Multi-level service resource pushing method and device, computer device and storage medium

Abstract

This invention discloses a multi-level service resource push method, apparatus, computer equipment, and storage medium, relating to the field of big data processing technology, specifically applicable to the financial and medical fields. The multi-level service resource push method integrates and coordinates user needs with business objectives, ensuring that users in the financial and medical fields can access suitable resources. The method includes: responding to a resource push request, dynamically calculating the number of service leads to be pushed that day based on the time the request is received; filtering multiple target candidate service resources based on service resource location and user preference profiles; grouping the multiple target candidate service resources according to a preset output value target corresponding to each target candidate service resource, prioritizing the resulting resource groups to obtain a priority ranking result; and pushing service resources based on the number of service leads to be pushed that day, combined with the priority ranking result.

Description

Technical Field

[0001] This invention relates to the field of big data processing technology, and can be specifically applied to the financial and medical fields. In particular, it relates to a multi-level service resource push method, apparatus, computer equipment, and storage medium. Background Technology

[0002] In the financial and healthcare sectors, the precise delivery of service resources plays a crucial role in enhancing customer experience, optimizing resource allocation, and strengthening corporate competitiveness. In the financial sector, such as the insurance industry, when customers encounter unexpected situations like car accidents, quickly and accurately directing them to suitable service outlets not only resolves their problems promptly but also enhances customer trust and loyalty to the insurance company through a superior service experience. Similarly, in the healthcare sector, addressing patients' diverse health needs by appropriately directing medical resources, such as specialist consultations and examinations, is essential for improving diagnostic and treatment efficiency and enhancing the patient's healthcare experience.

[0003] In related technologies, service resources are pushed to users based on unified standards or systems. For example, in the insurance industry, some insurance companies recommend branches based on geographical proximity; in the medical field, service resources such as medical institutions are pushed based on the busyness of departments or the doctors' schedules.

[0004] However, the applicant recognizes that the relevant technology has at least the following technical problems in its implementation: Relying on a single push standard makes it impossible to meet the diverse needs of customers in terms of service resource delivery. In the insurance field, customers may be dissatisfied with the poor quality of repairs at service outlets or the long waiting time. In the medical field, patients may delay their treatment because they cannot receive the most suitable expert diagnosis and treatment. It is difficult to guarantee the accuracy and timeliness of resource delivery, which increases operating costs and management difficulty, makes it impossible to achieve optimal resource allocation at the system level, and results in poor flexibility. Summary of the Invention

[0005] In view of this, the present invention provides a multi-level service resource push method, apparatus, computer equipment and storage medium, the main purpose of which is to solve the problems that it is difficult to guarantee the accuracy and timeliness of resource push, which increases operating costs and management difficulty, makes it impossible to achieve optimal resource allocation at the system level and results in poor flexibility.

[0006] According to a first aspect of the present invention, a multi-level service resource push method is provided, the method comprising: In response to a resource push request, the number of service leads to be pushed on that day is dynamically calculated for multiple candidate service resources included in the candidate resource pool, based on the time point at which the resource push request is received. Determine the service resource location indicated by the resource push request and the user preference profile corresponding to the user who initiated the resource push request; based on the service resource location and the user preference profile, filter multiple target candidate service resources from the multiple candidate service resources. Based on the preset output value target corresponding to each of the target candidate service resources, the multiple target candidate service resources are grouped, and based on the user preference profile and the preset resource guidance target corresponding to each of the target candidate service resources, the multiple resource groups obtained after grouping are prioritized to obtain the priority ranking result; Based on the number of service leads to be pushed out that day, and in conjunction with the priority ranking results, service resources are pushed to the user.

[0007] According to a second aspect of the present invention, a multi-level service resource push device is provided, the device comprising: The calculation module is used to respond to a resource push request and dynamically calculate the number of service leads that should be pushed on the same day for multiple candidate service resources included in the candidate resource pool, based on the time point when the resource push request is received. The filtering module is used to determine the service resource location indicated by the resource push request and the user preference profile corresponding to the user who initiated the resource push request, and to filter multiple target candidate service resources from the multiple candidate service resources based on the service resource location and the user preference profile. The sorting module is used to group the multiple target candidate service resources according to the preset output value target corresponding to each target candidate service resource, and to sort the multiple resource groups obtained after grouping according to the user preference profile and the preset resource guidance target corresponding to each target candidate service resource, so as to obtain the priority sorting result; The push module is used to push service resources to the user based on the number of service leads to be pushed on that day and the priority ranking result.

[0008] According to a third aspect of the present invention, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any of the first aspects above.

[0009] According to a fourth aspect of the present invention, a storage medium is provided having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0010] By employing the above technical solutions, this invention provides a multi-level service resource push method, apparatus, computer equipment, and storage medium. In industries such as finance and healthcare, this invention utilizes a multi-level service resource push method to achieve the integration and synergy of user needs and business objectives. It can adjust resource push in real time based on user needs and business conditions, satisfying diverse user needs while further ensuring the accuracy and timeliness of resource push, reducing operating costs and management difficulty. It can achieve optimal resource allocation at the system level, offering good flexibility and ensuring that users in the financial and healthcare fields can obtain appropriate service resources in a timely manner.

[0011] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0012] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a schematic diagram of an application environment for a multi-level service resource push method according to an embodiment of the present invention; Figure 2 This is a flowchart illustrating a multi-level service resource push method in one embodiment of the present invention; Figure 3 This is a flowchart illustrating a specific implementation of step S10; Figure 4 This is a flowchart illustrating a specific implementation of step S20; Figure 5 This is a flowchart illustrating a specific implementation of step S30; Figure 6 This is a flowchart illustrating a specific implementation of step S33; Figure 7 This is a flowchart illustrating a specific implementation of step S40; Figure 8 This is a schematic diagram of a multi-level service resource push device in one embodiment of the present invention; Figure 9 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention; Figure 10 This is another structural schematic diagram of a computer device according to one embodiment of the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] The multi-level service resource push method provided in this invention can be applied to, for example... Figure 1 In this application environment, the client communicates with the server via a network. The server can receive resource push requests from the client and respond to them. Based on the time the resource push request is received, the server dynamically calculates the number of service leads to be pushed to the client that day from multiple candidate service resources in the candidate resource pool. The server determines the location of the service resource indicated by the resource push request and the user preference profile corresponding to the user who initiated the request. Based on the service resource location and user preference profile, the server filters multiple target candidate service resources from the multiple candidate service resources. Based on the preset output value target corresponding to each target candidate service resource, the server groups the multiple target candidate service resources. Based on the user preference profile and the preset resource guidance target corresponding to each target candidate service resource, the server prioritizes the multiple resource groups obtained after grouping, and obtains the priority ranking result. Based on the number of service leads to be pushed that day and the priority ranking result, the server pushes the service resources to the client's user.

[0015] In this invention, a multi-level service resource push method is adopted in industries such as finance and healthcare to achieve the integration and synergy of user needs and business objectives. Resource push can be adjusted in real time based on user needs and business conditions, satisfying diverse user needs while ensuring the accuracy and timeliness of resource push, reducing operating costs and management difficulty. It can achieve optimal resource allocation at the system level, offering good flexibility and ensuring that users in the financial and healthcare fields can obtain appropriate service resources in a timely manner. The client can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a dedicated server or a server cluster consisting of multiple servers. The invention will be described in detail below through specific embodiments.

