing components: Process crowd packages based on marketing user tags, and salvage user lists for crowd packages to locate target users. Decision-making components: Data flow in different directions can be carried out according to strategies, such as decision table components and decision tree components, and user tags / attributes of user marketing can be further subdivided for more precise marketing target user selection. Proportional diversion components: Traffic can be distributed according to a certain proportion, such as percentage diversion, random diversion, user tail number diversion, etc. After diversion through proportional diversion components, fixed marketing strategies can be formulated for marketing strategy result analysis, comparison of different marketing methods, and further optimization of marketing plans and marketing strategies. Label processing components: can be used for user attribute / label processing / update, can process complex labels, or update labels, and feed back to data receiving components for further marketing of processed and updated user labels. Basic marketing channel components: can be used for marketing reach, such as SMS marketing, APP push marketing, smart telemarketing and other specific marketing methods.

[0138] In some embodiments, see Figure 7 , Figure 7 This is a schematic diagram of the process of the information recommendation method provided in the embodiment of the present application. Figure 2 The information recommendation method provided in the embodiment of the present application can be performed by Figure 7 Steps 201 to 207 are shown to be implemented.

[0139] In step 201, pull the marketing population: pull the customers who registered in the past 30 days.

[0140] In some embodiments, the above step 201 can be implemented by configuring a data processing component. The data processing component retrieves newly registered users in the past 30 days through the configured population package tags, SQL, etc., and transmits them to the decision component through the Kafka message queue.

[0141] In step 202, customer segmentation is performed using a decision table.

[0142] In some embodiments, after receiving user information, the decision-making component divides the customers with and without credit limits through decision stratification, wherein the customers with credit limits are asynchronously transmitted to the basic marketing SMS channel component through Kafka. The SMS marketing component sends the marketing content to the user through the SMS channel. If the user completes the credit limit through the marketing SMS, the label processing component processes the user into a user with a credit limit label.

[0143] In some embodiments, customer segmentation is performed by segmenting customers with credit limits and customers without credit limits. For customers without credit limits, step 203 is performed to improve credit limit evaluation by SMS marketing for customers without credit limits through the basic marketing channel: SMS. For customers with credit limits, step 205 is performed.

[0144] In step 203, basic marketing channel: SMS.

[0145] In some embodiments, by configuring basic marketing components, SMS marketing for customers without credit limits improves credit limit assessment.

[0146] In step 204, the label is processed.

[0147] In some embodiments, by configuring label processing components, labels are processed according to the customer's complete credit limit information after SMS marketing, and the customer's credit limit is updated.

[0148] In some embodiments, the SMS marketing component sends the marketing content to the user through the SMS channel. If the user completes the credit limit through the marketing SMS, the label processing component processes the user into a user with a credit limit label.

[0149] In step 205, the flow is divided proportionally.

[0150] In some embodiments, the decision-making component transmits the customers with credit limits to the proportional diversion component through Kafka asynchronously, configures the diversion component, and performs proportional diversion on the customers with credit limits. Proportional diversion and comparison marketing are performed on customers with credit limits.

[0151] In step 206, after receiving the marketing customer information, the proportional distribution component transmits it to the APP marketing component and the SMS marketing component asynchronously through Kafka at a ratio of 1:9.

[0152] In some embodiments, the APP marketing component sends marketing content to customers through the AP terminal.

[0153] In step 207, SMS marketing is performed.

[0154] In some embodiments, the SMS marketing component sends marketing content to users via an SMS channel.

[0155] In some embodiments, the configuration and release of marketing strategies, after each marketing strategy is configured, the configuration module will generate a corresponding strategy configuration data structure. This process is called component configuration, and different parameters can be selected when configuring the components. After the components are configured, the marketing data flow needs to be determined by connecting lines. A strategy configuration data structure includes nodes (Node) and node lines (Line). A complete marketing strategy configuration data structure includes multiple Nodes and Lines. Nodes correspond to each component in componentization, and Lines are used to connect Nodes to form a marketing strategy channel. Lines can be configured with conditions. When the conditions are met, data flows through the line. A complete marketing strategy has a start node and an end node. The data salvage component serves as the start node, and the end node is generally a label processing node or a basic marketing channel node. Each Node and Line has a globally unique ID. When the marketing strategy configuration is completed, it takes effect immediately after being released.

