Method, apparatus and electronic device for transmitting service information

Abstract

The application discloses a service information sending method and device and electronic equipment. It relates to the field of artificial intelligence, and the method comprises the following steps: obtaining user information of M initial users, and screening attribute information associated with a target service from the user information to obtain M attribute information; obtaining operation information of each initial user on a target platform, and extracting positive feature information and negative feature information from each piece of operation information to obtain M sets of target feature information; determining a service conversion probability value of each initial user according to the associated attribute information and the target feature information; determining an initial user with a service conversion probability value greater than a preset threshold value in the M initial users as a to-be-sent user, and sending service information of the target service to the to-be-sent user. Through the application, the problem of high load, high resource consumption and low running efficiency of an information sending device caused by the full-quantity mass sending strategy in the related art is solved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more specifically, to a method, apparatus, and electronic device for transmitting business information. Background Technology

[0002] Currently, when financial institutions send business information to customers, such as loan application reminders and credit card upgrade notifications, they generally adopt a mass messaging strategy, which means sending business information to all customers.

[0003] However, the aforementioned business information sending strategy does not dynamically identify users' business processing intentions, resulting in a large amount of information being sent to users who have no intention to process the business. This causes information sending devices (such as SMS gateways, message middleware, SMS receiving terminal clusters, mail servers, etc.) to face high-frequency and inefficient call pressure, a surge in system concurrent requests, message queue backlog, increased response latency, and device resources (such as processors, memory, network bandwidth) being occupied by a large number of invalid requests. This significantly reduces system throughput efficiency and increases operation and maintenance costs.

[0004] There is currently no effective solution to the problem of excessive load, high resource consumption, and low operating efficiency of information sending devices caused by the use of full-volume mass messaging strategies in related technologies. Summary of the Invention

[0005] The main purpose of this application is to provide a method, apparatus, and electronic device for sending business information, so as to solve the problems of excessive load, high resource consumption, and low operating efficiency of information sending devices caused by using a full-volume mass sending strategy for information push in related technologies.

[0006] To achieve the above objectives, according to one aspect of this application, a method for sending business information is provided. The method includes: acquiring user information of M initial users, and filtering attribute information associated with a target business from the user information to obtain M attribute information, where M is a positive integer; acquiring operation information of each initial user on a target platform, and extracting positive and negative feature information from each operation information to obtain M sets of target feature information, wherein each set of target feature information includes positive and negative feature information, where positive feature information is positively correlated with the target business, and negative feature information is negatively correlated with the target business; determining the business conversion probability value of each initial user based on the associated attribute information and the target feature information; identifying initial users among the M initial users whose business conversion probability value is greater than a preset threshold as users to be sent to, and sending the business information of the target business to the users to be sent to.

[0007] Optionally, filtering attribute information related to the target business from user information includes: determining the associated attribute dimensions of the target business based on the preset business rule base of the target business, thus obtaining multiple attribute dimensions; and determining the attribute information under each attribute dimension based on the user information, thus obtaining the attribute information of the initial user.

[0008] Optionally, extracting positive feature information from various operation information includes: for any target user, obtaining the target user's first operation information within a first time period, and obtaining the target user's historical transactions within the first time period; obtaining second operation information associated with historical transactions from the first operation information, and determining the target user's operation feature information based on the second operation information; obtaining the target user's third operation information within a second time period, wherein the second time period is later than the first time period and later than the launch time of the target transaction; and selecting positive feature information from the third operation information based on the operation feature information to obtain the target user's positive feature information.

[0009] Optionally, extracting negative feature information from various operation information includes: for any target user, obtaining fourth operation information that is related to the target business from the target user's operation information; obtaining operation information that meets the negative correlation characteristics from the fourth operation information to obtain negative feature information, wherein the negative correlation characteristics are related to the target business.

[0010] Optionally, determining the business conversion probability value for each initial user based on the associated attribute information and target feature information includes: for any target user, inputting the positive feature information from the target user's target feature information into a first prediction model to obtain a first probability value for the target user to handle the target business, wherein the first probability value is trained using a first training set, which includes multiple first sample data, including positive feature information of historical users for historical businesses and the handling status of historical businesses, and the historical businesses and the target business belong to the same business type; inputting the negative feature information from the target user's target feature information into a second prediction model to obtain a second probability value for the target user to handle the target business, wherein the second probability value is trained using a second training set, which includes multiple second sample data, including negative feature information of historical users for historical businesses and the handling status of historical businesses; calculating the average of the first probability value and the second probability value to obtain a target probability value, and determining the decay rate of the target probability value based on the attribute information; multiplying the target probability value by the decay rate to obtain the business conversion probability value of the target user.

[0011] Optionally, determining the decay rate of the target probability value based on attribute information includes: determining the matching degree between the target user and the target service based on attribute information to obtain a first matching degree; obtaining the target user's historical services and determining the matching degree between the historical services and the target service to obtain a second matching degree; determining the target matching degree between the target user and the target service based on the first matching degree and the second matching degree, and determining the decay rate based on the target matching degree, wherein the target matching degree and the decay rate are inversely proportional.

