Customer arrival opportunity recommendation method and device, equipment and storage medium

Abstract

The invention belongs to the technical field of customer portraits, and discloses a customer arrival opportunity recommendation method and device, equipment and a storage medium. The first connection probability corresponding to each time period is determined according to the historical dialing record code, the second connection probability corresponding to each time period is determined according to the historical behavior time code of the to-be-recommended customer, and the third connection probability corresponding to each time period is determined according to the customer information corresponding to the to-be-recommended customer; and according to at least one of the first connection probability, the second connection probability and the third connection probability, determining the arrival opportunity corresponding to the to-be-recommended customer. According to the method, the connection probability of the to-be-recommended customer in each time period is jointly determined in combination with the historical dialing record code, the historical behavior time code of the to-be-recommended customer and the customer information, and then the arrival time corresponding to the to-be-recommended customer is determined according to the connection probability, so that the accuracy of arrival time recommendation is improved when the to-be-recommended customer is a new customer.

Description

Technical Field

[0001] This application relates to the field of customer profiling technology, and in particular to a method, apparatus, device, and storage medium for recommending customer contact timing. Background Technology

[0002] With the continuous development and popularization of information technology, customer profiling technology has become an indispensable part of the marketing and sales strategies of many enterprises. By analyzing and mining customer data, we can better understand customer preferences, needs, and behavioral characteristics, thereby providing more accurate and targeted guidance for enterprise marketing and sales activities. Outreach timing recommendation is one application of customer profiling technology in marketing and sales activities. By analyzing and mining customer profile data, the optimal outreach timing can be recommended, thereby improving the effectiveness and conversion rate of outbound calls. With the development of artificial intelligence and big data technologies, the outbound calling market has also begun to gradually introduce model algorithms to improve the efficiency and effectiveness of outbound calls. However, most current outreach timing recommendation systems rely on a single data source, such as outbound call records or user behavior logs. This single-source modeling approach ignores the complementarity and synergistic effect between multi-source heterogeneous data, especially in the context of new customers, where the accuracy of outreach timing recommendation drops significantly. Summary of the Invention

[0003] The main objective of this application is to provide a method, apparatus, device, and storage medium for recommending customer outreach timing, aiming to solve the technical problem of how to improve the accuracy of new customer outreach timing recommendations.

[0004] To achieve the above objectives, this application provides a customer outreach timing recommendation method, which includes the following steps: The first connection probability for each time period is determined based on the historical call record codes. The second connection probability for each time period is determined based on the historical behavior time code of the customer to be recommended. The third connection probability for each time period is determined based on the customer information corresponding to the customer to be recommended. The timing for reaching the customer to be recommended is determined based on at least one of the first connection probability, the second connection probability, and the third connection probability.

[0005] Optionally, determining the second connection probability corresponding to each time period based on the historical behavior time code of the customer to be recommended includes: Determine the historical behavior time code of the customer to be recommended, the historical behavior time code includes: the time code for receiving the activity coupon and the time code for the operation behavior; The time code for receiving the activity coupon and the time code for the operation behavior are concatenated to obtain a time vector; The time vector is input into a preset timing self-adjustment model to obtain the second connection probability corresponding to each time period.

[0006] Optionally, the step of inputting the time vector into a preset timing self-adjustment model to obtain the second connection probability corresponding to each time period includes: The time vector is input into a preset timing self-adjustment model, and the time vector is converted into a time matrix through the embedding mechanism in the preset timing self-adjustment model; The importance of each time period is determined by the self-adjustment mechanism in the preset timing self-adjustment model. The second connection probability for each time period is determined by the forward feedback network in the preset timing self-adjustment model and the importance.

[0007] Optionally, determining the third connection probability for each time period based on the customer information corresponding to the customer to be recommended includes: Based on the customer information corresponding to the customer to be recommended, identify associated customers and obtain the associated customer characteristics corresponding to the associated customers; Based on the associated customer characteristics, determine the characteristics of the customer to be recommended corresponding to the customer to be recommended; Based on the characteristics of the customer to be recommended, select the target customer corresponding to the customer to be recommended from the associated customers; Determine the time distribution of the target customer's target historical call records, and determine the third call probability of the customer to be recommended in each time period based on the time distribution.