[0156] In some embodiments, the processing and data flow of marketing strategy data, the data processing component is executed through a scheduled task, and the data processing is performed according to the data processing time configured for different marketing strategies. Other components listen to the specified message queue when the component is started. Each component listens to a different message queue. After the data processing component completes the data processing, it patrols the line according to the Line configured for different marketing strategies. Patrolling the line is to find the next or multiple Nodes based on the connection between the configured Nodes. When the conditions of the Line are met, the data will be sent to the message queue that the component of the corresponding Node listens to. The selection of components is achieved in this way.

[0157] In some embodiments, the data processing component: can be configured to select different population packages and different execution times. During implementation, the data selection SQL is dynamically generated according to the configured population package conditions. After the configuration is completed, a Node is generated. The component type of the Node is a data processing component, and the data processing component is responsible for execution. After startup, the data processing component will periodically scan all published marketing strategies, check the Node of the data processing component type, and perform data selection and data processing according to the processing time configured for the Node. When the data processing is completed, the line is patrolled according to the unique ID of the Node, all Lines under the Node are found, and the marketing user data is checked to see if it meets the conditions of the Line. If so, the ID of the next one or more Nodes is found through the Line. The flow of data to the next component is achieved through the inter-component communication mechanism, and the data is sent to the component of the downstream Node of the Line through the message queue.

[0158] In some embodiments, the decision component: can configure different decision tables / trees, such as selecting user attributes such as user age, gender, credit limit, etc. to make decisions. After the decision component configures the decision table, it generates the Node of the decision component, and the decision component is responsible for execution. When the decision component is started, it will listen to the specified message queue A. For example, after receiving the list sent by the data processing component, it obtains the decision configuration information according to the ID of the Node specified by the sender, and makes decisions according to the configured decision rules. The column header of the decision table is the user attribute, and the column content is the attribute condition. Multiple columns can be configured to configure multiple attributes, and multiple attribute conditions can be configured at the same time. When the user's attributes meet all the conditions of the column during the decision, the decision is successful and the decision result is output. After the decision component completes the decision according to the configured list, it passes the decision result to the next component, and patrols the next node in method 1 (b). For example, "Consumer Finance Registration New User Marketing" is taken as an example. If the result of the decision is that there is a credit limit, it satisfies the condition of the proportional diversion Line and flows to the corresponding Node. If there is no credit limit, it satisfies the condition of the basic marketing channel SMS Line and flows to the corresponding Node. The data flow to the next component is realized through the inter-component communication mechanism, and the data is sent to the component of the downstream Node of the Line through the message queue.

[0159] In some embodiments, the proportional diversion component: can be configured with different proportions and diversion methods, and can configure percentage diversion, and can output the results in a random percentage. When the proportional diversion is started, it will listen to the specified message queue B. For example, after receiving the list sent by the decision-maker, it will obtain the proportional diversion configuration information according to the ID of the Node specified by the sender, and distribute the proportion according to the configured proportion. When executing, the proportional diversion component uses a random value method to distribute according to the configured proportion. For example, taking "Consumer Finance Registration of New User Marketing" as an example, the distribution is distributed in a random value method, with a random value of 1 to 100. If the random value is less than or equal to 10, it satisfies the conditions of the APP marketing Line and flows to the corresponding Node. If the random value is greater than 10, it satisfies the conditions of the SMS marketing Line and flows to the corresponding Node. The data flow to the next component is realized through the inter-component communication mechanism, and the data is sent to the component of the downstream Node of the Line through the message queue.

[0160] In some embodiments, the label processing component can be configured to output different user labels and label values, and the label processing component will process data according to the configured label information. When the data label component is started, it will listen to the specified message queue C. For example, after receiving the data, it will obtain the label configuration information according to the NodeId specified by the sender, process the data label, generate SQL for inserting / updating user attributes based on the label and label value, and update the user label attributes.

