Intelligent allocation method, device and equipment for customer incoming line resources

Abstract

This invention relates to an intelligent allocation method, apparatus, and device for customer call resources, belonging to the field of resource allocation technology. The method identifies customers by their customer identifiers to determine if they have historical behavior data, thereby differentiating customer call history and other information for targeted resource allocation. When no historical behavior data exists, resources are allocated according to real-time allocation rules. When historical behavior data exists, resources are pre-allocated based on call time prediction models, ensuring timely allocation. By pre-allocating historical customers, the allocation results are directly invoked when a customer calls, which is convenient and fast. Simultaneously, it solves the current problem of not being able to achieve accurate and personalized resource allocation based on dynamic information such as customer call history, leading to resource mismatch, duplicate customer operations, poor user experience, and uneven utilization of employee or community resources. This method enables accurate and personalized resource allocation after a customer calls.

Description

Technical Field

[0001] This invention relates to the field of resource allocation technology, and specifically to an intelligent allocation method, apparatus, and equipment for customer inbound resources. Background Technology

[0002] In the field of customer relationship management and marketing, companies typically use WeChat customer service, mini-programs and other channels to guide customers to add employees' personal WeChat accounts or join communities in order to establish long-term reach channels.

[0003] In existing technologies, resource allocation after a customer comes in is relatively simple, with common methods including random allocation, fixed allocation (e.g., channel 1 always allocated to A, channel 2 always allocated to B, etc.), or simple round-robin allocation. However, these methods all have the following technical problems:

[0004] The inability to accurately and personally allocate resources based on dynamic information such as customer call history, friend addition records, and group joining records leads to problems such as resource mismatch, duplicate customer operations, poor user experience, and uneven utilization of employee or community resources. For example, the same customer may be repeatedly assigned to different employees if they call multiple times in a short period of time, causing customer confusion; customers who have already added an employee or joined a group may still be repeatedly pushed the same resources, resulting in invalid operations; after an employee leaves or a group is disbanded, the system may still assign customers to expired resources, leading to customer churn.

[0005] Therefore, how to accurately and personally allocate resources after a customer comes online has become a pressing technical problem that needs to be solved in the existing technology. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to provide a method, apparatus and device for intelligent allocation of customer call resources, so as to overcome the current problem that it is impossible to achieve accurate and personalized resource allocation based on dynamic information such as customer call history, friend addition records, and group entry records, resulting in resource mismatch, repeated customer operations, poor user experience and uneven utilization of employee or community resources, so as to achieve accurate and personalized resource allocation after the customer calls.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: On the one hand, an intelligent allocation method for customer inbound resources includes: In response to a customer's incoming call request, the customer identifier of the current customer is obtained from the incoming call request; wherein, the customer identifier of each customer is unique; Determine whether there is historical behavior data for a preset time period under the current customer identifier; wherein, the historical behavior data is the behavior record associated with the customer identifier and stored before and after each customer completes the call process; If no historical behavior data for the preset time period exists under the current customer identifier, then resources are allocated to the current customer according to the real-time allocation rules to obtain the target resources; If historical behavior data for the preset time period exists under the current customer identifier, then based on the customer identifier, the corresponding pre-allocated resources are queried from the pre-allocation result storage area, and the queried pre-allocated resources are used as the target resources for the current customer; wherein, the pre-allocated resources in the pre-allocation result storage area are stored after the time of the customer's future call request is predicted in advance based on the call time prediction model, and resources are allocated in advance based on the prediction results; the call time prediction model is trained and constructed based on the customer's historical behavior data; Send the target resource to the client.

[0008] Optionally, determining whether historical behavior data for a preset time period exists under the current customer identifier includes: If there is no historical behavior data under the current customer identifier, or if the current customer's current call time is more than a preset period apart from the last call time, then it is determined that there is no historical behavior data for the preset time period under the current customer identifier; wherein, the current customer's current call time is determined based on the customer's call request. If there is historical behavior data under the current customer identifier, and the time between the current customer's last call and the time between the last call does not exceed a preset period, then it is determined that there is historical behavior data for a preset time period under the current customer identifier.