[0012] Optionally, sending the business information of the target business to the user to be sent includes: obtaining preset promotional content of the target business to obtain N preset promotional content, where N is a positive integer; for any user to be sent, generating promotional information based on each preset promotional content to obtain N promotional information; inputting each promotional information, attribute information, and business conversion probability value into a causal effect estimation model, and processing to obtain N estimation results, where each estimation result includes an updated business conversion probability value; selecting the target estimation result with the largest updated business conversion probability value from the N estimation results; and determining the promotional information generated by the preset promotional content to which the target estimation result belongs as the business information of the user to be sent.

[0013] To achieve the above objectives, according to another aspect of this application, a service information transmission apparatus is provided. The apparatus includes: a first acquisition unit, configured to acquire user information of M initial users and filter attribute information associated with a target service from the user information to obtain M attribute information, where M is a positive integer; a second acquisition unit, configured to acquire operation information of each initial user on a target platform and extract positive and negative feature information from each operation information to obtain M sets of target feature information, wherein each set of target feature information includes positive and negative feature information, where positive feature information is positively correlated with the target service and negative feature information is negatively correlated with the target service; a first determination unit, configured to determine the service conversion probability value of each initial user based on the associated attribute information and the target feature information; and a second determination unit, configured to determine the initial users among the M initial users whose service conversion probability value is greater than a preset threshold as users to be sent to, and to send the service information of the target service to the users to be sent to.

[0014] To achieve the above objectives, according to another aspect of this application, an electronic device is provided, the electronic device including a memory storing an executable program; and a processor for running the program, wherein the program executes the above-described method for sending business information during runtime.

[0015] To achieve the above objectives, according to another aspect of this application, a computer program product is provided, including computer instructions that, when executed by a processor, implement the steps of the above-described method for sending business information.

[0016] In this embodiment, user information of M initial users is obtained, and attribute information associated with the target service is filtered out from the user information to obtain M attribute information, where M is a positive integer; operation information of each initial user on the target platform is obtained, and positive and negative feature information is extracted from each operation information to obtain M sets of target feature information, where each set of target feature information includes positive and negative feature information, positive feature information being feature information positively correlated with the target service, and negative feature information being feature information negatively correlated with the target service; the service conversion probability value of each initial user is determined based on the associated attribute information and target feature information; initial users whose service conversion probability value is greater than a preset threshold among the M initial users are determined as users to be sent to. This method involves sending target business information to users only. By determining the business conversion probability value of each initial user based on associated attribute information and target feature information, users with a business conversion probability value greater than a preset threshold are selected as target users. This avoids sending information to all initial users indiscriminately, achieving the goal of sending target business information only to users with a business conversion probability value higher than the preset threshold. This realizes the strategy shift from mass sending to targeted push based on conversion probability value selection. While maintaining the accuracy and effectiveness of information sending, it significantly reduces the amount of information that the device needs to send. This solves the technical problem of excessive load, high resource consumption, and low operating efficiency of information sending devices caused by using mass sending strategies in related technologies. Attached Figure Description

[0017] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 A hardware structure block diagram of a computer terminal for implementing a method of sending business information is shown.

[0019] Figure 2 This is a flowchart of a method for sending business information according to Embodiment 1 of this application;

[0020] Figure 3 This is a schematic diagram of a service information transmission device according to Embodiment 2 of this application;

[0021] Figure 4This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0022] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] It should be noted that the methods, devices, and electronic equipment for transmitting business information as defined in this disclosure can be used in the field of artificial intelligence, or in any field other than artificial intelligence. The application fields of the methods, devices, and electronic equipment for transmitting business information as defined in this disclosure are not limited.

[0026] It should be noted that all information, user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, and displayed data) used in this application are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with the relevant regulations and standards of the relevant regions, have taken necessary measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse use. If the user chooses to refuse, the process proceeds to the expert decision-making process. For example, this system has interfaces with relevant users or organizations. Before obtaining relevant information, a request to obtain the information needs to be sent to the aforementioned user or organization through the interface. After receiving consent from the aforementioned user or organization, the relevant information is obtained. Users can view the purpose of data use in real time through the authorization interface and have the right to withdraw authorization or delete data at any time. After authorization is withdrawn, the system will terminate the relevant data processing within 24 hours.

[0027] The embodiments or examples disclosed herein are not exhaustive, but merely illustrative of some embodiments or examples, and are not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment or example can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment or example can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment or example can be arbitrarily interchanged. Furthermore, optional methods or examples in a particular embodiment or example can be arbitrarily combined; moreover, embodiments or examples can be arbitrarily combined. For example, some or all steps of different embodiments or examples can be arbitrarily combined, and a particular embodiment or example can be arbitrarily combined with optional methods or examples of other embodiments or examples.

[0028] Example 1

[0029] According to an embodiment of this application, an embodiment of a method for sending business information is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1 A hardware block diagram of a computer terminal for implementing a method of sending business information is shown. Figure 1As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, processing devices such as microprocessors or programmable logic devices), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface, a universal serial bus port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0031] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0032] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the business information transmission method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned business information transmission method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0033] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0034] The display may be, for example, a touchscreen LCD display that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0035] Under the aforementioned operating environment, this application provides the following: Figure 2 The method for sending business information is shown. Figure 2 This is a flowchart of a method for sending service information according to Embodiment 1 of this application, as shown below. Figure 2 As shown, the method includes:

[0036] Step S201: Obtain user information for M initial users, and filter out attribute information related to the target business from the user information to obtain M attribute information, where M is a positive integer.