[0161] In some embodiments, the basic marketing SMS channel component can be configured to send different SMS contents and call the SMS operator channel for SMS reach. When the basic marketing SMS channel component is started, it will listen to the specified message queue D. For example, after receiving the data, it will obtain the marketing SMS content according to the NodeId specified by the sender, and dynamically replace the user parameters of the SMS content before sending the marketing SMS to the user. After the SMS is sent, the SMS sending result can be obtained.

[0162] In some embodiments, after the SMS is executed, the line is patrolled according to the unique ID of the Node, all Lines under the Node are found, and the marketing user data is checked to see if it meets the conditions of the Line. If so, the ID of the next or multiple Nodes is found through the Line. The data flow to the next component is realized through the inter-component communication mechanism, and the data is sent to the components of the downstream Node of the Line through the message queue.

[0163] In some embodiments, the basic marketing channel APP component can be configured to send different push content and push marketing content through the APP channel (same as the basic marketing SMS channel component).

[0164] In this way, by screening N objects to be recommended, the target object to be recommended that meets the recommendation conditions is obtained, and the target information recommendation method corresponding to the target object to be recommended is determined from the M information recommendation methods, and information recommendation is performed on the target object to be recommended according to the target information recommendation method. In this way, by determining the target information recommendation method corresponding to the target object to be recommended from the M information recommendation methods, diversified recommendations are performed through M different information recommendation methods, and the information recommendation strength is effectively improved. By determining the target information recommendation method corresponding to the target object to be recommended from the M information recommendation methods, a customized target information recommendation method is provided for the target object to be recommended. By screening N objects to be recommended, the target object to be recommended that meets the recommendation conditions is obtained, and the interference of redundant objects to be recommended on the information recommendation strength is effectively reduced, thereby effectively improving the information recommendation effect.

[0165] It is understandable that in the embodiments of the present application, when data related to the objects to be recommended is involved, when the embodiments of the present application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0166] The following is a description of an exemplary structure of the information recommendation device 455 provided in the embodiment of the present application implemented as a software module. In some embodiments, Figure 2 As shown, the software modules stored in the information recommendation device 455 of the memory 450 may include: a screening module, used to screen N objects to be recommended to obtain a target object to be recommended that meets the recommendation conditions, where N is a positive integer greater than 1; a determination module, used to determine a target information recommendation method corresponding to the target object to be recommended from M information recommendation methods, where M is a positive integer greater than 1; and a recommendation module, used to recommend information to the target object to be recommended according to the target information recommendation method.

[0167] In some embodiments, the above-mentioned screening module is also used to compare the object information of the object to be recommended with the recommendation condition for each object to be recommended, so as to obtain a condition comparison result of the object to be recommended; if the condition comparison result indicates that the object information of the object to be recommended meets the recommendation condition, the object to be recommended is determined as the target object to be recommended.

[0168] In some embodiments, the above-mentioned screening module is also used to determine the object to be recommended as a reference object to be recommended if the condition comparison result indicates that the object information of the object to be recommended does not meet the recommendation condition; and send a prompt message to the reference object to be recommended, wherein the prompt message is used to prompt the reference object to be recommended to update the object information of the reference object to be recommended to obtain updated object information that meets the recommendation condition.

[0169] In some embodiments, the above-mentioned determination module is also used to determine the M types of information recommendation methods as the target information recommendation methods corresponding to the target objects to be recommended if the total number of the target objects to be recommended is one; if the total number of the target objects to be recommended is multiple, then based on the number of the information recommendation methods and the total number of the target objects to be recommended, determine the target information recommendation method corresponding to each of the target objects to be recommended.

[0170] In some embodiments, the above-mentioned determination module is also used to compare the number of the information recommendation methods and the total number of the target objects to be recommended to obtain a comparison result; if the comparison result indicates that the number of the information recommendation methods is greater than or equal to the total number of the target objects to be recommended, then for each target object to be recommended, at least one of the information recommendation methods is determined as the target information recommendation method corresponding to the target object to be recommended; if the comparison result indicates that the number of the information recommendation methods is less than the total number of the target objects to be recommended, the target objects to be recommended are divided to obtain object sets corresponding to each of the information recommendation methods, and for each object set, the information recommendation method corresponding to the object set is determined as the target information recommendation method corresponding to the target object to be recommended in the object set.