[0016] Please see Figure 2 As shown, Figure 2 A flowchart illustrating a multi-level service resource push method provided in an embodiment of the present invention includes the following steps: S10: In response to a resource push request, dynamically calculate the number of service leads that should be pushed that day for multiple candidate service resources included in the candidate resource pool, based on the time point at which the resource push request is received.

[0017] The multi-level service resource push method provided by this invention can be applied to resource push systems in various application scenarios. Resource push systems are typically implemented through a server, which can receive resource push requests initiated by users from clients in real time. In response to a resource push request, the system first identifies the time point of the request, including the time interval (beginning of the month, middle of the month, non-beginning of the month, and middle of the month) and the progress of the current service quota (e.g., comparison of the monthly repair amount with the target output). The dynamic closed-loop calculation model built into the resource push system combines these variables with historical conversion rates (e.g., store visit rate) to calculate the number of service leads to be pushed that day. The dynamic closed-loop calculation model is an algorithm that can automatically adjust parameters based on real-time operating data, ensuring a high degree of match between resource allocation and business objectives.

[0018] In this way, the above process enables dynamic and scientific resource allocation, avoiding the rigidity of traditional fixed allocation models, ensuring maximum resource utilization, and strongly supporting the achievement of business objectives. For example, in the insurance sector, if an insurance company finds at the beginning of the month that its vehicle repair output target has not been met, the resource allocation system automatically increases the number of leads pushed to high-potential outlets that day, accelerating the completion of the target. As another example, in the medical field, if a hospital finds insufficient outpatient visits during the peak flu season (mid-month), the resource allocation system dynamically adjusts the number of specialist appointment slots pushed, prioritizing allocation to departments with high demand.

[0019] Among them, such as Figure 3 As shown, in step S10, which involves dynamically calculating the number of service leads to be pushed that day for multiple candidate service resources in the candidate resource pool based on the time point at which the resource push request is received, the following steps are included: S11: Identify the time interval in which the time point is located.

[0020] In this embodiment of the invention, the resource push system identifies the time interval of a given time point, that is, based on the specific date the resource push request is received, it determines whether it is at the beginning or middle of the current month. Here, the time interval is a time division concept used to differentiate calculation strategies, indicating whether the time point is at the beginning or middle of the current month. Specifically, the beginning of the month can refer to the first few days of the month (e.g., the 1st to the 5th), and the middle of the month can refer to the period after a certain number of days have passed but is still in the middle stage (e.g., the 6th to the 25th). The resource push system obtains the current date and compares it with the preset beginning or middle range to determine which clue calculation model should be used.

[0021] This distinction is made because at the beginning of the month, the actual data for that month is insufficient, and predictions need to rely on historical data; while in the middle of the month, the actual data for that month is available for reference, and the calculations can be closer to the real-time situation. Therefore, through the above process, it can be ensured that the resource push system can adaptively select the calculation basis according to different time stages, enhance the applicability and accuracy of the model at different stages of the cycle, and avoid push deviations caused by insufficient or delayed data.

[0022] For example, in the scenario of vehicle damage insurance claims, if a user initiates a request to recommend repair shops on May 3, and the resource recommendation system identifies it as the beginning of the month, then the average number of cases in historical months will be used for calculation; as another example, in the scenario of medical resource allocation, if a user initiates a request to recommend specialist clinics on May 18, and the resource recommendation system identifies it as the middle of the month, then the medical data that has already occurred this month will be used directly for calculation.

[0023] S12: Combine time intervals to calculate the average number of service cases.

[0024] In this embodiment of the invention, the resource push system combines time intervals to calculate the average number of service cases. Specifically, when the time interval indicates the beginning of the current month, the resource push system obtains the total number of service cases over the past several months (e.g., the last six months) and calculates the monthly average number of cases as the average number of service cases. When the time interval indicates the middle of the current month, the resource push system directly calculates the cumulative number of service cases from the beginning of the month to the current date, divides it by the number of days elapsed, and obtains the daily average number of cases for the month as the average number of service cases. The average number of service cases reflects the average number of service requests that the resource push system needs to process per unit of time, and is a basic parameter for subsequent estimation of lead demand. By differentiating time intervals and selecting different statistical sources, the resource push system can still have reasonable predictive capabilities at the beginning of the month when data is incomplete, and gradually transitions to real-time data-driven approaches in the middle of the month, thereby improving the accuracy of lead push prediction and supporting dynamic target management.

[0025] For example, in the vehicle repair recommendation scenario, if there is not enough data for this month's repairs at the beginning of May, the average of 200 repairs per month in the previous six months is taken as the average number of service cases; in the medical department scheduling scenario, if 2,000 visits have been registered by mid-May, this data is used directly to calculate the average daily service volume, which is used to plan the number of patients recommended each day.

[0026] S13: Query the service amount and preset target amount for the current month. If the service amount has not reached the preset target amount, calculate the number of service leads pushed and the theoretical number of service leads pushed for the current month by combining the service amount, the average number of service cases, and the historical conversion rate currently recorded.

[0027] In this embodiment of the invention, the resource push system will query the service quota and the preset target quota for the current month. The service quota refers to the actual amount or quantity of services completed this month, such as the amount of repairs completed this month; the preset target quota is the target output value for the current month.

[0028] If the service expenditure has not reached the preset target, the resource push system will combine the service expenditure, the average number of service cases, and the currently recorded historical conversion rate to calculate the number of service leads already pushed and the theoretical number of service leads to be pushed in the current month. The historical conversion rate, such as the in-store visit rate, refers to the proportion of pushed leads that actually convert into in-store services. The theoretical number of service leads to be pushed can be estimated by "service expenditure ÷ average number of service cases ÷ historical conversion rate," representing the theoretical number of leads that should be pushed to achieve the current output value. The number of service leads already pushed is the actual number of leads pushed to the branch this month. By comparing the two, the resource push system can determine whether the current push has met the theoretical expectations, thus providing a basis for adjusting subsequent push strategies, achieving a dynamic comparison between targets and actual results, and ensuring that push activities are always scientifically controlled around the output value target.

[0029] For example, in a car insurance repair scenario, assuming the target repair revenue for the month is 500,000 yuan, and the current repair revenue is 300,000 yuan, with 600 leads pushed to the branch this month and a historical in-store visit rate of 30%, then the theoretically required number of leads to be pushed is approximately 333 (300,000 yuan ÷ average revenue per vehicle ÷ 30%). By comparing this with the actual 600 leads pushed, the system identifies that the theoretical value has been exceeded and adjusts the subsequent push schedule accordingly. As another example, in healthcare management, if a rehabilitation department aims to serve 300 patients this month, and has currently served 200 with a conversion rate of 40%, then the theoretically required number of leads to be pushed is 500. This is compared with the actual number of leads pushed to assess the efficiency of the push.

[0030] S14: Determine the remaining days of the current month based on the current time, and calculate the number of service leads to be pushed on the current day by combining the number of service leads already pushed, the theoretical number of service leads to be pushed, and the remaining days.