[0037] It should be noted that the execution entity in this embodiment can be a business information sending system. This system can be configured in the information sending device, thereby executing the business information sending method according to the business information sending system, determining the users who need to send information, and sending business information to each user through the information sending device. This achieves the effect of reducing the amount of information that the device needs to send, thereby improving the device's operating efficiency and reducing the device's operating burden.

[0038] It should be noted that the target business is a business that requires the transmission of business information. This business information can be promotional information or business notification information, etc. This embodiment does not limit the type and function of the business information. User information can be authorized information related to the user that can be obtained by financial institutions, such as the user's gender, location, and transactions already processed. Attribute information can be information that is related to the target business, such as the number and names of related transactions processed by the user.

[0039] It should be noted that in each step of this embodiment, whenever it is necessary to obtain information related to the user, the user must be authorized before obtaining the information, and the user's information can only be obtained after the user has completed the authorization operation.

[0040] For example, when determining the users to be sent, it is first necessary to read the user information of M initial users from the user data source of the financial business, and then, according to the business definition of the target business, filter out the attribute information related to the target business from the user information of each user. For example, if the target business is the recommendation of high-end wealth management products, the attribute information may include net asset size, years of investment experience, number of times historically purchased high-risk products, etc., so as to obtain the attribute information of each user, that is, M attribute information.

[0041] Step S202: Obtain the operation information of each initial user on the target platform, and extract positive feature information and negative feature information from each operation information to obtain M sets of target feature information. Each set of target feature information includes positive feature information and negative feature information. Positive feature information is feature information that is positively correlated with the target business, and negative feature information is feature information that is negatively correlated with the target business.

[0042] It should be noted that the operational information on the target platform refers to all interactive behavior logs generated by users on the online service platform deployed by the financial business, including but not limited to page browsing, function clicks, form filling, dwell time, exit path and other behavioral events. Positive feature information refers to the characteristics of user behavior on the target platform that are positively correlated with the target business, that is, the behavioral patterns that indicate that the user shows interest, attention or tendency to accept the business. Negative feature information refers to the characteristics of user behavior that are negatively correlated with the target business, that is, the behavioral patterns that indicate that the user has a tendency to avoid, hesitate or resist the business.

[0043] For example, when determining the user to be sent, it is also necessary to obtain the user's target feature information. First, the system retrieves all the operation information of each initial user on the target platform. The operation information includes page browsing time, number of clicks on functional modules, form filling progress, exit path, collection behavior, cancellation of submission and other behavioral event logs.

[0044] Furthermore, based on the pre-defined positive and negative behavioral rules of the target business, the system analyzes each piece of operation information, identifies and extracts two types of features: positive feature information is the behavioral pattern in which the user shows attention, interest or acceptance tendency, and negative feature information is the behavioral pattern in which the user shows avoidance, hesitation or resistance tendency.

[0045] Furthermore, the system constructs a set of target feature information for each initial user. This set of information consists of a positive feature set and a negative feature set, which together constitute a complete structured expression of the user's behavioral tendencies and are associated with the attribute information corresponding to the user in step S201 through a unique identifier.

[0046] Finally, the system completes the processing of M initial users, forming M sets of target feature information, each set containing both positive and negative features.

[0047] For example, the system first retrieves the user's operation log in the mobile banking application. The log shows that the user clicked on the "Consumer Credit" product page three times consecutively, with a total browsing time of 62 seconds, used the "Interest Rate Calculator" twice, and added the product to favorites, but did not enter the application form filling process. Then, the system determines the user's positive characteristics based on the rules: "viewed the application page ≥ 3 times", "total browsing time ≥ 30 seconds", and "added to favorites", while the negative characteristics are "did not submit an application" and "did not enter the filling process". Further, the system combines the above positive and negative characteristics into a set of target feature information to obtain the user's target feature information.

[0048] Step S203: Determine the business conversion probability value for each initial user based on the associated attribute information and target feature information.

[0049] It should be noted that the business conversion probability value is used to characterize the probability that the initial user will complete the target business or be associated with the target business.

[0050] For example, after obtaining the attribute information and target feature information, the attribute information and target feature information of each initial user can be obtained, and the two can be associated through the user's unique identifier.

[0051] Furthermore, the system can perform rule-based weighted calculations on the user's attribute information and target feature information based on a preset business conversion logic model. The system accumulates scores item by item according to the weighting rules, adding points for positive features and deducting points for negative features. Attribute fields contribute fixed scores according to preset weights, and finally outputs a unique value as the user's business conversion probability value.

[0052] Finally, the system completes the calculation for the M initial users, forming M business conversion probability values, with each value corresponding to one user, thus completing the determination of the business conversion probability values.

[0053] Step S204: Determine the initial users among the M initial users whose business conversion probability value is greater than a preset threshold as users to be sent, and send the business information of the target business to the users to be sent.

[0054] It should be noted that "users to be sent" is a general term for users to whom business information needs to be sent.

[0055] For example, after obtaining the business conversion probability value, the system compares the business conversion probability value of each user with a preset threshold one by one, filters out all initial users whose business conversion probability value is greater than the threshold, marks them as users to be sent to, and sends the business information of the target business only to users marked as users to be sent to. The information can be sent through the registered account of the target platform, SMS channel or application push channel, and the business information can be static information directly related to the target business, such as product description, application portal, preferential terms, etc., thereby completing the entire process of sending business information.