[0171] In some embodiments, the above-mentioned determination module is also used to determine the target number of target objects to be recommended required to be recommended by each of the information recommendation methods based on the weight of each of the information recommendation methods and the total number of target objects to be recommended; and divide the target objects to be recommended according to the target number corresponding to each of the information recommendation methods to obtain the object sets corresponding to each of the information recommendation methods.

[0172] In some embodiments, the above-mentioned determination module is also used to multiply the weight of the information recommendation method by the total number for each of the information recommendation methods to obtain the number of candidates; if the number of candidates is an integer, the number of candidates is determined as the target number of the information recommendation method; if the number of candidates is not an integer, the number of candidates is rounded up to obtain the target number of the information recommendation method.

[0173] In some embodiments, the above-mentioned determination module is also used to generate a random value for each of the target objects to be recommended; the target number corresponding to the first information recommendation method is determined as the first target number, and the target objects to be recommended whose random values ​​are less than the first target number are aggregated to obtain an object set corresponding to the first information recommendation method; the target numbers corresponding to the first i information recommendation methods are summed to obtain the ith target number, and the target objects to be recommended whose random values ​​are less than the ith target number and not less than the i-1th target number are aggregated to obtain an object set corresponding to the ith information recommendation method, where i is a positive integer greater than 1; traverse i to perform the above-mentioned processing until the object set corresponding to the Mth information recommendation method is obtained.

[0174] In some embodiments, the above-mentioned determination module is also used to perform the following processing for each of the information recommendation methods: obtaining the carrying quantity of the target objects to be recommended that can be recommended by the information recommendation method; if the carrying quantity is greater than or equal to the total quantity, then aggregating the target objects to be recommended to obtain an object set corresponding to the information recommendation method; if the carrying quantity is less than the total quantity, then aggregating the target objects to be recommended with the carrying quantity to obtain an object set corresponding to the information recommendation method.

[0175] In some embodiments, the above-mentioned determination module is also used to obtain the feedback degree of the target object to be recommended for each of the information recommendation methods, and the feedback degree is used to indicate the feedback level of the target object to be recommended when information is recommended to the target object to be recommended according to the information recommendation method; the information recommendation method with the feedback degree greater than the feedback degree threshold is determined as the target information recommendation method corresponding to the target object to be recommended.

[0176] In some embodiments, the above-mentioned recommendation module is also used to, if the total number of the target objects to be recommended is one, send the information to be recommended to the target objects to be recommended respectively according to each target information recommendation method; if the total number of the target objects to be recommended is multiple, then for each target object to be recommended, send the information to be recommended to the target object according to the target information recommendation method corresponding to the target object to be recommended.

[0177] The embodiment of the present application provides a computer program product, which includes a computer program or a computer executable instruction, and the computer program or the computer executable instruction is stored in a computer-readable storage medium. The processor of the electronic device reads the computer executable instruction from the computer-readable storage medium, and the processor executes the computer executable instruction, so that the electronic device performs the information recommendation method described in the embodiment of the present application.

[0178] The present application embodiment provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the processor will execute the information recommendation method provided by the present application embodiment, for example, Figure 3 The information recommendation method is shown.

[0179] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEP ROM, flash memory, magnetic surface storage, optical disk, or CD-ROM; or it may be various electronic devices including one or any combination of the above memories.

[0180] In some embodiments, computer executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0181] As an example, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file storing other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0182] As an example, computer executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed at multiple sites and interconnected by a communication network.

[0183] In summary, the embodiments of the present application have the following beneficial effects:

[0184] (1) By screening N objects to be recommended, a target object to be recommended that meets the recommendation conditions is obtained, and a target information recommendation method corresponding to the target object to be recommended is determined from M information recommendation methods, and information recommendation is performed on the target object to be recommended according to the target information recommendation method. In this way, by determining the target information recommendation method corresponding to the target object to be recommended from M information recommendation methods, diversified recommendations are performed through M different information recommendation methods, and the information recommendation strength is effectively improved. By determining the target information recommendation method corresponding to the target object to be recommended from M information recommendation methods, a customized target information recommendation method is provided for the target object to be recommended. By screening N objects to be recommended, a target object to be recommended that meets the recommendation conditions is obtained, and the interference of redundant objects to be recommended on the information recommendation strength is effectively reduced, thereby effectively improving the information recommendation effect.