[0009] Optionally, the step of allocating resources to the current customer according to real-time allocation rules and obtaining target resources includes: The incoming call request includes an incoming call mode, which includes a one-to-one friend adding mode and a community joining mode; According to the incoming line mode, a target resource is allocated from the corresponding candidate resource pool in a round-robin order; The candidate resource pool for the one-to-one friend-adding mode consists of currently online and available employee personal QR codes or customer acquisition links; the candidate resource pool for the community joining mode consists of currently active group QR codes that are not yet full.

[0010] Optionally, the method for constructing the arrival time prediction model includes: Collect customers’ historical behavior data, which includes at least the customer’s call time, QR code scanning time and scanning channel; Using the historical behavior data as training samples, an independent call-in time prediction model is constructed for each customer with historical behavior data. The model outputs the probability value of the customer initiating a call-in request within a first preset time window; wherein, the customer's QR code scan corresponds one-to-one with the call-in request.

[0011] Optionally, the method for constructing the pre-allocated result storage area includes: The probability of a customer initiating an incoming call request within a first preset time window is predicted based on the aforementioned time prediction model. When it is predicted that the probability of a customer initiating an incoming call request in the next time window exceeds a preset threshold, a candidate resource is locked for the customer before the next time window. The locked candidate resource is used as a pre-allocated resource, and the locking relationship is stored in the pre-allocation result storage area as a pre-allocation result. The candidate resource is determined based on the customer's historical allocation records and resource load status. The update methods for the pre-allocation result storage area include: recalculating the pre-allocated resources of all or part of the customers in batches according to a fixed time period, and / or triggering the recalculation of the pre-allocated resources of any customer when the historical behavior data of any customer is updated.

[0012] Optionally, the candidate resources are determined based on the customer's historical allocation records and resource load status, including: Obtain the number of times each resource has been allocated in the customer's historical allocation records, and calculate the historical preference score for each resource; Get the real-time load rate of each resource and calculate the load score for each resource; For each resource, calculate the overall score = α × historical preference score + β × (1 - real-time load rate), where α and β are preset weighting coefficients; The resource with the highest overall score will be selected as the candidate resource for this customer.

[0013] Optionally, if historical behavior data for the preset time period exists under the current customer identifier, but the pre-allocated resources queried in the pre-allocated result storage area are unavailable, a multi-level degradation allocation strategy is executed; the multi-level degradation allocation strategy includes: Obtain historical allocation records from the historical behavior data under the customer identifier. The historical allocation records include at least the most recently successfully allocated resource and the allocation time for the customer. Determine whether the most recently successfully allocated resource is available. If it is available, use the resource as the target resource. If the most recently successfully allocated resource is unavailable, then based on the customer's historical scanning records in the customer's historical behavior data, obtain the resource that the customer scanned the most times in the most recent second preset time window, and determine whether the resource is available. If the resource is available, then use the resource as the target resource. If the resource with the most scans by the customer within the most recent second preset time window is unavailable, a dynamic selection sub-strategy will be executed; the dynamic selection sub-strategy includes: Obtain the status information of all currently available resources, including at least the real-time load rate of each resource, the number of currently queued customers, and the average service duration; Based on the historical allocation records in the customer's historical behavior data, calculate the historical association strength S_i between the customer and each available resource i. The historical association strength S_i includes at least one or more of the following: the number of times the customer has been allocated to resource i in history, the number of times the customer has scanned resource i in history, and the number of times the customer has completed a call through resource i in history. For each available resource i, calculate its comprehensive score. The calculation rule for the comprehensive score is as follows: R_i=λ·(1-L_i)+μ·(1-Q_i / Q_max)+ν·(S_i / S_max), Where L_i is the real-time load rate of resource i, Q_i is the current number of queued customers of resource i, Q_max is the preset maximum queuing threshold, S_max is the maximum historical association strength among all resources, and λ, μ, and ν are preset weight coefficients. Select the resource with the highest overall score R_i as the target resource; After the downgrade allocation is completed, the pre-allocated resources for that customer are recalculated, and the newly calculated pre-allocated resources are updated in the pre-allocation result storage area.

[0014] Optionally, after sending the target resource to the client, the process further includes: The parameters of the incoming call time prediction model are updated, and the deviation between the customer's incoming call request initiation time and the predicted incoming call request initiation time is used as a feedback signal to adjust the prediction accuracy of the model.