[0031] In this embodiment of the invention, when calculating the number of service leads to be pushed on a given day, the resource push system first determines whether the number of leads already pushed in the current month is greater than the theoretical number of leads to be pushed. If it is determined that the number of leads already pushed in the current month is greater than the theoretical number of leads to be pushed, it indicates that the push effort in the early stages was too strong, and the calculation logic needs to be corrected. Therefore, the resource push system first calculates the number of leads for the remaining month based on the remaining target output value, the average number of service cases, and the historical conversion rate. Specifically, the number of leads for the remaining month can be estimated by calculating "remaining target output value ÷ average number of service cases ÷ historical conversion rate," and the current push volume is dynamically determined using Formula 1 below. Formula 1: Number of leads to be pushed on the current day = Number of leads remaining in the month - (Number of service leads already pushed - Number of service leads that can be pushed on the theory). Formula 1 is primarily used to adjust the pace of subsequent leads based on the initial oversubscription rate. However, if it is determined that the number of leads already pushed in the current month is less than the theoretical number of leads to be pushed, then the basic allocation formula shown in Formula 2 below is used to calculate the number of service leads to be pushed that day. Formula 2: Daily lead generation target = Remaining monthly lead count ÷ Remaining days in the current month. In this embodiment of the invention, the above description addresses the situations at the beginning and middle of the month. However, in practical applications, optionally, if the identified time interval falls outside the beginning and middle of the current month (e.g., the end of the month, date parsing errors, or non-standard time periods), the resource push system will activate backup calculation logic to ensure the continuity of the push plan. In this scenario, the resource push system uses the current month's already served quota and the preset target quota to calculate the remaining service quota, i.e., the difference between the preset target quota and the already served quota is used as the remaining service quota. It also queries the preset default average case volume, a stable parameter based on historical data or business experience, used to replace the dynamically calculated average service case volume when real-time data is unreliable. The resource push system combines the remaining service quota, the default average case volume, and the remaining days to calculate the number of service leads to be pushed that day. By introducing default parameters, it avoids calculation interruptions caused by abnormal time interval identification or data loss, ensuring stable system operation and decision reliability in edge situations, thereby maintaining the timeliness and business continuity of resource pushes. For example, in the insurance sector, when recommending repair shops for vehicles, if the resource push system receives a push request on May 31st (the end of the month) and identifies it as an abnormal period, it uses the default daily average repair output (e.g., 0.4 million yuan per case) and the store visit rate (e.g., 25%) to calculate the number of leads to be pushed that day. As another example, in medical resource scheduling, if a platform in a certain region triggers an abnormal period on the last day of the month due to data synchronization delays, it uses the default daily average number of visits (e.g., 50 visits) and conversion rate (e.g., 30%) to estimate the number of specialist doctors that should be recommended that day.

[0032] Furthermore, in this embodiment of the invention, optionally, when the service amount has reached a preset target amount, the resource push system will set the number of service leads to be pushed that day to 0. This judgment is based on real-time monitoring of the completion of business targets. Once the output value has covered the monthly target, continuing to push leads will not only not help achieve the target, but may also cause excessive resource allocation and burden on the service side. Therefore, by setting the push volume to zero, the resource push system can promptly stop unnecessary resource scheduling, reduce operating costs, improve customer experience, and avoid information overload or service congestion. For example, in the vehicle repair recommendation scenario, if the target output value for this month is 500,000 yuan, and the repair amount so far is 520,000 yuan, the resource push system will automatically stop pushing leads for new repair outlets that day; as another example, in the field of medical and health management, if a rehabilitation department's target for serving 200 patients this month is 210 patients already served, the resource push system will suspend the patient recommendation task for that day and instead focus on follow-up work for the patients already served.

[0033] Furthermore, in this embodiment of the invention, optionally, when the number of service leads already pushed exceeds the theoretical number of service leads to be pushed, the resource push system will execute a dynamic adjustment mechanism to optimize subsequent resource allocation. Specifically, it will determine the remaining quota (i.e., the difference between the two) based on the service quota already served and the preset target quota, calculate the remaining monthly lead volume by combining the average number of service cases served, the remaining quota, and the historical conversion rate, and calculate the number of service leads to be pushed on that day by combining the remaining monthly lead volume, the number of service leads already pushed, and the theoretical number of service leads to be pushed. The specific calculation method is shown in Formula 3 below. Formula 3: Number of service leads to be pushed on the current day = Remaining monthly leads - (Number of service leads already pushed - Theoretical number of service leads to be pushed). In this way, the calculation using Formula 3 enables flexible planning based on actual push efficiency, which can correct potential speed deviations in the early stages and ensure that monthly targets are reasonably decomposed within the remaining time, thereby improving the accuracy of resource allocation and the flexibility at the system level.

[0034] S20: Determine the location of the service resource indicated by the resource push request and the user preference profile corresponding to the user who initiated the resource push request. Based on the service resource location and the user preference profile, filter multiple target candidate service resources from multiple candidate service resources.

[0035] In this embodiment of the invention, the resource push system extracts the service resource location (such as the location of the accident or the user's residence) from the resource push request. At the same time, it calls the user preference profile constructed based on the user's historical behavior data. The user preference profile is a personalized tag system formed by in-depth mining of the user's historical interaction data, which is used to accurately describe the user's needs. Specifically, it may include factors such as the customer's past service outlet repair preferences, service evaluations, and brand preferences. The resource push system will select multiple target candidate service resources that meet the user's geographical location needs and match preferences from the candidate resource pool, thereby breaking through the traditional recommendation logic of only distance, significantly improving the user experience and recommendation acceptance, and enhancing customer stickiness.

[0036] For example, in the insurance industry, when a car owner has an accident in another city, the resource recommendation system filters out high-scoring repair shops that match their past preferences (such as frequently chosen 4S stores) and reviews of nearby service centers based on their experience. Similarly, in the medical field, when a patient schedules an appointment, the resource recommendation system combines their past medical records (preference for chief physicians) and current location to filter out nearby hospital departments that match their preferences.

[0037] Among them, such as Figure 4 As shown, step S20, which involves filtering multiple target candidate service resources from multiple candidate service resources based on the service resource location and user preference profile, includes the following steps: S21: Identify the preset service location of each candidate service resource, and extract multiple first candidate service resources from multiple candidate service resources whose preset service location is consistent with the service resource location.

[0038] In this embodiment of the invention, the resource push system traverses each candidate service resource in the candidate resource pool and reads its pre-defined preset service location information. The preset service location is a fixed service coverage area for each candidate service resource. Then, the resource push system compares these preset service locations with the service resource locations in the received resource push request. The service resource location is the specific location where the user currently needs to obtain the service. If the two match, the corresponding candidate service resource is extracted to form multiple first candidate service resources.

[0039] In this way, by accurately matching the preset service locations of candidate service resources with the service resource locations in the request, the system can initially filter out service resources that meet the user's geographical needs. This provides a foundation for further filtering based on user preferences, ensuring that the recommended resources meet geographical accessibility requirements and improving the effectiveness of resource delivery. For example, in the insurance field, if a user's vehicle is involved in an accident in region A, the resource delivery system identifies the preset service locations of various repair shops and extracts the repair shops with the preset service location in region A as the first candidate service resource. As another example, in the medical field, if a user needs medical services in region B, the resource delivery system filters out hospitals or clinics with the preset service location in region B as the first candidate service resource.

[0040] S22: Determine the user's preferred service location based on the user preference profile, and extract multiple second candidate service resources from multiple candidate service resources whose preset service location matches the user's preferred service location.