[0008] Optionally, determining the characteristics of the customer to be recommended corresponding to the associated customer characteristics includes: For any two customers among the associated customers and the customers to be recommended, calculate the Mahalanobis distance between the two customers based on the characteristics of the associated customers and the characteristics of the customers to be recommended; Cluster the associated customers and the customers to be recommended based on the Mahalanobis distance to obtain the clustering results; Based on the clustering results, the target customer corresponding to the customer to be recommended is selected from the associated customers.

[0009] Optionally, determining the first connection probability corresponding to each time period based on the historical call record code includes: Determine the default connection probability for each time period based on the number of time periods; Determine whether there is a historical call record corresponding to the historical call record code in each time period, and obtain the determination result; Based on the judgment result, the default connection probability is adjusted to obtain the first connection probability corresponding to each time period.

[0010] Optionally, determining the contact timing for the customer to be recommended based on at least one of the first connection probability, the second connection probability, and the third connection probability includes: The first connection probability, the second connection probability, and the third connection probability are weighted and summed to obtain the predicted connection probability for each time period, and a connection probability set is constructed. Determine the maximum connection probability in the set of connection probabilities; The time period corresponding to the maximum connection probability is used as the contact timing for the customer to be recommended.

[0011] Furthermore, to achieve the above objectives, this application also provides a customer outreach timing recommendation device, the customer outreach timing recommendation device comprising: The connection probability determination module is used to determine the first connection probability for each time period based on the encoding of historical call records. The connection probability determination module is also used to determine the second connection probability corresponding to each time period based on the historical behavior time code of the customer to be recommended. The connection probability determination module is also used to determine the third connection probability corresponding to each time period based on the customer information corresponding to the customer to be recommended. The contact timing recommendation module is used to determine the contact timing corresponding to the customer to be recommended based on at least one of the first connection probability, the second connection probability, and the third connection probability.

[0012] In addition, to achieve the above objectives, this application also proposes a customer outreach timing recommendation device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the customer outreach timing recommendation method as described above.

[0013] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the customer outreach timing recommendation method as described above.

[0014] This application determines a first connection probability for each time period based on historical call record codes, a second connection probability for each time period based on the historical behavior time codes of the customer to be recommended, and a third connection probability for each time period based on the customer information of the customer to be recommended. Then, it determines the contact timing for the customer to be recommended based on at least one of the first, second, and third connection probabilities. This application combines historical call record codes, the historical behavior time codes of the customer to be recommended, and customer information to jointly determine the connection probability of the customer to be recommended in each time period, and then determines the contact timing for the customer to be recommended based on the connection probability, thereby improving the accuracy of contact timing recommendations when the customer to be recommended is a new customer. Attached Figure Description

[0015] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating the first embodiment of the method for recommending customer outreach timing in this application; Figure 2 A flowchart illustrating the second embodiment of the method for recommending customer outreach timing in this application; Figure 3 A flowchart illustrating the third embodiment of the method for recommending customer outreach timing in this application; Figure 4 A schematic diagram of the overall process for recommending an embodiment of the timing of customer outreach in this application; Figure 5 A structural block diagram of the first embodiment of the customer outreach timing recommendation device for this application; Figure 6 This is a schematic diagram of the structure of the customer outreach timing recommendation device in the hardware operating environment involved in the embodiments of this application.

[0018] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0020] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0021] The main solution of this application embodiment is: to determine the first connection probability corresponding to each time period based on the historical call record code; to determine the second connection probability corresponding to each time period based on the historical behavior time code of the customer to be recommended; to determine the third connection probability corresponding to each time period based on the customer information corresponding to the customer to be recommended; and to determine the contact timing corresponding to the customer to be recommended based on at least one of the first connection probability, the second connection probability, and the third connection probability.

[0022] With the continuous development and popularization of information technology, customer profiling technology has become an indispensable part of the marketing and sales strategies of many enterprises. By analyzing and mining customer data, we can better understand customer preferences, needs, and behavioral characteristics, thereby providing more accurate and targeted guidance for enterprise marketing and sales activities. Outreach timing recommendation is one application of customer profiling technology in marketing and sales activities. By analyzing and mining customer profile data, the optimal outreach timing can be recommended, thereby improving the effectiveness and conversion rate of outbound calls. With the development of artificial intelligence and big data technologies, the outbound calling market has also begun to gradually introduce model algorithms to improve the efficiency and effectiveness of outbound calls. However, most current outreach timing recommendation systems rely on a single data source, such as outbound call records or user behavior logs. This single-source modeling approach ignores the complementarity and synergistic effect between multi-source heterogeneous data, especially in the context of new customers, where the accuracy of outreach timing recommendation drops significantly.