[0015] On another front, an intelligent allocation device for customer incoming line resources includes: The acquisition module is used to obtain the customer identifier of the current customer in response to a customer's incoming call request; wherein, the customer identifier of each customer is unique; The judgment module is used to determine whether there is historical behavior data for a preset time period under the current customer identifier; wherein, the historical behavior data is the behavior record associated with the customer identifier and stored before and after each customer completes the call process; The allocation module is used to allocate resources to the current customer according to real-time allocation rules and obtain target resources when there is no historical behavior data for the preset time period under the current customer identifier. The retrieval module is used to, when historical behavior data for the preset time period exists under the current customer identifier, query the corresponding pre-allocated resources from the pre-allocated result storage area based on the customer identifier, and use the queried pre-allocated resources as the target resources for the current customer; wherein, the pre-allocated resources in the pre-allocated result storage area are stored after the time of the customer's future call request is predicted in advance based on the call time prediction model, and resources are allocated in advance based on the prediction results; the call time prediction model is trained and constructed based on the customer's historical behavior data; The sending module is used to send the target resource to the client.

[0016] On another front, an intelligent allocation device for customer incoming line resources includes a processor and a memory, wherein the processor is connected to the memory: The processor is used to call and execute the program stored in the memory; The memory is used to store the program, which is at least used to execute the intelligent allocation method for customer incoming line resources as described in any of the preceding claims.

[0017] The technical solutions described in the embodiments of the present invention have at least the following beneficial effects: The purpose of this invention is to provide an intelligent allocation method, apparatus, and device for customer call resources. It identifies customers by their customer identifiers to determine if they have historical behavior data, thereby differentiating customer call history and other information for targeted resource allocation. When no historical behavior data exists, resources are allocated according to real-time allocation rules. When historical behavior data exists, resources are pre-allocated based on call time prediction models, ensuring timely allocation. By pre-allocating historical customers, the allocation results are directly invoked when a customer calls, which is convenient and fast. It also solves the current problem of not being able to achieve accurate and personalized resource allocation based on dynamic information such as customer call history, friend addition records, and group joining records, leading to resource mismatch, duplicate customer operations, poor user experience, and uneven utilization of employee or community resources. This invention provides accurate and personalized resource allocation after a customer calls. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating an intelligent allocation method for customer inbound resources according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an intelligent allocation device for customer incoming line resources according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an intelligent allocation device for customer incoming line resources provided in an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0021] As described in the background section, existing technologies cannot achieve accurate and personalized resource allocation based on dynamic information such as customer call history, friend addition records, and group joining records. This leads to problems such as resource mismatch, duplicate customer operations, poor user experience, and uneven utilization of employee or community resources. For example, the same customer may be repeatedly assigned to different employees if they call multiple times in a short period of time, causing customer confusion; a customer may still be repeatedly pushed the same resources after adding an employee or joining a group, resulting in invalid operations; and after an employee leaves or a group is disbanded, the system may still assign customers to expired resources, leading to customer churn.

[0022] Therefore, how to accurately and personally allocate resources after a customer comes online has become a pressing technical problem that needs to be solved in the existing technology.

[0023] Based on this, embodiments of the present invention provide a method, apparatus, and device for intelligent allocation of customer call resources to solve the problems of current methods being unable to achieve accurate and personalized resource allocation based on dynamic information such as customer call history, friend addition records, and group joining records, resulting in resource mismatch, duplicate customer operations, poor user experience, and uneven utilization of employee or community resources. This invention enables accurate and personalized resource allocation after a customer initiates a call. Example 1 Figure 1 This is a flowchart illustrating an intelligent allocation method for customer inbound resources according to an embodiment of the present invention. Please refer to [link / reference]. Figure 1 This embodiment may include the following steps: Step S101: In response to a customer's incoming call request, obtain the current customer's customer identifier from the incoming call request; wherein, the customer identifier of each customer is unique; In this embodiment, the intelligent allocation method for customer incoming line resources proposed in this application can be integrated into a processor for execution and deployed in a server.

[0024] In a specific implementation, a customer can initiate an incoming call request through the client. The incoming call request can include the customer's customer identifier, incoming call time, incoming call mode, scanning link / channel, etc. Each customer's customer identifier is unique.

[0025] Step S102: Determine whether there is historical behavior data for a preset time period under the current customer identifier; wherein, the historical behavior data is the behavior record associated with the customer identifier and stored before and after each customer completes the call process; The historical period can be one month, or 10 days, 2 months, etc., and this application does not make a specific limitation.