[0041] In this embodiment of the invention, the resource recommendation system invokes a user preference profile constructed for the user who initiated the resource recommendation request. This profile is generated based on the user's historical behavior data and includes information about the user's preferred service locations. The system determines the user's preferred service locations based on the profile; these might be frequently chosen service areas or locations with which the user has emotional attachment. Next, the system iterates through the candidate resource pool, comparing the preset service locations of the candidate service resources with the user's preferred service locations. If they match, they are extracted, forming multiple second candidate service resources. This process, combined with the user preference profile, determines the preferred service locations and extracts the corresponding service resources, fully considering the user's personalized needs. This makes the recommended resources more aligned with the user's psychological expectations, improving user acceptance and satisfaction with the recommended resources, and further optimizing the quality of resource recommendations.

[0042] For example, in the insurance field, if a user has repeatedly chosen region C for vehicle repairs, the resource recommendation system determines their preference for region C based on the user's preference profile and extracts repair shops in region C as second-choice service resources. As another example, in the healthcare field, if a user is accustomed to or trusts a hospital in region D, the resource recommendation system extracts hospitals in region D as second-choice service resources based on their preference profile.

[0043] S23: Use multiple first candidate service resources and multiple second candidate service resources as multiple target candidate service resources.

[0044] In this embodiment of the invention, the resource recommendation system integrates multiple extracted first candidate service resources and multiple second candidate service resources together as multiple target candidate service resources for subsequent processes. These target candidate service resources satisfy both geographical matching with the service resource locations and user personalized preferences, representing a set of service resources initially selected from the candidate resource pool that better meet user needs. By comprehensively considering both geographical location and user preferences in determining target candidate service resources, the selected resources demonstrate good performance in terms of geographical convenience and user acceptance, laying a solid foundation for subsequent accurate service resource recommendations. This effectively improves the accuracy of resource recommendations and user satisfaction, achieving a better integration of user needs and resource recommendations.

[0045] For example, in the insurance sector, resource recommendation systems integrate repair shops whose service locations match the accident location and the user's preferred area into target candidate service resources, which are then used to recommend suitable repair shops for the vehicle. Similarly, in the healthcare sector, hospitals that meet both the service location requirements and the user's preferred location are integrated into target candidate service resources to recommend suitable medical resources to the user.

[0046] S30: Based on the preset output value target corresponding to each target candidate service resource, group multiple target candidate service resources, and based on the user preference profile and the preset resource guidance target corresponding to each target candidate service resource, prioritize the multiple resource groups obtained after grouping to obtain the priority ranking result.

[0047] In this embodiment of the invention, the resource recommendation system groups each target candidate service resource according to its preset output value target (such as the monthly repair amount required by a service point). The preset output value target is a quantitative indicator set by a service point or department to achieve its business objectives, used to guide the direction of resource allocation. Simultaneously, combining user preference profiles and the preset resource guidance objectives corresponding to each resource (such as prioritizing service points that have not met their output value targets), the resource groups are prioritized to obtain a priority ranking result.

[0048] In this way, the recommendation logic is upgraded from service-driven to business-driven, ensuring that resources are accurately directed to branches or departments with revenue-generating needs. For example, in the insurance industry, if a branch's revenue is below target but customer preference is high, the resource recommendation system will appropriately adjust its ranking to prioritize it, satisfying user needs while promoting business goals. As another example, in the medical field, if a department needs to increase outpatient volume, the resource recommendation system will prioritize recommending appointments to patients who prefer that department, while balancing resources from other departments to avoid idle resources.

[0049] Among them, such as Figure 5As shown, in step S30, that is, according to the preset output value target corresponding to each target candidate service resource, multiple target candidate service resources are grouped, and according to the user preference profile and the preset resource guidance target corresponding to each target candidate service resource, the multiple resource groups obtained after grouping are prioritized to obtain the priority ranking result, including the following steps: S31: Query the preset output value target corresponding to each target candidate service resource, and count the completed output value corresponding to each target candidate service resource.

[0050] In this embodiment of the invention, the resource recommendation system queries a pre-set output value target from the corresponding database records for each selected target candidate service resource. The pre-set output value target is the expected output value set based on factors such as the service resource's business plan and market potential. Simultaneously, the resource recommendation system calculates the output value already achieved by each target candidate service resource since the start of the statistical period, i.e., the completed output value. This can be specifically calculated by obtaining actual business data through integration with the business system.

[0051] In this way, by accurately querying preset output targets and calculating completed output, a crucial data foundation is provided for the subsequent rational grouping and sorting of service resources. This ensures that resource delivery is closely integrated with the operational status of service resources, guaranteeing that the delivery strategy aligns with actual business needs. For example, in the insurance sector, the resource delivery system queries the preset annual output targets of each vehicle repair outlet and simultaneously calculates the vehicle repair revenue completed by each outlet this month. As another example, in the healthcare sector, it queries the preset quarterly medical revenue targets of each hospital and calculates the medical revenue amount achieved this month.

[0052] S32: Based on the preset output value target and completed output value corresponding to each target candidate service resource, divide multiple target candidate service resources into multiple resource groups.

[0053] In this embodiment of the invention, the resource recommendation system classifies each target candidate service resource into different resource groups based on the relationship between its preset output value target and its completed output value. Specifically, there are three categories: First, resource groups where the preset output value target is not 0 and the completed output value is less than the preset output value target, indicating that these service resources still have room for output value improvement; second, resource groups where the preset output value target is not 0 and the completed output value is greater than the preset output value target, indicating that they have completed or exceeded their target; and third, resource groups where the preset output value is 0, indicating that these resources may be in a special operating state or newly established without a target set.

[0054] This grouping method clearly identifies service resources in different operational states, facilitating the development of differentiated delivery strategies based on the characteristics of each group. This ensures that resource delivery is more aligned with the actual operational status of the service resources, improving the rationality of resource allocation. For example, in the insurance sector, outlets that did not meet their monthly repair revenue targets, outlets that met or exceeded their targets, and outlets without pre-set targets can be grouped into corresponding resource groups. Similarly, in the healthcare sector, hospitals that did not meet their quarterly medical revenue targets, hospitals that met or exceeded their targets, and hospitals without pre-set targets can be grouped.

[0055] S33: Within each resource group, the target candidate service resources included in each resource group are sorted based on the user preference profile, and multiple resource groups are sorted according to the group priority corresponding to each resource group to obtain the priority ranking result.

[0056] In this embodiment of the invention, within each predefined resource group, the resource recommendation system sorts the target candidate service resources within the group based on the user's preference profile. Simultaneously, multiple resource groups are sorted according to their respective group priorities. For example, resource groups with a preset output value target of not being zero but not yet achieved have higher priority, while resource groups with a preset output value target of zero may have lower priority. This sorting by user preference within resource groups fully considers users' personalized needs, increasing user acceptance of recommended resources. Furthermore, sorting resource groups optimizes the resource recommendation order from an overall operational perspective. The combination of these two approaches achieves the integration of user needs and operational goals, ensuring the accuracy and timeliness of resource recommendations and realizing optimal resource allocation at the system level.

[0057] For example, in the insurance sector, repair outlets that haven't met their revenue targets are sorted by user preference for distance and brand; outlets with a pre-set revenue target of 0 are also sorted by user preference, and then the overall ranking is determined by group priority, which is used to recommend repair outlets to users. Similarly, in the healthcare sector, hospitals that haven't met their revenue targets are sorted by user preference for hospital departments, distance, etc.; hospitals with a pre-set revenue target of 0 are sorted by user preference, and then the final ranking is determined by group priority, pushing medical resources to users.