[0023] This application determines a first connection probability for each time period based on historical call record codes, a second connection probability for each time period based on the historical behavior time codes of the customer to be recommended, and a third connection probability for each time period based on the customer information of the customer to be recommended. Then, it determines the contact timing for the customer to be recommended based on at least one of the first, second, and third connection probabilities. This application combines historical call record codes, the historical behavior time codes of the customer to be recommended, and customer information to jointly determine the connection probability of the customer to be recommended in each time period, and then determines the contact timing for the customer to be recommended based on the connection probability, thereby improving the accuracy of contact timing recommendations when the customer to be recommended is a new customer.

[0024] It should be noted that the executing entity of this application can be a computing service device with data processing, network communication and program execution functions, such as a computer.

[0025] Based on this, embodiments of this application provide a method for recommending customer outreach timing, referring to... Figure 1 , Figure 1A flowchart illustrating the first embodiment of the method for recommending customer outreach timing for this application.

[0026] In this embodiment, the customer outreach timing recommendation method includes the following steps: Step S10: Determine the first connection probability for each time period based on the historical call record codes.

[0027] Understandably, historical call records can be retrieved based on accessible time ranges, such as 8:00-12:00 and 14:00-20:00 daily. This could involve retrieving call records from the past year and encoding each record. In one feasible embodiment, this can be done in 15-minute increments, followed by sequential encoding. For example, 8:00-8:15 is encoded as 0, 8:15-8:30 as 1, 8:30-8:45 as 2, and so on. If a historical call record is for 8:10, 8:20, 8:25, or 8:35, it would be encoded as 0112. Historical call records outside of the 8:00-12:00 and 14:00-20:00 time ranges can be directly deleted.

[0028] In practice, the first connection probability for each time period can be determined based on the historical call record encoding. The time period can be the same as the time period division mentioned above, that is, divided according to a 15-minute granularity, and each time period can correspond to a first connection probability.

[0029] Furthermore, in this embodiment, in order to accurately obtain the first connection probability corresponding to each time period, step S10 includes: determining the default connection probability corresponding to each time period based on the number of time periods; determining whether there is a historical call record corresponding to the historical call record code in each time period, and obtaining a determination result; adjusting the default connection probability according to the determination result to obtain the first connection probability corresponding to each time period.

[0030] It should be understood that the number of time periods k can be determined first. For example, the time periods 8:00-12:00 and 14:00-20:00 can be divided into granular periods of 15 minutes each. The number of time periods k is 40, and the default connection probability for each time period is 1 / k, that is, 1 / 40.

[0031] In the implementation, if there are no unanswered calls within a certain time period, the first connection probability for that time period is adjusted to 0; if there are only answered calls within a certain time period, the first connection probability for that time period is adjusted to 2 / k. For example, if there are no answered calls between 8:00 and 8:15, the first connection probability for that time period can be adjusted to 0. Finally, the first connection probabilities for each time period can be normalized so that the sum of the processed first connection probabilities is 1.

[0032] Step S20: Determine the second connection probability for each time period based on the historical behavior time code of the customer to be recommended.

[0033] Understandably, the customers to be referred can be new customers, i.e., customers who have never initiated marketing or sales activities. Historical behavior time can include the time when the customer to be referred received the promotional coupon, and the time when the customer to be referred interacted with the app, mini-program, and webpage. Furthermore, the historical behavior time is encoded using the same method as the historical call records described above, and this encoding is used to determine the second connection probability.

[0034] Step S30: Determine the third connection probability for each time period based on the customer information corresponding to the customer to be recommended.

[0035] It should be understood that the customer information corresponding to the customer to be recommended may include a mobile phone number. In one feasible embodiment, the third connection probability corresponding to each time period can be determined based on the mobile phone number of the customer to be recommended and the mobile phone number of the associated customer.

[0036] Step S40: Determine the contact timing for the customer to be recommended based on at least one of the first connection probability, the second connection probability, and the third connection probability.

[0037] Understandably, the final connection probability can be obtained based on at least one of the first connection probability, the second connection probability, and the third connection probability. Specifically, the first connection probability can be used as the final connection probability; the second connection probability can also be used as the final connection probability; the third connection probability can also be used as the final connection probability; the first and second connection probabilities can be weighted and summed to obtain the final connection probability; the first and third connection probabilities can be weighted and summed to obtain the final connection probability; the second and third connection probabilities can also be weighted and summed to obtain the final connection probability; and the first, second, and third connection probabilities can also be weighted and summed to obtain the final connection probability. The time period with the highest final connection probability is then used as the contact timing for the customer to be recommended.