[0026] Each time a customer initiates a call request and completes a call, the customer's identifier is associated with the record and the behavior is stored for later retrieval.

[0027] After obtaining the customer's customer identifier, it is determined whether there is historical behavioral data for a preset time period under this customer identifier.

[0028] Step S103: If there is no historical behavior data for the preset time period under the current customer identifier, then allocate resources to the current customer according to the real-time allocation rules and obtain the target resources; When it is determined that there is no historical behavior data for a preset time period under the current customer identifier, the real-time allocation rules are executed to obtain the target resources. These target resources can be employee personal QR codes, customer acquisition links, or group QR codes.

[0029] Step S104: If historical behavior data for the preset time period exists under the current customer identifier, then according to the customer identifier, query the corresponding pre-allocated resources from the pre-allocation result storage area, and use the queried pre-allocated resources as the target resources for the current customer; wherein, the pre-allocated resources in the pre-allocation result storage area are stored after the time of the customer's future call request is predicted in advance based on the call time prediction model, and resources are allocated in advance according to the prediction results; the call time prediction model is trained and constructed based on the customer's historical behavior data; Understandably, in this application, historical behavior data of each customer is pre-collected as samples for training, thereby constructing a call-in time prediction model for that user, and outputting a probability value for initiating a call-in request in a future time period. Based on this probability value, resources are allocated in advance to obtain pre-allocated resources, and then these pre-allocated resources are associated with the customer and placed in the pre-allocation result storage area for later retrieval.

[0030] When it is determined that there is no historical behavior data for a preset time period under a customer identifier, the pre-allocated resource associated with that customer identifier is directly called from the pre-allocated result storage area and used as the target resource.

[0031] Step S105: Send the target resource to the client.

[0032] Once a customer joins the call, the system obtains the target resource and returns it to the client so that the customer can scan the code to join the call.

[0033] Understandably, the technical solution provided in this implementation allows for the identification of customers based on their historical behavior data, thereby differentiating customer call history and other information for targeted resource allocation. In the absence of historical behavior data, resources are allocated according to real-time allocation rules. When historical behavior data is available, resources are pre-allocated based on call time prediction models, ensuring timely allocation. Specifically, by pre-allocating historical customers, the allocation results are directly retrieved when a customer calls, which is convenient and fast. Furthermore, it solves the current problem of not being able to achieve accurate and personalized resource allocation based on dynamic information such as customer call history, friend addition records, and group joining records, leading to resource mismatch, duplicate customer operations, poor user experience, and uneven utilization of employee or community resources. This allows for accurate and personalized resource allocation after a customer calls.

[0034] Example 2 Based on the above embodiment 1, the present invention also provides another embodiment to illustrate the determination of whether there is historical behavior data for a preset time period under the current customer identifier.

[0035] In this embodiment, determining whether historical behavior data for a preset time period exists under the current customer identifier includes: If there is no historical behavior data under the current customer identifier, or if the current customer's current call time is more than a preset period apart from the last call time, then it is determined that there is no historical behavior data for the preset time period under the current customer identifier; wherein, the current customer's current call time is determined based on the customer's call request. If there is historical behavior data under the current customer identifier, and the time between the current customer's last call and the time between the last call does not exceed a preset period, then it is determined that there is historical behavior data for a preset time period under the current customer identifier.

[0036] The preset period can be 1 month, 2 months, 6 months, etc.

[0037] It is understood that by adopting the technical solution provided in this embodiment, and setting a preset period, the scope of new users can be adjusted. When there is no historical behavior data under a customer identifier, or when the time since the last call is too long, the user is determined to be a new user, thus confirming that there is no historical behavior data for the preset time period under that customer identifier. Conversely, the user is determined to be an old user, thereby facilitating the pre-allocation process.

[0038] Example 3 Based on the above embodiment 1, the present invention also provides another embodiment to illustrate the real-time allocation rules.

[0039] In this embodiment, the step of allocating resources to the current customer according to real-time allocation rules and obtaining target resources includes: The incoming call request includes an incoming call mode, which includes a one-to-one friend adding mode and a community joining mode; According to the incoming line mode, a target resource is allocated from the corresponding candidate resource pool in a round-robin order; The candidate resource pool for the one-to-one friend-adding mode consists of currently online and available employee personal QR codes or customer acquisition links; the candidate resource pool for the community joining mode consists of currently active group QR codes that are not yet full.