[0058] Among them, such as Figure 6 As shown, in step S33, that is, within each resource group, the target candidate service resources included in each resource group are sorted based on the user preference profile, including the following steps: S331: For each resource group, the multiple target candidate service resources included in the resource group are treated as multiple resources to be sorted.

[0059] In this embodiment of the invention, for each resource group, the multiple target candidate service resources contained in the group are directly determined as multiple resources to be sorted. These resources to be sorted are the objects of the subsequent sorting operation. They have different attributes and characteristics. For example, in the insurance vehicle repair shop recommendation scenario, each repair shop to be sorted has its unique location, service capabilities, user reviews and other attributes.

[0060] In this way, by clearly defining the resources to be sorted, a foundation is laid for subsequent precise sorting based on user preferences. This ensures that the sorting operation is carried out on specific service resources, giving the entire sorting process a clear target. For example, in the insurance industry, in the group of repair outlets that have not met their preset revenue targets, each repair outlet is designated as a resource to be sorted. Similarly, in the healthcare industry, for the group of hospitals that have not met their preset healthcare revenue targets, each hospital is identified as a resource to be sorted.

[0061] S332: Based on user preference profiles, count the number of times a user has accessed each resource to be sorted in history, and calculate the distance between each resource to be sorted and the user's current location based on the user's current location.

[0062] In this embodiment of the invention, the resource recommendation system combines the user's past selection and interaction with various service resources recorded in the user preference profile to count the number of times the user has accessed each resource to be ranked. Simultaneously, based on the user's current location, it uses technologies such as Geographic Information Systems (GIS) to calculate the distance between each resource to be ranked and the user's current location. For example, in an insurance scenario, it counts the number of times the user has visited different repair shops in the past, and calculates the straight-line distance or actual travel distance between each repair shop and the user's current location.

[0063] In this way, by statistically analyzing historical access counts, the system reflects users' preferences for different service resources, while calculating distance intervals considers the ease with which users access services. Obtaining these two sets of data provides crucial information for subsequent rational sorting based on user preferences and geographical location. For example, in the insurance sector, the resource recommendation system queries user preference profiles to determine the number of times a user has visited certain repair shops in the past, and simultaneously uses map software to calculate the distance between these shops and the user's current location. As another example, in the healthcare sector, the system statistically analyzes the number of times a user has visited different hospitals in the past and calculates the distance between each hospital and the user's current location.

[0064] S333: Extract multiple first unsorted resources from multiple unsorted resources whose historical access count reaches the standard threshold and whose distance interval meets the distance limit. Sort the multiple first unsorted resources according to the historical access count and distance interval corresponding to each first unsorted resource to obtain the first sorting result.

[0065] In this embodiment of the invention, the resource push system sets a standard access threshold and a distance limit. It extracts multiple first-order resources from a pool of resources to be sorted that have reached the standard access threshold and whose distance intervals meet the distance limit. Then, it performs a comprehensive sorting based on the historical access count (higher priority for more accesses) and distance interval (closer priority for closer resources) of each first-order resource to obtain a first sorting result. For example, if the standard access threshold is set to 3 times and the distance limit is 5 kilometers, then repair outlets with 3 historical accesses and within 5 kilometers of the user will be sorted as first-order resources.

[0066] This approach prioritizes resources that meet the criteria of higher access frequency and closer proximity, quickly filtering out resources that users are more likely to be interested in and that are easier to access. This improves the quality of recommended resources and user acceptance, making the ranking results more aligned with users' actual needs. For example, in the insurance industry, a standard access frequency threshold of 2 visits and a distance limit of 10 kilometers are set, prioritizing repair shops that users have visited more than twice and are within 10 kilometers of them as the first resources to be ranked. Similarly, in the healthcare industry, a standard access frequency threshold of 3 visits and a distance limit of 8 kilometers are used to rank hospitals that meet these criteria.

[0067] S334: In addition to the multiple first unsorted resources, extract other unsorted resources whose historical access count reaches the minimum threshold and whose distance interval meets the distance limit. Extract multiple second unsorted resources whose resource attributes meet the preset attribute labeling from the other unsorted resources. Sort the multiple second unsorted resources according to the historical access count and distance interval corresponding to each second unsorted resource to obtain the second sorting result.

[0068] In this embodiment of the invention, the resource push system extracts other resources to be sorted from those other than the multiple first-to-sort resources, whose historical access count reaches a minimum threshold and whose distance interval meets the distance limit. Then, from these resources, it further filters out multiple second-to-sort resources whose resource attributes match preset attribute labels. The preset attribute labels can be attributes such as the brand, level, or special services of the service resources. Then, the resources are sorted according to the historical access count and distance interval corresponding to each second-to-sort resource to obtain a second sorting result. For example, if the minimum access threshold is set to 1 time, and the preset attribute label is "comprehensive repair shop and vehicle center or key repair shop," then repair shops with 1 historical access, meeting the distance limit, and being comprehensive repair shops and vehicle centers or key repair shops will be sorted as second-to-sort resources.

[0069] In this way, based on the initial ranking, resources that meet certain conditions and possess specific attributes are further explored, enriching the types and levels of recommended resources, meeting diverse user needs, and taking into account user preferences and geographical location factors. For example, in the insurance field, assuming a minimum threshold of 1 visit and a preset attribute label of "chain repair brand," repair outlets that meet the criteria are ranked. As another example, in the medical field, assuming a minimum threshold of 2 visits and a preset attribute label of "top-tier hospitals," hospitals that meet the requirements are ranked.

[0070] S335: Determine the remaining unsorted resources, integrate the remaining unsorted resources to obtain the third sorting result, and combine the first sorting result, second sorting result and third sorting result according to the priority corresponding to the first sorting result, second sorting result and third sorting result to complete the sorting operation of the target candidate service resources included in the resource group. The remaining unsorted resources are the unsorted resources other than the multiple first unsorted resources and multiple second unsorted resources among the multiple unsorted resources.

[0071] In this embodiment of the invention, the resource push system identifies the remaining resources to be sorted from multiple unsorted resources, excluding multiple first unsorted resources and multiple second unsorted resources. These resources are integrated to obtain a third sorting result. Then, according to the priority of the first sorting result, the second sorting result, and the third sorting result (usually the first sorting result has the highest priority, the second sorting result is next, and the third sorting result is last), the three are combined to complete the sorting operation of the target candidate service resources included in the resource group.

[0072] In this way, the above process ensures that all resources to be sorted are rationally arranged, forming a complete sorting system. This ensures that the service resources ultimately recommended to users consider both high-frequency access and proximity needs, while also covering other resources that meet certain conditions, comprehensively meeting user needs and achieving optimal resource delivery configuration. For example, in the insurance field, repair outlets not included in the first and second sorting results are integrated and sorted as remaining resources, and then the three sorting results are combined according to priority to recommend repair outlets to users. As another example, in the medical field, after integrating and sorting the remaining hospital resources, the three sorting results are combined to push medical resources to users.

[0073] S40: Based on the number of service leads to be pushed out that day, and combined with the priority ranking results, push service resources to users.

[0074] In this embodiment of the invention, the resource push system generates a final recommendation list based on the calculated number of service leads to be pushed that day, combined with the priority ranking results. High-priority resources are prioritized for recommendation, while ensuring the total number does not exceed the daily push limit. Furthermore, if high-priority resources are insufficient, the resource push system automatically supplements secondary-priority resources, forming a dynamically adjusted push strategy. Through this process, the accuracy and timeliness of resource push can be achieved, meeting diverse user needs while optimizing system-level resource allocation and reducing operating costs and management complexity.