[0056] The method for sending service information provided in this application involves: acquiring user information of M initial users and filtering attribute information associated with the target service from the user information to obtain M attribute information, where M is a positive integer; acquiring operation information of each initial user on the target platform and extracting positive and negative feature information from each operation information to obtain M sets of target feature information, where each set of target feature information includes positive and negative feature information, with positive feature information being positively correlated with the target service and negative feature information being negatively correlated with the target service; determining the service conversion probability value of each initial user based on the associated attribute information and target feature information; and identifying initial users among the M initial users whose service conversion probability value is greater than a preset threshold. This method involves sending target business information to users who are to be sent to. By determining the business conversion probability value of each initial user based on associated attribute information and target feature information, users with a business conversion probability value greater than a preset threshold are selected as users to be sent to. This avoids sending information to all initial users indiscriminately, achieving the goal of sending target business information only to users with a business conversion probability value higher than the preset threshold. This realizes the strategy shift from mass sending to targeted push based on conversion probability value selection. While maintaining the accuracy and effectiveness of information sending, it significantly reduces the amount of information that the device needs to send. This solves the technical problem of excessive load, high resource consumption, and low operating efficiency of information sending devices caused by using mass sending strategies in related technologies.

[0057] Optionally, in the method for sending business information provided in this application embodiment, filtering attribute information associated with the target business from user information includes: determining the associated attribute dimensions of the target business according to the preset business rule base of the target business to obtain multiple attribute dimensions; and determining the attribute information under each attribute dimension according to the user information to obtain the attribute information of the initial user.

[0058] For example, when obtaining the initial user's attribute information, the preset business rule library of the target business is loaded first. This rule library is static configuration data, which contains attribute dimensions directly related to the target business and their definition rules. For example, "credit business" corresponds to dimensions such as "credit score level", "historical number of loan defaults", and "whether or not the user holds a bank card of this bank".

[0059] After reading the rule base, the system parses all related attribute dimensions item by item, forming a structured dimension list, so that each dimension has a locatable field mapping relationship in the data source. Subsequently, for each initial user's original user information, the system matches item by item according to the dimension list, and for each attribute dimension, extracts its corresponding field value in the original data. For example, for the "credit score level" dimension, the system extracts its credit score field and maps it to three-level labels: "A", "B", and "C".

[0060] Finally, an attribute information set consisting of multiple attribute dimension values ​​is generated for each initial user. This set only contains the dimensions defined in the rule base and their corresponding values, and does not contain any other fields not explicitly defined in the rule base, thus completing the acquisition of the initial user's attribute information.

[0061] This embodiment determines the attribute dimensions that need to be filtered by setting a preset business rule base, and filters the attribute information by the attribute dimensions, thereby achieving the technical effect of accurately filtering the attribute information of the initial user.

[0062] Optionally, in the method for sending business information provided in this application embodiment, extracting positive feature information from each operation information includes: for any target user, obtaining the target user's first operation information within a first time period, and obtaining the target user's historical business processed within the first time period; obtaining second operation information associated with the historical business from the first operation information, and determining the target user's operation feature information based on the second operation information; obtaining the target user's third operation information within a second time period, wherein the second time period is later than the first time period and later than the launch time of the target business; selecting positive feature information from the third operation information based on the operation feature information to obtain the target user's positive feature information.

[0063] It should be noted that: the first time period refers to the historical behavior observation window of users on the target platform before the target business goes live, used to establish user behavior benchmarks; the first operation information refers to a series of behavior logs generated by users on the target platform during the first time period, including but not limited to original interaction records such as page visits, function clicks, and information searches; historical business refers to other financial businesses that the target user actually handled during the first time period that have similar functions or scenarios to the target business; the second operation information refers to specific behavior patterns identified from the first operation information that are associated with historical business, used to infer user behavior preferences in similar business scenarios; operation feature information refers to a set of structured features abstracted from the second operation information that reflects the user's behavioral tendencies in specific business types; the second time period refers to the behavior observation window of users on the target platform after the target business goes live, with its start time later than the first time period and later than the official release time of the target business; the third operation information refers to the latest behavior logs generated by users during the second time period, which is the direct data source for positive feature extraction.

[0064] For example, when determining positive feature information, since the benchmarks for positive feature information differ among different users, it is necessary to first determine the user's benchmark information based on the user's historical operation information.

[0065] For any target user, the first step is to determine the user's first operation information within the first time period, which means extracting all the user's operation records within the first time period from the target platform logs to form a set of first operation information.

[0066] Subsequently, the user's historical transactions within the first time period are determined, and second operational information related to these historical transactions is obtained from the first operational information. This includes, for example, repeatedly viewing the details page of the financial product in the historical transaction, repeatedly comparing return parameters, completing the risk assessment questionnaire, and staying on the product description page for more than a threshold duration. This forms a set of behavioral characteristics that have been revealed in the user's historical transactions, namely operational characteristic information. For example, operational characteristic information can be described by a set of structured rules, such as "viewing product pages ≥ 3 times and ≥ 40 seconds in total", "completing the risk assessment and selecting a conservative preference", etc.

[0067] After determining the operational characteristic information, the system can obtain the user's third operational information within a second time period, which is the observation window starting from the official launch of the target service. The system uses the operational characteristic information as a matching template to compare the third operational information item by item, identifying a subset that is highly consistent with the operational characteristic information in terms of behavior type, frequency, duration, path order, etc. This subset is taken as the target user's positive response to the target service, and finally forms the user's positive characteristic information.