[0040] It is understood that by adopting the technical solution provided in this embodiment, when executing the real-time allocation rule, a target resource is allocated from the corresponding candidate resource pool according to the polling order, which facilitates allocation.

[0041] The online access modes can include one-on-one friend adding mode and community joining mode. It is understood that in the one-on-one friend adding mode, the candidate resource pool consists of currently online and available employee personal QR codes or customer acquisition links; the candidate resource pool for the community joining mode consists of currently active group QR codes that are not yet full.

[0042] Example 4 Based on the above embodiment 1, the present invention also provides another embodiment, which is illustrated by the method of constructing the line-in time prediction model.

[0043] In this embodiment, the method for constructing the incoming line time prediction model includes: Collect customers’ historical behavior data, which includes at least the customer’s call time, QR code scanning time and scanning channel; Using the historical behavior data as training samples, an independent call-in time prediction model is constructed for each customer with historical behavior data. The model outputs the probability value of the customer initiating a call-in request within a first preset time window; wherein, the customer's QR code scan corresponds one-to-one with the call-in request.

[0044] In this invention, the server receives an incoming call request every time a customer scans a code. Therefore, the probability of a customer initiating an incoming call request is equivalent to the probability of a customer scanning a code.

[0045] It is understood that by adopting the technical solution provided in the embodiments of the present invention, by pre-constructing an incoming call time prediction model, customer incoming calls can be predicted in advance, and resources can be pre-allocated.

[0046] Example 5 Based on the above embodiment 1, the present invention also provides another embodiment to illustrate the method for constructing a pre-allocated result storage area.

[0047] In this embodiment, the method for constructing the pre-allocated result storage area includes: The probability of a customer initiating an incoming call request within a first preset time window is predicted based on the aforementioned time prediction model. When it is predicted that the probability of a customer initiating an incoming call request in the next time window exceeds a preset threshold, a candidate resource is locked for the customer before the next time window. The locked candidate resource is used as a pre-allocated resource, and the locking relationship is stored in the pre-allocation result storage area as a pre-allocation result. The candidate resource is determined based on the customer's historical allocation records and resource load status. The update methods for the pre-allocation result storage area include: recalculating the pre-allocated resources of all or part of the customers in batches according to a fixed time period, and / or triggering the recalculation of the pre-allocated resources of any customer when the historical behavior data of any customer is updated.

[0048] In this embodiment, each time window can be defined according to requirements, such as one minute, one second, etc., and no specific limitation is made in this embodiment. The preset threshold can be 95% or 92%, and no specific limitation is made in this application.

[0049] It is understood that by adopting the technical solution provided in this embodiment, predictions can be made for each predictable customer in real time, and pre-allocation can be performed when the probability of a customer making a request in any time window exceeds a preset threshold. This allows the pre-allocation results of multiple customers to form a pre-allocation result storage area for easy and quick retrieval. Furthermore, setting update calculations enables real-time updates of the pre-allocated resources.

[0050] Example 6 Based on Embodiment 5 above, the present invention also provides another embodiment to illustrate the candidate resources.

[0051] In this embodiment, the candidate resources are determined based on the customer's historical allocation records and resource load status, including: Obtain the number of times each resource has been allocated in the customer's historical allocation records, and calculate the historical preference score for each resource; Get the real-time load rate of each resource and calculate the load score for each resource; For each resource, calculate the overall score = α × historical preference score + β × (1 - real-time load rate), where α and β are preset weighting coefficients; The resource with the highest overall score will be selected as the candidate resource for this customer.

[0052] In this application, α and β can be 0.4 and 0.6, or 0.7 and 0.3, and no specific limitation is made.

[0053] It is understood that by adopting the technical solution provided in this embodiment, the most relevant resources to customers can be obtained as candidate resources by comprehensively considering the historical preference score and load score of each resource, thereby improving the rationality of resource allocation.

[0054] Example 7 Based on the above embodiment 1, the present invention also provides another embodiment, which describes the situation where there is historical behavior data for the preset time period under the current customer identifier, but the pre-allocated resources queried in the pre-allocated result storage area are unavailable.