[0038] Furthermore, in order to accurately determine the timing of the outreach, in this embodiment, step S40 includes: performing a weighted summation of the first connection probability, the second connection probability, and the third connection probability to obtain the predicted connection probability corresponding to each time period, and constructing a set of connection probabilities; determining the maximum connection probability in the set of connection probabilities; and using the time period corresponding to the maximum connection probability as the outreach timing for the customer to be recommended.

[0039] It should be understood that, in order to improve the accuracy of the contact timing recommendation, this embodiment can combine the first connection probability, the second connection probability, and the third connection probability to obtain the predicted connection probability for each time period. In a feasible embodiment, if the first connection probability Y1, the second connection probability Y2, and the third connection probability Y3 all exist, the predicted connection probability Y = αY1 + βY2 + γY3, where α + β + γ = 1; if the first connection probability Y1 and the second connection probability Y2 exist, but the third connection probability Y3 does not exist, the predicted connection probability Y = (αY1 + βY2) / (α + β); if the first connection probability Y1 and the third connection probability Y3 exist, but the second connection probability Y2 does not exist, the predicted connection probability Y = (αY1 + γY3) / (α + γ); when the first connection probability Y1 exists, but the second connection probability Y2 and the third connection probability Y3 do not exist, it means that there is no data for the customer to be recommended, and Y = Y1. The above weights are set in advance based on experience, and this embodiment does not limit the specific values.

[0040] Understandably, the first connection probability Y1 is calculated based on historical call records for each time period. Even without historical connection records, the first connection probability is assumed to be the same for each time period, therefore, the first connection probability Y1 is guaranteed to exist. The existence of the second connection probability Y2 depends on whether the customer to be recommended has received an activity coupon or operated on the APP, mini-program, or website. If not, Y2 does not exist. The existence of the third connection probability Y3 depends on whether the customer's phone number is associated with any other customer. If the customer's phone number has never been registered in internal data (e.g., as an emergency contact or spouse's phone number), then Y3 does not exist.

[0041] In practical implementation, a set of connection probabilities can be constructed based on the predicted connection probabilities corresponding to all time periods, and the maximum connection probability in the set can be determined. The time period corresponding to the maximum connection probability is then used as the contact timing for the customer to be recommended.

[0042] This embodiment determines the first connection probability for each time period based on historical call record codes, the second connection probability for each time period based on the historical behavior time codes of the customer to be recommended, and the third connection probability for each time period based on the customer information of the customer to be recommended. Then, it determines the contact timing for the customer to be recommended based on at least one of the first, second, and third connection probabilities. This embodiment combines historical call record codes, the historical behavior time codes of the customer to be recommended, and customer information to jointly determine the connection probability of the customer to be recommended in each time period, and then determines the contact timing for the customer to be recommended based on the connection probability, thereby improving the accuracy of contact timing recommendations when the customer to be recommended is a new customer.

[0043] refer to Figure 2, Figure 2 A flowchart illustrating the second embodiment of the method for recommending customer outreach timing for this application.

[0044] Based on the first embodiment described above, in this embodiment, step S20 includes: Step S201: Determine the historical behavior time code of the customer to be recommended. The historical behavior time code includes: the time code for receiving the activity coupon and the time code for the operation behavior.

[0045] Understandably, several behaviors can be selected from all historical behaviors based on the time codes of all the customers to be recommended. Specifically, the most recent m times when coupons were claimed and their corresponding codes can be selected, and the most recent n times when operations were performed on the APP, mini-program, and webpage and their corresponding codes can be selected. The selected n times when operations were performed on the APP, mini-program, and webpage are not on the same day.

[0046] Step S202: Concatenate the time code for receiving the activity coupon and the time code for the operation behavior to obtain a time vector.

[0047] It should be understood that the time codes for receiving activity coupons and the time codes for operation behaviors can be concatenated to obtain a time vector H. In particular, if the concatenated time vector H has fewer than m+n elements, zeros are added to the end to make it m+n.

[0048] Step S203: Input the time vector into the preset timing self-adjustment model to obtain the second connection probability corresponding to each time period.

[0049] Understandably, the preset timing self-adjustment model can be a pre-set convolutional neural network model. By inputting a time vector of length m+n into the preset timing self-adjustment model, the second connection probability corresponding to each time period can be obtained.