[0075] For example, in the insurance sector, if a car owner needs vehicle repairs, the resource recommendation system will recommend three service centers (two high-priority and one low-priority) based on the owner's preferences and the service center's revenue targets, ensuring a balance between recommendation quality and business objectives. As another example, in the medical field, if a patient schedules surgery, the resource recommendation system will prioritize recommending specialist teams with tighter surgical schedules, based on the patient's preferences and the department's capacity.

[0076] Among them, such as Figure 7 As shown, step S40, which involves pushing service resources to users based on the number of service leads to be pushed that day and the priority ranking results, includes the following steps: S41: Query the number of service leads pushed on the same day for each target candidate service resource in the priority ranking results, and calculate the difference between the number of service leads that should be pushed on the same day and the number of service leads that have been pushed on the same day for each target candidate service resource.

[0077] In this embodiment of the invention, the resource recommendation system first queries each target candidate service resource in the priority ranking results, and accurately finds the number of service leads that have been pushed out for each target candidate service resource on that day from the relevant data records. This data reflects the number of times the resource has been recommended to users. At the same time, based on the previously calculated number of service leads that should be pushed to each target candidate service resource on that day, the difference between the number that should be pushed and the number that has been pushed is calculated. This difference in the number of leads directly reflects how many push opportunities for each target candidate service resource are still unused.

[0078] In this way, by accurately querying and calculating the difference in the number of leads, the system can clearly grasp the push progress and remaining push space for each target candidate service resource. This provides precise data support for subsequent reasonable resource pushes, ensuring that resource pushes are neither excessive nor incomplete, meeting the balance requirements of business objectives and user needs. For example, in the insurance field, the resource push system might find that a vehicle repair shop has pushed 3 leads that day, while the required number is 8, resulting in a lead difference of 5. As another example, in the medical field, for a hospital, the number of leads pushed that day is 2, while the required number is 6, resulting in a difference of 4.

[0079] S42: According to the order of each target candidate service resource in the priority ranking result, push the corresponding target candidate service resources with a positive difference in the number of clues to the user in sequence to complete the service resource push to the user.

[0080] In this embodiment of the invention, the resource push system checks the difference in the number of leads for each target candidate service resource according to the order in which each target candidate service resource is ranked in the priority ranking result. This order is derived by comprehensively considering factors such as user preferences and business objectives. Starting from the first target candidate service resource, the system checks whether the difference in the number of leads is positive. If the difference is positive, the system pushes the target candidate service resource to the user until the total number of service leads to be pushed that day is reached or all qualified resources have been pushed.

[0081] In this way, by prioritizing and pushing target candidate service resources with positive differences, we can ensure that high-priority resources are recommended to users first, meeting their core needs and preferences, while also rationally allocating resources based on remaining push opportunities to avoid waste. This achieves accuracy and timeliness in resource delivery and optimizes resource allocation at the system level. For example, in the insurance industry, we can prioritize and check the difference in the number of leads for each repair outlet, recommending outlets with positive differences to users until the daily push target is met. Similarly, in the healthcare industry, we can prioritize and recommend hospitals with positive difference in the number of leads, ensuring users receive appropriate medical resources promptly.

[0082] In summary, the resource push system in this embodiment of the invention exhibits highly intelligent and refined characteristics in service resource push, achieving efficient and rational resource allocation through multi-dimensional strategies. Regarding the determination of daily lead push volume, it breaks through the traditional fixed or evenly distributed model, constructing a dynamic closed-loop calculation model that deeply integrates with actual business operations. The model incorporates the monthly cycle as a time dimension element in real time. Different stages of business focus and resource allocation needs vary; at the beginning of the month, the emphasis may be on target planning and resource reserves, while in the middle of the month, more attention is paid to progress control and adjustments. Simultaneously, the resource push system is closely linked to the completion of business targets, comparing the monthly repair amount with the target output value to clearly understand the business situation. It also fully considers historical conversion rates such as store visit rates, comprehensively performing multi-variable calculations. Once a delay in target progress is detected, the model automatically increases the weight of theoretical lead estimation, forming a complete closed loop of "target-execution-feedback-adjustment," ensuring refined and scientific lead resource management and effectively promoting the achievement of business targets.

[0083] Furthermore, to enhance user experience and conversion rates, the resource recommendation system employs a customer preference-first strategy centered on user historical behavior. It abandons the simplistic "nearest" recommendation approach and deeply analyzes customer historical behavior data, incorporating past service point preferences and multi-dimensional information about the customer and service points—such as customer characteristics, case information, and service point characteristics—as key weights for service point ranking. This approach prioritizes recommending familiar and trusted service points to users, fully considering the match between users and service points, greatly respecting user preferences, and increasing user acceptance of recommended service points. This effectively enhances customer loyalty and fosters long-term partnerships.

[0084] Furthermore, for the complex and unique scenario of "out-of-town accidents" in the financial insurance sector, the resource recommendation system is designed with a dedicated repair recommendation model that always prioritizes customer preferences. For example, customers who wish to have their repairs done near the accident site will be given priority. Building on this, the model deeply integrates operational guidance, comprehensively considering factors such as the achievement of output targets by surrounding branches. In this way, it provides out-of-town customers with intelligent repair solutions that meet both convenience needs and align with the overall interests of the service network, effectively addressing the problem of simplistic recommendation logic in traditional systems and improving service quality and resource allocation rationality in special scenarios.

[0085] In summary, the resource recommendation system in this embodiment of the invention implements a goal-driven, refined operational priority strategy that deeply integrates recommendation logic with operational indicators. Based on the operational status of outlets, such as "whether output value is 0" and "whether output value has been achieved," the system accurately classifies and ranks outlets, enabling recommendation behavior to transcend mere service and become a precise operational adjustment tool. The resource recommendation system can accurately direct leads to outlets with revenue-generating needs or those that have not yet achieved their goals, fully exploring the commercial value of every recommendation opportunity. This achieves an upgrade from the traditional "service-driven" to "operation-driven" model, optimizing overall operational efficiency.

[0086] The method provided in this invention employs a multi-level service resource push approach in industries such as finance and healthcare to achieve the integration and synergy of user needs and business objectives. It can adjust resource push in real time according to user needs and business conditions, meeting diverse user needs while further ensuring the accuracy and timeliness of resource push, reducing operating costs and management difficulty, achieving optimal resource allocation at the system level, and providing good flexibility to ensure that users in the financial and healthcare fields can obtain appropriate service resources in a timely manner.

[0087] Furthermore, as Figure 1 In a specific implementation of the method, this embodiment of the invention provides a multi-level service resource push device, such as... Figure 8As shown, the device includes: a calculation module 801, a filtering module 802, a sorting module 803, and a push module 804.

[0088] The calculation module 801 is used to respond to a resource push request and dynamically calculate the number of service leads that should be pushed on the same day for multiple candidate service resources included in the candidate resource pool, based on the time point at which the resource push request is received. The filtering module 802 is used to determine the service resource location indicated by the resource push request and the user preference profile corresponding to the user who initiated the resource push request, and to filter multiple target candidate service resources from the multiple candidate service resources according to the service resource location and the user preference profile. The sorting module 803 is used to group the multiple target candidate service resources according to the preset output value target corresponding to each target candidate service resource, and to sort the multiple resource groups obtained after grouping according to the user preference profile and the preset resource guidance target corresponding to each target candidate service resource, so as to obtain the priority sorting result. The push module 804 is used to push service resources to the user based on the number of service leads to be pushed on that day and the priority ranking result.