[0185] (2) If the total number of target objects to be recommended is one, then each information recommendation method can be determined as the target information recommendation method corresponding to the target object to be recommended, so that the target object to be recommended can be repeatedly recommended through M information recommendation methods, thereby effectively improving the recommendation effect.

[0186] (3) By comparing the number of information recommendation methods with the total number of target objects to be recommended, the complexity of the recommendation process can be effectively controlled. If the number of information recommendation methods is greater than or equal to the total number of target objects to be recommended, it means that each target object can correspond to at least one information recommendation method, which can ensure that each object can receive appropriate recommendations. If the number of information recommendation methods is less than the total number of target objects to be recommended, the system will divide the target objects to be recommended to ensure that each information recommendation method has a corresponding set of objects. This can avoid too many objects waiting without sufficient recommendation methods, thereby optimizing resource allocation. Ensuring that each target object to be recommended has a corresponding target information recommendation method can improve the accuracy and coverage of recommendations. Each object has the opportunity to be discovered and selected by the user through at least one recommendation method. When the number of information recommendation methods is not enough to cover all target objects to be recommended, by dividing the objects, it is ensured that each recommendation method can be fully utilized, which helps to provide effective recommendations under limited resources.

[0187] (4) A feedback mechanism is established by sending prompt information to the recommended objects that do not meet the recommendation conditions. This gives the objects the opportunity to update their information according to the conditions of the recommendation system, thereby increasing the possibility of being selected in the future. After receiving the prompt information, the reference recommended objects can dynamically adjust their object information based on this information. This dynamic optimization helps the objects better adapt to the screening criteria of the recommendation system and improve their competitiveness in the recommendation list. The user participation link is increased. Users are no longer passively accepting recommendations, but can actively adjust their information to meet the recommendation conditions of the system, thereby obtaining more satisfactory recommendation results. Through the prompt information, the reference recommended objects can more accurately understand the requirements of the recommendation system, thereby providing object information that better meets the requirements and improving the overall accuracy and efficiency of the recommendation system. The prompt information can help the reference recommended objects better adapt to system changes and demand changes, making the recommendation system more adaptable and flexible.

[0188] (5) By comparing the information of the recommended object with the recommendation conditions, the target object that meets the user's needs or scenario requirements can be accurately screened out, thereby improving the accuracy and relevance of the recommendation. When the information or products required by the user can be accurately provided, the user's satisfaction and experience can be significantly improved, thereby enhancing the user's loyalty to the service. The recommendation conditions can be efficiently processed and matched, resource utilization can be optimized, and computing costs can be reduced. For shopping, recruitment, matching and other services, providing qualified target objects can help make decisions faster and improve the conversion rate of transactions. The recommendation conditions can be quickly adjusted according to market changes and changes in user needs, so that the recommendation results are always timely and adaptable.

[0189] (6) By multiplying the weight of the information recommendation method by the total number, it can be ensured that the recommendation method with a higher weight receives more attention. This method can assign different priorities to the recommendation methods based on different business needs and goals. If the candidate number is an integer, it can be directly determined as the target number of the information recommendation method, which can ensure that each recommendation method can get an accurate target number, which helps to improve the accuracy and efficiency of the recommendation. If the candidate number is not an integer, the target number of the information recommendation method is obtained by rounding up, which can ensure that each recommendation method can get at least one target object. This processing method avoids the problem of insufficient target number allocation due to non-integer results. According to the real-time performance of the information recommendation method and business needs, the target number of each recommendation method can be dynamically adjusted by adjusting the weight, thereby realizing a flexible recommendation strategy. By accurately allocating the target number of information recommendation methods, resource utilization can be effectively improved, which can avoid resource waste and ensure that each recommendation method can be fully utilized.