[0068] This embodiment determines the user's operation feature information under the target service type based on the user's first operation information related to the historical service type, and then identifies the positive feature information in the user's third operation information based on the operation feature information, thus achieving the technical effect of accurately obtaining the positive feature information of the initial user.

[0069] Optionally, in the method for sending service information provided in this application embodiment, extracting negative feature information from each operation information includes: for any target user, obtaining fourth operation information that is related to the target service from the target user's operation information; obtaining operation information that meets the negative correlation feature from the fourth operation information to obtain negative feature information, wherein the negative correlation feature is related to the target service.

[0070] It should be noted that negative correlation characteristics refer to behavioral behaviors that are inversely related to the conversion intention of the target business.

[0071] For example, when determining negative feature information, the fourth operation information that is functionally or contentally related to the target business can be filtered out from all the user's operation information, and the fourth operation information can be analyzed one by one according to the preset negative correlation feature rule library to identify the subset that meets the negative correlation features.

[0072] Among them, the negative correlation feature rule base is defined based on historical user churn behavior and rejection patterns, including explicit behavioral patterns such as "staying on the target business page for less than the threshold and without any interaction", "visiting multiple times but not entering the application process and eventually closing the page", and "closing the application form within 1 second after opening it".

[0073] When analyzing and identifying negative feature information one by one, the system performs pattern matching on each fourth operation information. If its behavior completely matches any negatively correlated feature rule, it is marked as negative feature information, thereby determining the negative feature information of each target user.

[0074] This embodiment achieves the technical effect of accurately determining negative feature information by identifying negative feature information from the fourth operation information based on negative correlation characteristics.

[0075] Optionally, in the method for sending business information provided in this application embodiment, determining the business conversion probability value of each initial user based on the associated attribute information and target feature information includes: for any target user, inputting the positive feature information from the target user's target feature information into a first prediction model, processing to obtain a first probability value of the target user handling the target business, wherein the first probability value is trained by a first training set, the first training set includes multiple first sample data, the first sample data includes sample positive feature information of historical users for historical businesses and the handling status of historical businesses, and the historical businesses and the target business belong to the same business type; inputting the negative feature information from the target user's target feature information into a second prediction model, processing to obtain a second probability value of the target user handling the target business, wherein the second probability value is trained by a second training set, the second training set includes multiple second sample data, the second sample data includes sample negative feature information of historical users for historical businesses and the handling status of historical businesses; calculating the average of the first probability value and the second probability value to obtain a target probability value, and determining the decay rate of the target probability value based on the attribute information; multiplying the target probability value by the decay rate to obtain the business conversion probability value of the target user.

[0076] It should be noted that the decay rate refers to the weighting coefficient used to adjust the target probability value based on user attribute information, and is used to reflect the inhibitory effect of the user's basic conditions on the conversion probability.

[0077] For example, when determining the conversion probability of any target user among M initial users, it is necessary to process positive and negative feature information separately. First, the system uses the positive feature information from the user's target feature information as input and feeds it into the first prediction model. The first prediction model is a pre-trained supervised learning model, trained on a first training set consisting of multiple historical sample data. Each sample contains positive feature information generated by a historical user in historical business transactions and the actual result of whether the transaction was completed, such as combinations of behaviors like "viewing product pages multiple times, completing risk assessments, and adding products to favorites" along with the label "whether to open an account". The first prediction model learns the non-linear mapping relationship between positive behavior and actual conversion, and outputs the conversion probability of the user based on positive behavior, i.e., the first probability value.

[0078] Similarly, the system uses the negative features from the user's target feature information as input to the second prediction model. The second prediction model is also a pre-trained supervised learning model, trained on a second training set consisting of multiple historical sample data. Each sample contains negative feature information generated by a historical user in historical transactions and their actual conversion results, such as combinations of behaviors like "canceling submission immediately after entering the application page" and "closing immediately after entering the consultation page" along with the label "not processed." The second prediction model learns the inhibitory effect of negative behavior on conversion and outputs the probability of reduced conversion due to the user's negative behavior, i.e., the second probability value.

[0079] After obtaining the first probability value and the second probability value, the first probability value and the second probability value can be arithmetically averaged to obtain the target probability value. Based on the user's attribute information, such as credit score and frequency of use of historical financial products, the corresponding decay rate is determined from the preset decay rate mapping table. The decay rate is used to reflect the strength of the inhibition of conversion probability by insufficient basic conditions. For example, the decay rate of a user with a credit rating of C is 0.6.

[0080] The system multiplies the target probability value by the decay rate to obtain the final business conversion probability value.

[0081] This embodiment achieves the technical effect of accurately determining the user's business conversion probability value by constructing a conversion probability calculation mechanism with dual-model collaboration and attribute decay correction.

[0082] Optionally, in the method for sending service information provided in the embodiments of this application, determining the attenuation rate of the target probability value based on attribute information includes: determining the matching degree between the target user and the target service based on attribute information to obtain a first matching degree; obtaining the historical services of the target user and determining the matching degree between the historical services and the target service to obtain a second matching degree; determining the target matching degree between the target user and the target service based on the first matching degree and the second matching degree, and determining the attenuation rate based on the target matching degree, wherein the target matching degree and the attenuation rate are inversely proportional.

[0083] It should be noted that the matching degree refers to the degree of fit between user attributes and the user conditions required by the target business.