[0089] In a specific application scenario, the calculation module 801 is used to identify the time interval in which the time point is located, where the time interval indicates that the time point is at the beginning or middle of the current month; combining the time interval, it calculates the average number of service cases, wherein when the time interval indicates that the time point is at the beginning of the current month, the average number of service cases is obtained by calculating the average number of cases in a specified historical time interval, and when the time interval indicates that the time point is in the middle of the current month, the average number of service cases is obtained by calculating the average number of cases in the current month; it queries the service quota and preset target quota for the current month, and if the service quota has not reached the preset target quota, it calculates the number of service leads pushed and the theoretical number of service leads pushed for the current month by combining the service quota, the average number of service cases pushed, and the currently statistically analyzed historical conversion rate; it determines the remaining days of the current month based on the current time point, and calculates the number of service leads to be pushed on the current day by combining the number of service leads pushed, the theoretical number of service leads pushed, and the remaining days.

[0090] In specific application scenarios, the calculation module 801 is further configured to, when identifying that the time interval of the time point indicates that the time point is in an interval other than the beginning and middle of the current month, calculate the remaining service quota using the service quota already served in the current month and the preset target quota, and query the preset default case average volume, and calculate the number of service leads to be pushed on the current day by combining the remaining service quota, the default case average volume, and the remaining days; or, when the service quota already served reaches the preset target quota, set the number of service leads to be pushed on the current day to 0; or, when the number of service leads already pushed is greater than the theoretical number of service leads to be pushed, determine the remaining quota based on the service quota already served and the preset target quota, calculate the remaining monthly lead volume by combining the service case average volume, the remaining quota, and the historical conversion rate, and calculate the number of service leads to be pushed on the current day by combining the remaining monthly lead volume, the number of service leads already pushed, and the theoretical number of service leads to be pushed.

[0091] In a specific application scenario, the filtering module 802 is used to identify the preset service location where each candidate service resource is located, extract a plurality of first candidate service resources whose preset service location is consistent with the service resource location from the plurality of candidate service resources; determine the user's preferred service location based on the user preference profile, extract a plurality of second candidate service resources whose preset service location is consistent with the user's preferred service location from the plurality of candidate service resources; and use the plurality of first candidate service resources and the plurality of second candidate service resources as the plurality of target candidate service resources.

[0092] In a specific application scenario, the sorting module 803 is used to query the preset output value target corresponding to each of the target candidate service resources, and to count the completed output value corresponding to each of the target candidate service resources; based on the preset output value target and completed output value corresponding to each of the target candidate service resources, the multiple target candidate service resources are divided into multiple resource groups, wherein the multiple resource groups include resource groups with a preset output value target not equal to 0 and a completed output value less than the preset output value target, resource groups with a preset output value target not equal to 0 and a completed output value greater than the preset output value target, and resource groups with a preset output value of 0; within each resource group, the target candidate service resources included in each resource group are sorted in conjunction with the user preference profile, and the multiple resource groups are sorted according to the group priority corresponding to each resource group to obtain the priority sorting result.

[0093] In a specific application scenario, the sorting module 803 is used to, for each resource group, treat multiple target candidate service resources included in the resource group as multiple resources to be sorted; combine the user preference profile to count the historical access count of the user to each resource to be sorted, and according to the user's current location, count the distance interval between each resource to be sorted and the user's current location; extract multiple first resources to be sorted from the multiple resources to be sorted that have a historical access count that reaches a standard threshold and a distance interval that meets the distance limit, and sort the multiple first resources to be sorted according to the historical access count and distance interval corresponding to each first resource to be sorted to obtain a first sorting result; and sort the remaining resources (excluding the multiple first resources to be sorted) that have a historical access count that reaches a minimum threshold and a distance interval that meets the distance limit. For other unsorted resources that are not subject to restrictions, extract multiple second unsorted resources whose resource attributes match preset attribute labels from the other unsorted resources, and sort the multiple second unsorted resources according to the historical access count and distance interval corresponding to each second unsorted resource to obtain a second sorting result; determine the remaining unsorted resources, integrate the remaining unsorted resources to obtain a third sorting result, and combine the first sorting result, the second sorting result and the third sorting result according to the priority corresponding to the first sorting result, the second sorting result and the third sorting result to complete the sorting operation of the target candidate service resources included in the resource group, wherein the remaining unsorted resources are the unsorted resources other than the multiple first unsorted resources and the multiple second unsorted resources among the multiple unsorted resources.

[0094] In a specific application scenario, the push module 804 is used to query the number of service leads pushed to each target candidate service resource on the same day in the priority ranking result, and to calculate the difference in the number of service leads to be pushed to each target candidate service resource on the same day and the number of service leads already pushed on the same day; according to the order of each target candidate service resource in the priority ranking result, the corresponding target candidate service resources with a positive difference in the number of leads are pushed to the user in sequence to complete the service resource push to the user.

[0095] The device provided in this invention employs a multi-level service resource push method in industries such as finance and healthcare to achieve the integration and synergy of user needs and business objectives. It can adjust resource push in real time according to user needs and business conditions, meeting diverse user needs while further ensuring the accuracy and timeliness of resource push, reducing operating costs and management difficulty, achieving optimal resource allocation at the system level, and providing good flexibility to ensure that users in the financial and healthcare fields can obtain appropriate service resources in a timely manner.

[0096] Specific limitations regarding the multi-level service resource push device can be found in the limitations of the multi-level service resource push method described above, and will not be repeated here. Each module in the aforementioned multi-level service resource push device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in the computer device in hardware form, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0097] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external clients via a network connection. When the computer program is executed by the processor, it implements the functions or steps of a multi-level service resource push method on the server side.

[0098] In one embodiment, a computer device is provided, which may be a client, and its internal structure diagram may be as follows: Figure 10 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with an external server via a network connection. When the computer program is executed by the processor, it implements the client-side functions or steps of a multi-level service resource push method.

[0099] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: In response to a resource push request, the number of service leads to be pushed on that day is dynamically calculated for multiple candidate service resources included in the candidate resource pool, based on the time point at which the resource push request is received. Determine the service resource location indicated by the resource push request and the user preference profile corresponding to the user who initiated the resource push request; based on the service resource location and the user preference profile, filter multiple target candidate service resources from the multiple candidate service resources. Based on the preset output value target corresponding to each of the target candidate service resources, the multiple target candidate service resources are grouped, and based on the user preference profile and the preset resource guidance target corresponding to each of the target candidate service resources, the multiple resource groups obtained after grouping are prioritized to obtain the priority ranking result; Based on the number of service leads to be pushed out that day, and in conjunction with the priority ranking results, service resources are pushed to the user.

[0100] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor: In response to a resource push request, the number of service leads to be pushed on that day is dynamically calculated for multiple candidate service resources included in the candidate resource pool, based on the time point at which the resource push request is received. Determine the service resource location indicated by the resource push request and the user preference profile corresponding to the user who initiated the resource push request; based on the service resource location and the user preference profile, filter multiple target candidate service resources from the multiple candidate service resources. Based on the preset output value target corresponding to each of the target candidate service resources, the multiple target candidate service resources are grouped, and based on the user preference profile and the preset resource guidance target corresponding to each of the target candidate service resources, the multiple resource groups obtained after grouping are prioritized to obtain the priority ranking result; Based on the number of service leads to be pushed out that day, and in conjunction with the priority ranking results, service resources are pushed to the user.