[0055] In this embodiment of the invention, if historical behavior data for the preset time period exists under the current customer identifier, but the pre-allocated resources queried in the pre-allocated result storage area are unavailable, a multi-level degradation allocation strategy is executed; the multi-level degradation allocation strategy includes: Obtain historical allocation records from the historical behavior data under the customer identifier. The historical allocation records include at least the most recently successfully allocated resource and the allocation time for the customer. Determine whether the most recently successfully allocated resource is available. If it is available, use the resource as the target resource. If the most recently successfully allocated resource is unavailable, then based on the customer's historical scanning records in the customer's historical behavior data, obtain the resource that the customer scanned the most times in the most recent second preset time window, and determine whether the resource is available. If the resource is available, then use the resource as the target resource. If the resource with the most scans by the customer within the most recent second preset time window is unavailable, a dynamic selection sub-strategy will be executed; the dynamic selection sub-strategy includes: Obtain the status information of all currently available resources, including at least the real-time load rate of each resource, the number of currently queued customers, and the average service duration; Based on the historical allocation records in the customer's historical behavior data, calculate the historical association strength S_i between the customer and each available resource i. The historical association strength S_i includes at least one or more of the following: the number of times the customer has been allocated to resource i in history, the number of times the customer has scanned resource i in history, and the number of times the customer has completed a call through resource i in history. For each available resource i, calculate its comprehensive score. The calculation rule for the comprehensive score is as follows: R_i=λ·(1-L_i)+μ·(1-Q_i / Q_max)+ν·(S_i / S_max), Where L_i is the real-time load rate of resource i, Q_i is the current number of queued customers of resource i, Q_max is the preset maximum queuing threshold, S_max is the maximum historical association strength among all resources, and λ, μ, and ν are preset weight coefficients. Select the resource with the highest overall score R_i as the target resource; After the downgrade allocation is completed, the pre-allocated resources for that customer are recalculated, and the newly calculated pre-allocated resources are updated in the pre-allocation result storage area.

[0056] The second preset time window can be set according to the learning progress, and can be up to 1 month.

[0057] Understandably, when the pre-allocated resources in the pre-allocation results happen to be occupied, the allocation is carried out according to the multi-level degradation allocation strategy, which improves the convenience of allocation.

[0058] Example 8 Based on Embodiment 1 above, the present invention also provides another embodiment.

[0059] In this embodiment, after sending the target resource to the client, the process further includes: The parameters of the incoming call time prediction model are updated, and the deviation between the customer's incoming call request initiation time and the predicted incoming call request initiation time is used as a feedback signal to adjust the prediction accuracy of the model.

[0060] It is understandable that by adopting the technical solution provided in this embodiment, the prediction accuracy of the model is continuously adjusted and updated by using time deviation as a feedback signal, thereby further improving the accuracy and convenience of resource allocation.

[0061] Example 9 Based on a general inventive concept, the present invention also provides an intelligent allocation device for customer incoming line resources, used to implement the above-described method embodiments. Figure 2 This is a schematic diagram of the structure of an intelligent allocation device for customer incoming line resources according to an embodiment of the present invention, as shown below. Figure 2 As shown, the apparatus provided in this embodiment may include the following structure: The acquisition module 21 is used to obtain the customer identifier of the current customer in response to a customer call request; wherein, the customer identifier of each customer is unique; The judgment module 22 is used to determine whether there is historical behavior data for a preset time period under the current customer identifier; wherein, the historical behavior data is the behavior record associated with the customer identifier and stored before and after each customer completes the call process; The allocation module 23 is used to allocate resources to the current customer according to the real-time allocation rules and obtain the target resources when there is no historical behavior data for the preset time period under the current customer identifier. The retrieval module 24 is used to, when historical behavior data for the preset time period exists under the current customer identifier, query the corresponding pre-allocated resources from the pre-allocated result storage area according to the customer identifier, and use the queried pre-allocated resources as the target resources for the current customer; wherein, the pre-allocated resources in the pre-allocated result storage area are stored after the time of the customer's future call request is predicted in advance according to the call time prediction model, and the resources are allocated in advance according to the prediction results; the call time prediction model is trained and constructed based on the customer's historical behavior data; The sending module 25 is used to send the target resource to the client.