[0050] Furthermore, in order to accurately obtain the second connection probability corresponding to each time period, in this embodiment, step S203 includes: inputting the time vector into a preset timing self-adjustment model, and converting the time vector into a time matrix through the embedding mechanism in the preset timing self-adjustment model; determining the importance of each time period through the self-adjustment mechanism in the preset timing self-adjustment model; and determining the second connection probability corresponding to each time period through the forward feedback network in the preset timing self-adjustment model and the importance.

[0051] It should be understood that after inputting the time vector into the preset timing self-adjusting model, the numbered time vector H can be converted into a time matrix through the embedding mechanism in the model. ,Right now =embedding(H).

[0052] Understandably, the importance of m+n time periods can be learned through a self-adjusting mechanism, i.e.

[0053] In the formula, , , , are the parameters used for model training, and softmax is the activation function of the model.

[0054] In practical implementation, the second connection probability Y2 corresponding to each time period can be determined through a feedforward network and importance. , , This refers to the bias of the linear layer, and SELU is the activation function of the model. The core of the pre-set timing self-adjusting model lies in its self-adjusting mechanism, which allows the model to learn the importance of m+n time periods and train the optimal weight ratio based on a large number of samples. Moreover, it has been verified that the importance ranking of the m+n time periods is highly correlated with the difference from the current time.

[0055] This embodiment determines the historical behavior time codes of the customers to be recommended. These historical behavior time codes include: the time code for claiming promotional coupons and the time code for performing actions. The time codes for claiming promotional coupons and performing actions are then concatenated to obtain a time vector. This time vector is then input into a preset timing self-adjustment model to obtain the second connection probability corresponding to each time period. This embodiment first concatenates the time codes for claiming promotional coupons and performing actions, and then inputs the time vector into the preset timing self-adjustment model, which can accurately obtain the second connection probability corresponding to each time period.

[0056] refer to Figure 3 , Figure 3 A flowchart illustrating the third embodiment of the method for recommending customer outreach timing in this application.

[0057] Based on the above embodiments, in this embodiment, step S30 includes: Step S301: Determine associated customers based on the customer information corresponding to the customer to be recommended, and obtain the associated customer characteristics corresponding to the associated customers.

[0058] Understandably, associated customers can be determined based on the customer information corresponding to the customer to be recommended. In one feasible embodiment, the customer information may include the mobile phone number of the customer to be recommended, and the associated customers may be customers already stored in the system that are related to the customer to be recommended, such as the emergency contact person or spouse of the customer to be recommended.

[0059] It should be understood that the characteristics of the associated customers that can be obtained may include the age of all associated customers, the distribution of historical connection time periods, the distribution of historical operation time periods, the proportion of occupations, the gender ratio, and other characteristics.

[0060] Step S302: Determine the characteristics of the customer to be recommended corresponding to the customer to be recommended based on the associated customer characteristics.

[0061] Understandably, the characteristics of the customers to be recommended can be determined based on the characteristics of the associated customers. For example, the average age of all associated customers of the customer to be recommended, the historical connection time period with the highest probability, and the historical operation time period with the highest probability.

[0062] Step S303: Select the target customer corresponding to the customer to be recommended from the associated customers based on the characteristics of the customer to be recommended.

[0063] Furthermore, in order to accurately select the target customer corresponding to the customer to be recommended, in this embodiment, step S303 includes: for any two customers among the associated customers and the customers to be recommended, calculating the Mahalanobis distance between the two customers based on the characteristics of the associated customers and the characteristics of the customers to be recommended; clustering the associated customers and the customers to be recommended based on the Mahalanobis distance to obtain clustering results; and selecting the target customer corresponding to the customer to be recommended from the associated customers based on the clustering results.

[0064] It should be understood that for any two customers among all associated customers and all potential customers, the Mahalanobis distance between them can be calculated based on the characteristics of the associated customers and the potential customers. A and B represent the feature values ​​of any two customers, which can be obtained by feature quantization, and ∑ is the covariance matrix of the features of the related customers.

[0065] Understandably, related customers and customers to be recommended can be clustered based on Mahalanobis distance. K-Medoids clustering can be used to group customers with close Mahalanobis distances into the same cluster. After clustering, if another customer appears, we can determine which cluster's Mahalanobis distance the customer's characteristics are closest to, and then that customer belongs to that cluster.

[0066] In practice, the cluster to which the customer to be recommended belongs can be determined first, and all related customers in that cluster can be used as target customers.