[0101] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0102] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this invention are all information and data authorized by the user or fully authorized by all parties.

[0103] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0104] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0105] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A multi-level service resource push method, characterized in that, include: In response to a resource push request, the number of service leads to be pushed on that day is dynamically calculated for multiple candidate service resources included in the candidate resource pool, based on the time point at which the resource push request is received. Determine the service resource location indicated by the resource push request and the user preference profile corresponding to the user who initiated the resource push request; based on the service resource location and the user preference profile, filter multiple target candidate service resources from the multiple candidate service resources. Based on the preset output value target corresponding to each of the target candidate service resources, the multiple target candidate service resources are grouped, and based on the user preference profile and the preset resource guidance target corresponding to each of the target candidate service resources, the multiple resource groups obtained after grouping are prioritized to obtain the priority ranking result; Based on the number of service leads to be pushed out that day, and in conjunction with the priority ranking results, service resources are pushed to the user.

2. The method according to claim 1, characterized in that, The step of dynamically calculating the number of service leads to be pushed that day for multiple candidate service resources included in the candidate resource pool based on the time point of receiving the resource push request includes: Identify the time interval in which the time point is located, wherein the time interval is used to indicate that the time point is at the beginning or middle of the current month; The average number of service cases is calculated by combining the time intervals. Specifically, when the time interval indicates that the time point is at the beginning of the current month, the average number of service cases is obtained by calculating the average number of cases in the specified historical time interval. When the time interval indicates that the time point is in the middle of the current month, the average number of service cases is obtained by calculating the average number of cases in the current month. Query the service quota and preset target quota for the current month. If the service quota has not reached the preset target quota, calculate the number of service leads pushed and the theoretical number of service leads pushed for the current month by combining the service quota, the average number of service cases, and the currently statistical historical conversion rate. Based on the current time, determine the remaining days of the current month, and combine the number of service leads already pushed, the theoretical number of service leads to be pushed, and the remaining days to calculate the number of service leads to be pushed on that day.

3. The method according to claim 2, characterized in that, The method further includes: If the time interval of the identified time point indicates that the time point is outside the beginning and middle of the current month, the remaining service quota is calculated using the service quota already served in the current month and the preset target quota, and the preset default case average is queried. Combining the remaining service quota, the default case average, and the remaining days, the number of service leads to be pushed out that day is calculated; or, If the already served amount reaches the preset target amount, the number of service leads to be pushed out that day will be set to 0; or, If the number of service leads already pushed is greater than the theoretical number of service leads to be pushed, the remaining quota is determined based on the service quota already served and the preset target quota. The remaining monthly lead volume is calculated by combining the average number of service cases, the remaining quota, and the historical conversion rate. The number of service leads to be pushed on the current day is calculated by combining the remaining monthly lead volume, the number of service leads already pushed, and the theoretical number of service leads to be pushed.

4. The method according to claim 1, characterized in that, The step of filtering multiple target candidate service resources from the multiple candidate service resources based on the service resource location and the user preference profile includes: Identify the preset service location where each of the candidate service resources is located, and extract a plurality of first candidate service resources whose preset service location is consistent with the service resource location from the plurality of candidate service resources; Based on the user preference profile, the user's preferred service location is determined, and multiple second candidate service resources whose preset service locations are consistent with the user's preferred service locations are extracted from the multiple candidate service resources; The plurality of first candidate service resources and the plurality of second candidate service resources are used as the plurality of target candidate service resources.

5. The method according to claim 1, characterized in that, The process of grouping the multiple target candidate service resources according to the preset output value target corresponding to each target candidate service resource, and prioritizing the multiple resource groups obtained after grouping according to the user preference profile and the preset resource guidance target corresponding to each target candidate service resource, to obtain a priority ranking result, includes: Query the preset output value target corresponding to each of the target candidate service resources, and count the completed output value corresponding to each of the target candidate service resources; Based on the preset output value target and the completed output value corresponding to each target candidate service resource, the multiple target candidate service resources are divided into multiple resource groups, wherein the multiple resource groups include resource groups with a preset output value target of not being 0 and a completed output value less than the preset output value target, resource groups with a preset output value target of not being 0 and a completed output value greater than the preset output value target, and resource groups with a preset output value of 0. In each resource group, the target candidate service resources included in each resource group are sorted according to the user preference profile, and the multiple resource groups are sorted according to the group priority corresponding to each resource group to obtain the priority sorting result.

6. The method according to claim 5, characterized in that, The step of sorting the target candidate service resources included in each resource group based on the user preference profile includes: For each resource group, the multiple target candidate service resources included in the resource group are treated as multiple resources to be sorted. Based on the user preference profile, the historical access count of the user to each of the resources to be sorted is counted, and the distance interval between each resource to be sorted and the user's current location is counted according to the user's current location. Extract multiple first resources from the multiple resources to be sorted that have reached the standard access count threshold and whose distance interval meets the distance limit. Sort the multiple first resources to be sorted according to the historical access count and distance interval corresponding to each first resource to be sorted to obtain a first sorting result. In addition to the multiple first unsorted resources, other unsorted resources whose historical access counts reach the minimum threshold and whose distance intervals meet the distance limit are extracted. Among these other unsorted resources, multiple second unsorted resources whose resource attributes meet the preset attribute labels are extracted. The multiple second unsorted resources are sorted according to the historical access counts and distance intervals corresponding to each second unsorted resource to obtain a second sorting result. The remaining unsorted resources are determined, and the remaining unsorted resources are integrated to obtain a third sorting result. According to the priority corresponding to the first sorting result, the second sorting result, and the third sorting result, the first sorting result, the second sorting result, and the third sorting result are combined to complete the sorting operation of the target candidate service resources included in the resource group. The remaining unsorted resources are the unsorted resources other than the multiple first unsorted resources and the multiple second unsorted resources among the multiple unsorted resources.

7. The method according to claim 1, characterized in that, The step of pushing service resources to the user based on the number of service leads to be pushed that day, combined with the priority ranking result, includes: Query the number of service leads pushed on the same day for each target candidate service resource in the priority ranking result, and calculate the difference in the number of service leads that should be pushed on the same day for each target candidate service resource and the number of service leads that have been pushed on the same day. According to the order of each target candidate service resource in the priority ranking result, the target candidate service resources with a positive difference in the number of clues are pushed to the user in sequence to complete the service resource push to the user.

8. A multi-level service resource push device, characterized in that, include: The calculation module is used to respond to a resource push request and dynamically calculate the number of service leads that should be pushed on the same day for multiple candidate service resources included in the candidate resource pool, based on the time point when the resource push request is received. The filtering module is used to determine the service resource location indicated by the resource push request and the user preference profile corresponding to the user who initiated the resource push request, and to filter multiple target candidate service resources from the multiple candidate service resources based on the service resource location and the user preference profile. The sorting module is used to group the multiple target candidate service resources according to the preset output value target corresponding to each target candidate service resource, and to sort the multiple resource groups obtained after grouping according to the user preference profile and the preset resource guidance target corresponding to each target candidate service resource, so as to obtain the priority sorting result; The push module is used to push service resources to the user based on the number of service leads to be pushed on that day and the priority ranking result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images