[0190] (7) By generating random values ​​to determine the allocation of target objects to be recommended, it is ensured that the chances of each object being recommended during the recommendation process are random and balanced, reducing manual intervention and unfairness. The random allocation method can flexibly respond to different recommendation needs and scenarios, and at the same time has a certain degree of scalability and can adapt to changes in the number of recommended objects. By aggregating target objects to be recommended whose random values ​​are less than the corresponding target number, it is possible to achieve optimal allocation of resources and ensure that each recommendation method can obtain a set of objects that matches its weight. By generating random values ​​and aggregating them according to the target number of information recommendation methods, it is possible to achieve optimal allocation of resources, improve the accuracy of recommendations and user experience, and support dynamic adjustment and stability.

[0191] (8) By screening out recommendations with high feedback, it is possible to better adapt to users' actual preferences, thereby improving the relevance of recommended content. When users find that the information provided by the recommendation system meets their interests and needs, they are more likely to continue using the service, which helps increase user stickiness to the platform. Effective recommendation methods can help reduce resource waste, such as avoiding recommending content that users are not interested in, thereby focusing resources on effective recommendations. For e-commerce platforms, highly relevant recommendations can increase user conversion rates.

[0192] (9) Obtaining the number of target objects to be recommended by the information recommendation method can ensure the optimal configuration under limited resources and avoid resource waste. When the number of targets is greater than or equal to the total number, all target objects to be recommended can be directly aggregated into an object set, which can improve the processing efficiency of the recommendation and reduce repeated calculations and resource consumption. When the number of targets is less than the total number, by aggregating the number of recommended objects, the load of different information recommendation methods can be balanced to ensure that each recommendation method can efficiently process the recommendation tasks within its scope.

[0193] The above is only an embodiment of the present application and is not intended to limit the protection scope of the present application. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the present application are included in the protection scope of the present application.

Claims

1. An information recommendation method, characterized in that: The method comprises: Screening N objects to be recommended to obtain target objects to be recommended that meet the recommendation conditions, where N is a positive integer greater than 1; Determine a target information recommendation method corresponding to the target object to be recommended from M information recommendation methods, where M is a positive integer greater than 1; According to the target information recommendation method, information is recommended to the target object to be recommended.

2. The method according to claim 1, characterized in that The screening of the N objects to be recommended to obtain the target objects to be recommended that meet the recommendation conditions includes: For each of the objects to be recommended, comparing the object information of the object to be recommended with the recommendation condition to obtain a condition comparison result of the object to be recommended; If the condition comparison result indicates that the object information of the object to be recommended meets the recommendation condition, the object to be recommended is determined as the target object to be recommended.

3. The method according to claim 2, characterized in that The method further comprises: If the condition comparison result indicates that the object information of the object to be recommended does not meet the recommendation condition, determining the object to be recommended as a reference object to be recommended; Prompt information is sent to the reference object to be recommended, where the prompt information is used to prompt the reference object to be recommended to update the object information of the reference object to be recommended to obtain updated object information that meets the recommendation condition.

4. The method according to claim 1, characterized in that: Determining the target information recommendation method corresponding to the target object to be recommended from the M information recommendation methods includes: If the total number of the target objects to be recommended is one, the M information recommendation methods are determined as the target information recommendation methods corresponding to the target objects to be recommended; If the total number of the target objects to be recommended is multiple, the target information recommendation method corresponding to each target object to be recommended is determined based on the number of the information recommendation methods and the total number of the target objects to be recommended.

5. The method according to claim 4, characterized in that The determining, based on the number of the information recommendation methods and the total number of the target objects to be recommended, a target information recommendation method corresponding to each target object to be recommended, includes: Compare the number of the information recommendation methods with the total number of the target objects to be recommended to obtain a comparison result; If the comparison result indicates that the number of the information recommendation methods is greater than or equal to the total number of the target objects to be recommended, then for each target object to be recommended, determining at least one of the information recommendation methods as the target information recommendation method corresponding to the target object to be recommended; If the comparison result indicates that the number of the information recommendation methods is less than the total number of the target objects to be recommended, the target objects to be recommended are divided to obtain object sets corresponding to each of the information recommendation methods; For each of the object sets, the information recommendation mode corresponding to the object set is determined as the target information recommendation mode corresponding to the target object to be recommended in the object set.