[0084] For example, a two-dimensional matching evaluation mechanism can be used when calculating the decay rate of the target probability value. First, the system compares the target user's attribute information with the business requirements of the target service to determine the first degree of matching. The business requirements can be preset characteristics of the target user group when setting up the service. The first degree of matching can be determined based on the degree to which the user's attribute information conforms to the business requirements.

[0085] Furthermore, the system obtains the user's historical business records, identifies financial products that the user has previously processed that have similar attributes to the target business, and performs a structured comparison between each historical business and the target business in terms of business type, funding period, risk level, deposit mechanism, etc., calculates its similarity score, and takes the average of the matching degree between all historical businesses and the target business as the second matching degree.

[0086] Finally, the first matching degree and the second matching degree are weighted and fused to form the target matching degree. Based on the preset attenuation rate mapping table, the target matching degree is converted into the attenuation rate. The higher the matching degree, the lower the attenuation rate, thus completing the determination of the user's attenuation rate.

[0087] This embodiment determines the matching degree between a user and the target service based on the user's basic attributes and historical characteristics, and then determines the user's attenuation rate based on the inverse relationship between the matching degree and the attenuation rate, thus achieving the technical effect of accurately determining the attenuation rate of the initial user.

[0088] Optionally, in the method for sending business information provided in this application embodiment, sending the business information of the target business to the user to be sent includes: obtaining preset promotional content of the target business to obtain N preset promotional content, where N is a positive integer; for any user to be sent, generating promotional information according to each preset promotional content to obtain N promotional information; inputting each promotional information, attribute information, and business conversion probability value into a causal effect estimation model, and processing to obtain N estimation results, where each estimation result includes an updated business conversion probability value; selecting the target estimation result with the largest updated business conversion probability value from the N estimation results; and determining the promotional information generated by the preset promotional content to which the target estimation result belongs as the business information of the user to be sent.

[0089] It should be noted that the preset promotional content refers to standardized text or structured content modules designed in advance by the business operations department and embedded in the system to convey core information of the target business to users, such as product descriptions and preferential terms. Promotional information refers to personalized push text generated by dynamically combining the preset promotional content with the user's personal attribute information and conversion probability value for a specific target user. The causal effect estimation model refers to a machine learning model trained on historical intervention data to simulate the potential impact of different promotional content on the user's conversion probability. The estimation result refers to the business conversion probability value output by the causal effect estimation model for each promotional message, after simulation correction, reflecting the expected increase in the target user's conversion intention due to the content.

[0090] For example, when sending business information to users, since the target business may have multiple preset promotional contents, such as using different promotional methods and strategies, it is necessary to determine the promotional content that will have the greatest impact on the promotional operation for each user, thereby improving the effectiveness of the business information.

[0091] When determining the promotional content that best resonates with users, promotional messages can be generated based on each preset promotional content, resulting in N promotional messages. These are then input into a causal effect estimation model. The causal effect estimation model is an offline-trained counterfactual inference model, trained on historical promotional experiment data, including actual conversion results for different user groups after receiving different promotional content. The model learns the causal effect of the combination of promotional content and user attributes on conversion probability. After receiving each promotional message, the model outputs an updated business conversion probability value, which represents the estimated increase in user conversion intention under the influence of that promotional content.

[0092] The system generates N promotional messages sequentially from N preset promotional content, inputs each message into the model, and obtains N estimation results. Each estimation result contains an updated business conversion probability value. The system compares all N estimation results and selects the maximum value among the updated business conversion probability values ​​as the target estimation result. The promotional message corresponding to this result is the optimal content that is most likely to induce the current user to convert.

[0093] Finally, the preset promotional content associated with the target estimation result is extracted as the final business information and sent to the target user, thereby completing the business information generation operation and improving the effectiveness of the business information.

[0094] This embodiment introduces a causal effect estimation model, upgrading content delivery from experience-based selection to optimized decision-making based on causal inference, thereby improving the effectiveness and accuracy of business information.

[0095] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0096] Example 2

[0097] This application also provides a service information sending apparatus. It should be noted that the service information sending apparatus of this application can be used to execute the service information sending method provided in the above embodiments. The service information sending apparatus provided in this application will be described below.

[0098] According to an embodiment of this application, an apparatus for implementing the above-described method for sending business information is also provided. Figure 3 This is a schematic diagram of a service information transmission device according to Embodiment 2 of this application, as shown below. Figure 3 As shown, the device includes:

[0099] The first acquisition unit 31 is used to acquire user information of M initial users and filter out attribute information related to the target business from the user information to obtain M attribute information, where M is a positive integer.

[0100] The second acquisition unit 32 is used to acquire the operation information of each initial user on the target platform, and extract positive feature information and negative feature information from each operation information to obtain M sets of target feature information. Each set of target feature information includes positive feature information and negative feature information. Positive feature information is feature information that is positively correlated with the target business, and negative feature information is feature information that is negatively correlated with the target business.

[0101] The first determining unit 33 is used to determine the business conversion probability value of each initial user based on the associated attribute information and target feature information, respectively.

[0102] The second determining unit 34 is used to determine the initial users among the M initial users whose business conversion probability value is greater than a preset threshold as users to be sent, and to send the business information of the target business to the users to be sent.