[0062] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0063] Example 10 Based on a general inventive concept, embodiments of the present invention also provide an intelligent allocation device for customer incoming line resources.

[0064] The present invention also provides an intelligent allocation device for customer incoming line resources, used to implement the above method embodiments. Figure 3 This is a schematic diagram of the structure of an intelligent allocation device for customer incoming line resources according to an embodiment of the present invention, as shown below. Figure 3 As shown, the intelligent allocation device for customer incoming line resources in this embodiment includes a processor 31 and a memory 32, with the processor connected to the memory. The processor is used to call and execute a program stored in the memory; the memory is used to store the program, which is at least used to execute the intelligent allocation method for customer incoming line resources in the above embodiments.

[0065] The specific implementation scheme of the intelligent allocation device for customer incoming line resources provided in this application embodiment can refer to the implementation scheme of the intelligent allocation method for customer incoming line resources in any of the above embodiments, and will not be repeated here.

[0066] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0067] It should be noted that in the description of this invention, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this invention, unless otherwise stated, "a plurality of" means at least two.

[0068] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0069] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0070] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it includes one or a combination of the steps of the method embodiments.

[0071] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0072] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0073] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0074] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for intelligent allocation of customer inbound resources, characterized in that, include: In response to a customer's incoming call request, the customer identifier of the current customer is obtained from the incoming call request; wherein, the customer identifier of each customer is unique; Determine whether there is historical behavior data for a preset time period under the current customer identifier; wherein, the historical behavior data is the behavior record associated with the customer identifier and stored before and after each customer completes the call process; If no historical behavior data for the preset time period exists under the current customer identifier, then resources are allocated to the current customer according to the real-time allocation rules to obtain the target resources; If historical behavior data for the preset time period exists under the current customer identifier, then based on the customer identifier, the corresponding pre-allocated resources are queried from the pre-allocation result storage area, and the queried pre-allocated resources are used as the target resources for the current customer; wherein, the pre-allocated resources in the pre-allocation result storage area are stored after the time of the customer's future call request is predicted in advance based on the call time prediction model, and resources are allocated in advance based on the prediction results; the call time prediction model is trained and constructed based on the customer's historical behavior data; Send the target resource to the client.

2. The method according to claim 1, characterized in that, The determination of whether historical behavior data for a preset time period exists under the current customer identifier includes: If there is no historical behavior data under the current customer identifier, or if the current customer's current call time is more than a preset period apart from the last call time, then it is determined that there is no historical behavior data for the preset time period under the current customer identifier; wherein, the current customer's current call time is determined based on the customer's call request. If there is historical behavior data under the current customer identifier, and the time between the current customer's last call and the time between the last call does not exceed a preset period, then it is determined that there is historical behavior data for a preset time period under the current customer identifier.

3. The method according to claim 1, characterized in that, The step of allocating resources to the current customer according to real-time allocation rules and obtaining target resources includes: The incoming call request includes an incoming call mode, which includes a one-to-one friend adding mode and a community joining mode; According to the incoming line mode, a target resource is allocated from the corresponding candidate resource pool in a round-robin order; The candidate resource pool for the one-to-one friend-adding mode consists of currently online and available employee personal QR codes or customer acquisition links; the candidate resource pool for the community joining mode consists of currently active group QR codes that are not yet full.

4. The method according to claim 1, characterized in that, The method for constructing the incoming line time prediction model includes: Collect customers’ historical behavior data, which includes at least the customer’s call time, QR code scanning time and scanning channel; Using the historical behavior data as training samples, an independent call-in time prediction model is constructed for each customer with historical behavior data. The model outputs the probability value of the customer initiating a call-in request within a first preset time window; wherein, the customer's QR code scan corresponds one-to-one with the call-in request.

5. The method according to claim 4, characterized in that, The method for constructing the pre-allocated result storage area includes: The probability of a customer initiating an incoming call request within a first preset time window is predicted based on the aforementioned time prediction model. When it is predicted that the probability of a customer initiating an incoming call request in the next time window exceeds a preset threshold, a candidate resource is locked for the customer before the next time window. The locked candidate resource is used as a pre-allocated resource, and the locking relationship is stored in the pre-allocation result storage area as a pre-allocation result. The candidate resource is determined based on the customer's historical allocation records and resource load status. The update methods for the pre-allocation result storage area include: recalculating the pre-allocated resources of all or part of the customers in batches according to a fixed time period, and / or triggering the recalculation of the pre-allocated resources of any customer when the historical behavior data of any customer is updated.