6. The method according to claim 5, characterized in that The step of dividing the target objects to be recommended to obtain sets of objects corresponding to the information recommendation methods includes: Determining the target number of the target objects to be recommended that need to be recommended by each of the information recommendation methods based on the weight of each of the information recommendation methods and the total number of the target objects to be recommended; According to the target quantity corresponding to each of the information recommendation methods, the target objects to be recommended are divided to obtain the object sets corresponding to each of the information recommendation methods.

7. The method according to claim 6, characterized in that The determining, based on the weight of each of the information recommendation modes and the total number of the target objects to be recommended, the target number of the target objects to be recommended required to be recommended by each of the information recommendation modes includes: For each information recommendation method, multiply the weight of the information recommendation method by the total number to obtain the candidate number; If the candidate number is an integer, the candidate number is determined as the target number of the information recommendation method; If the candidate quantity is not an integer, the candidate quantity is rounded up to obtain the target quantity of the information recommendation method.

8. The method according to claim 6, characterized in that The target objects to be recommended are divided according to the number of targets corresponding to each of the information recommendation methods to obtain the object sets corresponding to each of the information recommendation methods, including: Generate a random value for each target object to be recommended; The target number corresponding to the first information recommendation method is determined as the first target number, and the target objects to be recommended whose random values ​​are less than the first target number are aggregated to obtain an object set corresponding to the first information recommendation method; The target numbers corresponding to the first i information recommendation methods are summed to obtain the i-th target number, and the target objects to be recommended whose random values ​​are less than the i-th target number and not less than the i-1-th target number are aggregated to obtain the object set corresponding to the i-th information recommendation method, where i is a positive integer greater than 1; Traverse i and execute the above process until the object set corresponding to the Mth information recommendation information mode is obtained.

9. The method according to claim 5, characterized in that The target objects to be recommended are divided to obtain sets of objects corresponding to the information recommendation methods, including: The following processing is performed for each of the information recommendation methods: Obtaining the number of target objects to be recommended that can be recommended by the information recommendation method, and if the number of target objects to be recommended is greater than or equal to the total number, aggregating the target objects to be recommended to obtain an object set corresponding to the information recommendation method; If the carrying quantity is less than the total quantity, the target objects to be recommended of the carrying quantity are aggregated to obtain an object set corresponding to the information recommendation method.

10. The method according to claim 5, characterized in that The step of determining at least one of the information recommendation methods as a target information recommendation method corresponding to the target object to be recommended includes: Acquiring the feedback degree of the target object to be recommended for each of the information recommendation methods, wherein the feedback degree is used to indicate the feedback degree of the target object to be recommended when information is recommended to the target object to be recommended according to the information recommendation method; The information recommendation method with the feedback degree greater than the feedback degree threshold is determined as the target information recommendation method corresponding to the target object to be recommended.

11. The method according to claim 4, characterized in that The step of recommending information to the target object to be recommended according to the target information recommendation method includes: If the total number of the target objects to be recommended is one, sending the recommended information to the target objects to be recommended respectively according to the target information recommendation methods; If the total number of the target objects to be recommended is multiple, for each target object to be recommended, the information to be recommended is sent to the target object to be recommended according to the target information recommendation method corresponding to the target object to be recommended.

12. An information recommendation device, characterized in that: The method comprises: A screening module is used to screen N objects to be recommended to obtain target objects to be recommended that meet the recommendation conditions, where N is a positive integer greater than 1; A determination module, used to determine a target information recommendation method corresponding to the target object to be recommended from M information recommendation methods, where M is a positive integer greater than 1; The recommendation module is used to recommend information to the target object to be recommended according to the target information recommendation method.

13. An electronic device, characterized in that: The electronic device comprises: A memory for storing computer executable instructions or computer programs; A processor, configured to implement the information recommendation method according to any one of claims 1 to 11 when executing the computer executable instructions or computer programs stored in the memory.

14. A computer-readable storage medium storing computer-executable instructions, characterized in that: When the computer executable instructions are executed by a processor, the information recommendation method according to any one of claims 1 to 11 is implemented.

15. A computer program product comprising a computer program or computer executable instructions, characterized in that: When the computer program or computer executable instructions are executed by a processor, the information recommendation method according to any one of claims 1 to 11 is implemented.

Images