[0103] The service information sending device provided in this application embodiment acquires user information of M initial users through a first acquisition unit 31, and filters out attribute information associated with the target service from the user information to obtain M attribute information, where M is a positive integer; a second acquisition unit 32 acquires operation information of each initial user on the target platform, and extracts positive feature information and negative feature information from each operation information to obtain M sets of target feature information, where each set of target feature information includes positive feature information and negative feature information, positive feature information being feature information positively correlated with the target service, and negative feature information being feature information negatively correlated with the target service; a first determination unit 33 determines the service conversion probability value of each initial user based on the associated attribute information and target feature information; a second determination unit 34 determines the initial users among the M initial users whose service conversion probability value is greater than a preset threshold as users to be sent, and sends the service information of the target service to the users to be sent. By determining the business conversion probability value of each initial user based on associated attribute information and target feature information, users with a business conversion probability value greater than a preset threshold are selected as users to be sent information. This avoids sending information to all initial users indiscriminately, achieving the goal of sending target business information only to users with a business conversion probability value higher than the preset threshold. This realizes the strategy shift from mass sending to targeted push based on conversion probability value selection. While maintaining the accuracy and effectiveness of information sending, it significantly reduces the amount of information that the device needs to send. This solves the technical problem of excessive load, high resource consumption, and low operating efficiency of information sending devices caused by using mass sending strategies in related technologies.

[0104] Optionally, in the business information sending device provided in the embodiments of this application, the first acquisition unit 31 includes: a first determining module, used to determine the associated attribute dimensions of the target business according to the preset business rule library of the target business, and obtain multiple attribute dimensions; and a second determining module, used to determine the attribute information under each attribute dimension according to the user information, and obtain the attribute information of the initial user.

[0105] Optionally, in the business information sending device provided in this application embodiment, the second acquisition unit 32 includes: a first acquisition module, configured to acquire, for any target user, first operation information of the target user within a first time period, and acquire historical business handled by the target user within the first time period; a second acquisition module, configured to acquire second operation information associated with historical business from the first operation information, and determine the target user's operation feature information based on the second operation information; a third acquisition module, configured to acquire third operation information of the target user within a second time period, wherein the second time period is later than the first time period, and the second time period is later than the launch time of the target business; and a fourth acquisition module, configured to select positive feature information from the third operation information based on the operation feature information to obtain the target user's positive feature information.

[0106] Optionally, in the service information sending device provided in the embodiments of this application, the second acquisition unit 32 includes: a fifth acquisition module, used to acquire fourth operation information related to the target service from the operation information of any target user; and a sixth acquisition module, used to acquire operation information that meets negative correlation characteristics from the fourth operation information to obtain negative feature information, wherein the negative correlation characteristics are related to the target service.

[0107] Optionally, in the business information sending device provided in this application embodiment, the first determining unit 33 includes: a first processing module, used to input positive feature information from the target feature information of any target user into a first prediction model for processing to obtain a first probability value of the target user handling the target business, wherein the first probability value is obtained by training a first training set, the first training set includes multiple first sample data, the first sample data includes sample positive feature information of historical users for historical businesses and the handling status of historical businesses, and the historical businesses and the target business belong to the same business type; a second processing module, used to input negative feature information from the target feature information of the target user into a second prediction model for processing to obtain a second probability value of the target user handling the target business, wherein the second probability value is obtained by training a second training set, the second training set includes multiple second sample data, the second sample data includes sample negative feature information of historical users for historical businesses and the handling status of historical businesses; a first calculation module, used to calculate the average of the first probability value and the second probability value to obtain a target probability value, and determine the decay rate of the target probability value according to attribute information; a second calculation module, used to multiply the target probability value by the decay rate to obtain the business conversion probability value of the target user.

[0108] Optionally, in the service information sending device provided in the embodiments of this application, the first calculation module includes: a first determining submodule, used to determine the matching degree between the target user and the target service based on attribute information to obtain a first matching degree; a second determining submodule, used to obtain the historical services of the target user and determine the matching degree between the historical services and the target service to obtain a second matching degree; and a third determining submodule, used to determine the target matching degree between the target user and the target service based on the first matching degree and the second matching degree, and to determine the attenuation rate based on the target matching degree, wherein the target matching degree and the attenuation rate are inversely proportional.

[0109] Optionally, in the business information sending device provided in this application embodiment, the second determining unit 34 includes: a seventh obtaining module, used to obtain preset promotional content of the target business to obtain N preset promotional content, where N is a positive integer; a first generating module, used to generate promotional information for any user to be sent according to each preset promotional content to obtain N promotional information; a third processing module, used to input each promotional information, attribute information, and business conversion probability value into the causal effect estimation model respectively, and process to obtain N estimation results, where each estimation result includes an updated business conversion probability value; an eighth obtaining module, used to select the target estimation result with the largest updated business conversion probability value from the N estimation results; and a second generating module, used to determine the promotional information generated by the preset promotional content to which the target estimation result belongs as the business information of the user to be sent.

[0110] It should be noted that the first acquisition unit 31, the second acquisition unit 32, the first determination unit 33, and the second determination unit 34 mentioned above correspond to steps S201 to S204 in Embodiment 1. The instances and application scenarios implemented by each of the above units and the corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware components or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above modules can also be part of a device and can run in the computer terminal 10 provided in Embodiment 1.

[0111] Example 3

[0112] Embodiments of this application may provide an electronic device. Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (Only one is shown) processor 1002, memory 1004, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0113] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-described methods. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0114] Those skilled in the art will understand that Figure 4 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones, tablets, handheld computers, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 4 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 4 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 4 The different configurations shown.