[0067] Step S304: Determine the time distribution of the target historical call records corresponding to the target customer, and determine the third call probability of the customer to be recommended in each time period based on the time distribution.

[0068] Understandably, it is possible to determine all historical call connection records and historical non-connection records corresponding to the target customer. The time distribution can be the call connection status in each time period. Based on the call connection status, the call connection probability of the target customer in each time period can be calculated. For each time period, the call connection probability can be the number of calls in that time period divided by the total number of calls. The total number of calls can be the sum of the number of calls connected and the number of calls not connected. After obtaining the call connection probability for each time period, the call connection probability can be normalized, and the normalized probability can be used as the third call connection probability of the customer to be recommended in each time period.

[0069] In the specific implementation, refer to Figure 4 , Figure 4 A schematic diagram of the overall process for recommending an embodiment of the customer outreach timing for this application is shown below. Figure 4 As shown, the system can first obtain the first connection probability for each time period based on historical call records and with the help of rules. Then, based on the time when new customers receive promotional coupons and the time when new customers perform actions, the system can obtain the second connection probability for each time period through a preset timing self-adjustment model. Furthermore, based on the new customer's mobile phone number, the system can obtain the third connection probability for each time period through clustering. Finally, the first, second, and third connection probabilities for each time period are weighted and summed to obtain the predicted connection probability for each time period. This probability can then be used to predict the timing of reaching new customers.

[0070] This embodiment determines associated customers based on the customer information corresponding to the customer to be recommended, and obtains the associated customer characteristics corresponding to the associated customers. Then, based on the associated customer characteristics, it determines the characteristics of the customer to be recommended. Next, based on the characteristics of the customer to be recommended, it selects the target customer corresponding to the customer to be recommended from the associated customers. Then, it determines the time distribution of the target customer's historical call records, and determines the third call connection probability of the customer to be recommended in each time period based on the time distribution. This embodiment selects the target customer corresponding to the customer to be recommended from the associated customers based on the characteristics of the customer to be recommended, and then accurately obtains the third call connection probability of the customer to be recommended in each time period based on the call connection status of the target customer.

[0071] Reference Figure 5 , Figure 5 A structural block diagram of the first embodiment of the device for recommending customer outreach timing in this application.

[0072] like Figure 5 As shown, the customer outreach timing recommendation device proposed in this application includes: The connection probability determination module 10 is used to determine the first connection probability corresponding to each time period based on the historical dialing record encoding. The connection probability determination module 10 is also used to determine the second connection probability corresponding to each time period based on the historical behavior time code of the customer to be recommended. The connection probability determination module 10 is also used to determine the third connection probability corresponding to each time period based on the customer information corresponding to the customer to be recommended. The contact timing recommendation module 20 is used to determine the contact timing corresponding to the customer to be recommended based on at least one of the first connection probability, the second connection probability, and the third connection probability.

[0073] This embodiment determines the first connection probability for each time period based on historical call record codes, the second connection probability for each time period based on the historical behavior time codes of the customer to be recommended, and the third connection probability for each time period based on the customer information of the customer to be recommended. Then, it determines the contact timing for the customer to be recommended based on at least one of the first, second, and third connection probabilities. This embodiment combines historical call record codes, the historical behavior time codes of the customer to be recommended, and customer information to jointly determine the connection probability of the customer to be recommended in each time period, and then determines the contact timing for the customer to be recommended based on the connection probability, thereby improving the accuracy of contact timing recommendations when the customer to be recommended is a new customer.

[0074] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of this application. In practical applications, those skilled in the art can select some or all of it to achieve the purpose of this embodiment according to actual needs, and no restrictions are imposed here.

[0075] In addition, for technical details not described in detail in this embodiment, please refer to the customer outreach timing recommendation method provided in any embodiment of this application, which will not be repeated here.

[0076] Based on the first embodiment of the customer contact timing recommendation device described in this application, a second embodiment of the customer contact timing recommendation device of this application is proposed.

[0077] In this embodiment, the connection probability determination module 10 is further used to determine the historical behavior time code of the customer to be recommended. The historical behavior time code includes: the time code for receiving the activity coupon and the time code for the operation behavior; the time code for receiving the activity coupon and the time code for the operation behavior are concatenated to obtain a time vector; the time vector is input into a preset timing self-adjustment model to obtain the second connection probability corresponding to each time period.