6. The method according to claim 5, characterized in that, The candidate resources are determined based on the customer's historical allocation records and resource load status, including: Obtain the number of times each resource has been allocated in the customer's historical allocation records, and calculate the historical preference score for each resource; Get the real-time load rate of each resource and calculate the load score for each resource; For each resource, calculate the overall score = α × historical preference score + β × (1 - real-time load rate), where α and β are preset weighting coefficients; The resource with the highest overall score will be selected as the candidate resource for this customer.

7. The method according to claim 1, characterized in that, If historical behavior data for the preset time period exists under the current customer identifier, and the pre-allocated resources queried in the pre-allocated result storage area are unavailable, then a multi-level degradation allocation strategy is executed. The multi-level degradation allocation strategy includes: Obtain historical allocation records from the historical behavior data under the customer identifier. The historical allocation records include at least the most recently successfully allocated resource and the allocation time for the customer. Determine whether the most recently successfully allocated resource is available. If it is available, use the resource as the target resource. If the most recently successfully allocated resource is unavailable, then based on the customer's historical scanning records in the customer's historical behavior data, obtain the resource that the customer scanned the most times in the most recent second preset time window, and determine whether the resource is available. If the resource is available, then use the resource as the target resource. If the resource with the most scans by the customer within the most recent second preset time window is unavailable, a dynamic selection sub-strategy will be executed; the dynamic selection sub-strategy includes: Obtain the status information of all currently available resources, including at least the real-time load rate of each resource, the number of currently queued customers, and the average service duration; Based on the historical allocation records in the customer's historical behavior data, calculate the historical association strength S_i between the customer and each available resource i. The historical association strength S_i includes at least one or more of the following: the number of times the customer has been allocated to resource i in history, the number of times the customer has scanned resource i in history, and the number of times the customer has completed a call through resource i in history. For each available resource i, calculate its comprehensive score. The calculation rule for the comprehensive score is as follows: R_i=λ·(1-L_i)+μ·(1-Q_i / Q_max)+ν·(S_i / S_max), Where L_i is the real-time load rate of resource i, Q_i is the current number of queued customers of resource i, Q_max is the preset maximum queuing threshold, S_max is the maximum historical association strength among all resources, and λ, μ, and ν are preset weight coefficients. Select the resource with the highest overall score R_i as the target resource; After the downgrade allocation is completed, the pre-allocated resources for that customer are recalculated, and the newly calculated pre-allocated resources are updated in the pre-allocation result storage area.

8. The method according to claim 1, characterized in that, After sending the target resource to the client, the process further includes: The parameters of the incoming call time prediction model are updated, and the deviation between the customer's incoming call request initiation time and the predicted incoming call request initiation time is used as a feedback signal to adjust the prediction accuracy of the model.

9. An intelligent allocation device for customer inbound resources, characterized in that, include: The acquisition module is used to obtain the customer identifier of the current customer in response to a customer's incoming call request; wherein, the customer identifier of each customer is unique; The judgment module is used to determine whether there is historical behavior data for a preset time period under the current customer identifier; wherein, the historical behavior data is the behavior record associated with the customer identifier and stored before and after each customer completes the call process; The allocation module is used to allocate resources to the current customer according to real-time allocation rules and obtain target resources when there is no historical behavior data for the preset time period under the current customer identifier. The retrieval module is used to, when historical behavior data for the preset time period exists under the current customer identifier, query the corresponding pre-allocated resources from the pre-allocated result storage area based on the customer identifier, and use the queried pre-allocated resources as the target resources for the current customer; wherein, the pre-allocated resources in the pre-allocated result storage area are stored after the time of the customer's future call request is predicted in advance based on the call time prediction model, and resources are allocated in advance based on the prediction results; the call time prediction model is trained and constructed based on the customer's historical behavior data; The sending module is used to send the target resource to the client.

10. An intelligent allocation device for customer incoming line resources, characterized in that, It includes a processor and a memory, wherein the processor is connected to the memory: The processor is used to call and execute the program stored in the memory; The memory is used to store the program, which is at least used to execute the intelligent allocation method for customer incoming line resources as described in any one of claims 1-8.

Images