[0115] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0116] Example 4

[0117] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the service information transmission method provided in Embodiment 1.

[0118] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0119] Embodiments of this application also provide a computer program product, which, when executed on a data processing device, is a program adapted to perform the steps of a method for sending business information.

[0120] Embodiments of this application also provide a computer-readable storage medium, which includes a stored executable program, wherein the executable program controls the device where the computer-readable storage medium is located to perform the above-described method for sending business information when it runs.

[0121] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0122] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0123] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0124] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0125] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0126] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0127] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for sending business information, characterized in that, include: Obtain user information for M initial users, and filter out attribute information related to the target business from the user information to obtain M attribute information, where M is a positive integer; The operation information of each initial user on the target platform is obtained, and positive feature information and negative feature information are extracted from each operation information to obtain M sets of target feature information. Each set of target feature information includes positive feature information and negative feature information. The positive feature information is the feature information that is positively correlated with the target service, and the negative feature information is the feature information that is negatively correlated with the target service. The business conversion probability value for each initial user is determined based on the associated attribute information and target feature information, respectively. The initial users among the M initial users whose business conversion probability value is greater than a preset threshold are identified as users to be sent to, and the business information of the target business is sent to the users to be sent to.

2. The method according to claim 1, characterized in that, The attribute information related to the target business is filtered from the user information, including: Based on the preset business rule base of the target business, the associated attribute dimensions of the target business are determined, resulting in multiple attribute dimensions; Based on the user information, the attribute information under each attribute dimension is determined to obtain the attribute information of the initial user.

3. The method according to claim 1, characterized in that, Extracting positive feature information from various operational information includes: For any target user, obtain the target user's first operation information within a first time period, and obtain the target user's historical transactions within the first time period; Obtain second operation information associated with the historical service from the first operation information, and determine the operation characteristic information of the target user based on the second operation information; Obtain the third operation information of the target user within a second time period, wherein the second time period is later than the first time period and the second time period is later than the launch time of the target service; Based on the operation feature information, the positive feature information is selected from the third operation information to obtain the positive feature information of the target user.

4. The method according to claim 1, characterized in that, Extracting negative feature information from various operational information includes: For any target user, obtain fourth operation information that is related to the target service from the target user's operation information; The negative feature information is obtained by acquiring operation information that conforms to the negative correlation feature from the fourth operation information, wherein the negative correlation feature is related to the target service.

5. The method according to claim 1, characterized in that, The business conversion probability value for each initial user is determined based on the associated attribute information and target feature information, including: For any target user, the positive feature information from the target feature information of the target user is input into the first prediction model, and the model is processed to obtain the first probability value of the target user handling the target business. The first probability value is obtained by training a first training set, which includes multiple first sample data. The first sample data includes the sample positive feature information of historical users for historical businesses and the handling status of the historical businesses. The historical businesses and the target business belong to the same business type. The negative feature information in the target feature information of the target user is input into the second prediction model, and the second probability value of the target user handling the target business is obtained. The second probability value is obtained by training a second training set, which includes multiple second sample data. The second sample data includes negative feature information of historical users for historical business and the handling status of the historical business. Calculate the average of the first probability value and the second probability value to obtain the target probability value, and determine the decay rate of the target probability value based on the attribute information; Multiply the target probability value by the decay rate to obtain the business conversion probability value of the target user.

6. The method according to claim 5, characterized in that, Determining the decay rate of the target probability value based on the attribute information includes: The matching degree between the target user and the target service is determined based on the attribute information to obtain a first matching degree; Obtain the target user's historical transactions and determine the matching degree between the historical transactions and the target transaction to obtain a second matching degree; The target matching degree between the target user and the target service is determined based on the first matching degree and the second matching degree, and the attenuation rate is determined based on the target matching degree, wherein the target matching degree and the attenuation rate are inversely proportional.

7. The method according to claim 1, characterized in that, Sending the service information of the target service to the user to be sent includes: Obtain the preset promotional content of the target business to get N preset promotional content, where N is a positive integer; For any user to be sent to, generate promotional information based on each preset promotional content to obtain N promotional messages; Each promotional information, the attribute information, and the business conversion probability value are input into the causal effect estimation model, and N estimation results are obtained. Each estimation result includes an updated business conversion probability value. Select the target estimation result with the highest updated business conversion probability value from the N estimation results; The promotional information generated from the preset promotional content to which the target estimation result belongs is determined as the business information of the user to be sent.

8. A device for transmitting business information, characterized in that, include: The first acquisition unit is used to acquire user information of M initial users and filter out attribute information related to the target business from the user information to obtain M attribute information, where M is a positive integer; The second acquisition unit is used to acquire the operation information of each initial user on the target platform, and extract positive feature information and negative feature information from each operation information to obtain M sets of target feature information. Each set of target feature information includes positive feature information and negative feature information. The positive feature information is feature information that is positively correlated with the target service, and the negative feature information is feature information that is negatively correlated with the target service. The first determining unit is used to determine the business conversion probability value of each initial user based on the associated attribute information and target feature information, respectively. The second determining unit is used to determine the initial users among the M initial users whose business conversion probability value is greater than a preset threshold as users to be sent, and to send the business information of the target business to the users to be sent.

9. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the method for sending business information as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, executes the method for sending business information according to any one of claims 1 to 7.

Images