[0078] Furthermore, the connection probability determination module 10 is also used to input the time vector into a preset timing self-adjustment model, and convert the time vector into a time matrix through the embedding mechanism in the preset timing self-adjustment model; determine the importance of each time period through the self-adjustment mechanism in the preset timing self-adjustment model; and determine the second connection probability corresponding to each time period through the forward feedback network in the preset timing self-adjustment model and the importance.

[0079] Furthermore, the connection probability determination module 10 is also used to determine associated customers based on the customer information corresponding to the customer to be recommended, and obtain associated customer characteristics corresponding to the associated customers; determine the characteristics of the customer to be recommended corresponding to the customer to be recommended based on the associated customer characteristics; select the target customer corresponding to the customer to be recommended from the associated customers based on the characteristics of the customer to be recommended; determine the time distribution of the target historical connection records corresponding to the target customer, and determine the third connection probability of the customer to be recommended in each time period based on the time distribution.

[0080] Furthermore, the connection probability determination module 10 is also used to calculate the Mahalanobis distance between any two customers among the associated customers and the customers to be recommended, based on the characteristics of the associated customers and the characteristics of the customers to be recommended; to cluster the associated customers and the customers to be recommended based on the Mahalanobis distance to obtain clustering results; and to select the target customer corresponding to the customer to be recommended from the associated customers based on the clustering results.

[0081] Furthermore, the connection probability determination module 10 is also used to determine the default connection probability corresponding to each time period based on the number of time periods; determine whether there is a historical call record corresponding to the historical call record code in each time period, and obtain a judgment result; adjust the default connection probability according to the judgment result to obtain the first connection probability corresponding to each time period.

[0082] Furthermore, the contact timing recommendation module 20 is also used to perform a weighted summation of the first connection probability, the second connection probability, and the third connection probability to obtain the predicted connection probability corresponding to each time period, and construct a connection probability set; determine the maximum connection probability in the connection probability set; and use the time period corresponding to the maximum connection probability as the contact timing corresponding to the customer to be recommended.

[0083] Other embodiments or specific implementations of the customer reach timing recommendation device in this application can be found in the above-described method embodiments, and will not be repeated here.

[0084] This application provides a customer outreach timing recommendation device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the customer outreach timing recommendation method in the first embodiment described above.

[0085] The following is for reference. Figure 6 The diagram illustrates a structural schematic suitable for implementing a customer outreach timing recommendation device according to embodiments of this application. The customer outreach timing recommendation device in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 6 The customer outreach timing recommendation device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0086] like Figure 6As shown, the customer reach timing recommendation device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the customer reach timing recommendation device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows the customer reach timing recommendation device to communicate wirelessly or wiredly with other devices to exchange data. While the figure shows customer reach timing recommendation devices with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0087] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0088] The customer outreach timing recommendation device provided in this application, employing the customer outreach timing recommendation method described in the above embodiments, can solve the technical problem of how to improve the accuracy of new customer outreach timing recommendations. Compared with the prior art, the beneficial effects of the customer outreach timing recommendation device provided in this application are the same as those of the customer outreach timing recommendation method described in the above embodiments, and other technical features of this customer outreach timing recommendation device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0089] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0090] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0091] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the customer outreach timing recommendation method described in the above embodiments.

[0092] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0093] The aforementioned computer-readable storage medium may be included in the customer outreach timing recommendation device; or it may exist independently and not assembled into the customer outreach timing recommendation device.

[0094] The aforementioned computer-readable storage medium carries one or more programs that, when executed by a customer contact timing recommendation device, cause the customer contact timing recommendation device to: determine a first connection probability corresponding to each time period based on historical call record codes; determine a second connection probability corresponding to each time period based on historical behavior time codes of the customer to be recommended; determine a third connection probability corresponding to each time period based on customer information corresponding to the customer to be recommended; and determine the contact timing corresponding to the customer to be recommended based on at least one of the first connection probability, the second connection probability, and the third connection probability.

[0095] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Python, Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0096] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0097] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0098] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described customer outreach timing recommendation method, thereby solving the technical problem of how to improve the accuracy of new customer outreach timing recommendations. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the customer outreach timing recommendation method provided in the above embodiments, and will not be repeated here.

[0099] The above description is only a part of the embodiments of this application and does not limit the scope of protection of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the scope of protection of this application.

Claims

1. A customer touch opportunity recommendation method, characterized by, The customer touch opportunity recommendation method comprises the following steps: According to the historical call record coding, the first connection probability corresponding to each time period is determined; According to the historical behavior time coding of the to-be-recommended customer, the second connection probability corresponding to each time period is determined; According to the customer information corresponding to the to-be-recommended customer, the third connection probability corresponding to each time period is determined; According to at least one of the first connection probability, the second connection probability and the third connection probability, the touch opportunity corresponding to the to-be-recommended customer is determined.

2. The customer touch opportunity recommendation method of claim 1, wherein, According to the historical behavior time coding of the to-be-recommended customer, the second connection probability corresponding to each time period is determined, comprising: The historical behavior time coding of the to-be-recommended customer is determined, and the historical behavior time coding comprises: activity coupon collection time coding and operation behavior time coding; The activity coupon collection time coding and the operation behavior time coding are spliced to obtain a time vector; The time vector is input into a preset opportunity self-adjusting model to obtain the second connection probability corresponding to each time period.

3. The customer touch opportunity recommendation method of claim 2, wherein, The time vector is input into a preset opportunity self-adjusting model, and the time vector is converted into a time matrix through an embedding mechanism in the preset opportunity self-adjusting model; The importance of each time period is determined through a self-adjusting mechanism in the preset opportunity self-adjusting model; The second connection probability corresponding to each time period is determined through a forward feedback network in the preset opportunity self-adjusting model and the importance. According to the customer information corresponding to the to-be-recommended customer, the third connection probability corresponding to each time period is determined, comprising:

4. The customer touch opportunity recommendation method of claim 1, wherein, According to the customer information corresponding to the to-be-recommended customer, an associated customer is determined, and an associated customer feature corresponding to the associated customer is obtained; According to the associated customer feature, a to-be-recommended customer feature corresponding to the to-be-recommended customer is determined; According to the to-be-recommended customer feature, a target customer corresponding to the to-be-recommended customer is selected from the associated customer; The time distribution of the target historical call record corresponding to the target customer is determined, and the third connection probability corresponding to the to-be-recommended customer in each time period is determined according to the time distribution. According to the customer information corresponding to the to-be-recommended customer, the third connection probability corresponding to each time period is determined, comprising:

5. The customer touch opportunity recommendation method of claim 4, wherein, For any two customers in the associated customer and the to-be-recommended customer, the Mahalanobis distance between the any two customers is calculated according to the associated customer feature and the to-be-recommended customer feature; According to the Mahalanobis distance, the associated customer and the to-be-recommended customer are clustered to obtain a clustering result; According to the clustering result, a target customer corresponding to the to-be-recommended customer is selected from the associated customer. According to the historical call record coding, the first connection probability corresponding to each time period is determined, comprising:

6. The customer touch opportunity recommendation method according to any one of claims 1 to 5, wherein According to the number of time periods, the default connection probability corresponding to each time period is determined; Whether there is a historical call record corresponding to the historical call record coding in each time period is judged to obtain a judgment result; According to the judgment result, the default connection probability is adjusted to obtain the first connection probability corresponding to each time period. ​ 7. The customer touch opportunity recommendation method according to any one of claims 1 to 5, wherein The determining of the touch opportunity corresponding to the to-be-recommended customer according to at least one of the first call connection probability, the second call connection probability and the third call connection probability comprises: performing weighted summation on the first call connection probability, the second call connection probability and the third call connection probability to obtain a predicted call connection probability corresponding to each time period, and constructing a call connection probability set; determining a maximum call connection probability in the call connection probability set; taking a time period corresponding to the maximum call connection probability as the touch opportunity corresponding to the to-be-recommended customer.

8. A customer touch opportunity recommendation apparatus, characterized by, The customer touch opportunity recommendation device comprises: a call connection probability determination module configured to determine a first call connection probability corresponding to each time period according to historical call record encoding; the call connection probability determination module is further configured to determine a second call connection probability corresponding to each time period according to historical behavior time encoding of a to-be-recommended customer; the call connection probability determination module is further configured to determine a third call connection probability corresponding to each time period according to customer information corresponding to the to-be-recommended customer; a touch opportunity recommendation module configured to determine a touch opportunity corresponding to the to-be-recommended customer according to at least one of the first call connection probability, the second call connection probability and the third call connection probability.

9. A customer touch opportunity recommendation device, characterized by, The device comprises a memory, a processor and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the customer touch opportunity recommendation method according to any one of claims 1 to 7.

10. A storage medium, characterized by The storage medium is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the customer touch opportunity recommendation method according to any one of claims 1 to 7.

Images