Recommendation Method and Device for Outbound Users, and Electronic Device

Abstract

The present application provides a method and device for recommending outbound users, an electronic device, and a computer-readable storage medium, which relate to the technical field of data processing. The method sorts the value parameters corresponding to each outbound operator for each outbound user to obtain the sorted value parameters; obtains the outbound work condition data of multiple outbound operators, and constructs a cyclic allocation model according to the outbound work condition data of the multiple outbound operators; based on the sorted value parameters and the cyclic allocation model, cyclically allocates outbound users among the multiple outbound users for each outbound operator to obtain an ordered queue of pre-allocated outbound users for each outbound operator; recommends outbound users according to the ordered queue of pre-allocated outbound users for each outbound operator. By cyclically allocating, the embodiments of the present application pre-allocate outbound users to each outbound operator orderly and in accordance with the quantity, which can realize the calling ability of the outbound operator and improve the efficiency and order conversion rate of outbound calls.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and in particular, to a method and device for recommending outbound users, an electronic device, and a computer-readable storage medium. Background Art

[0002] With the development of mobile Internet technology, various new sales models have emerged, and the sales form has gradually become online. The outbound call system is an important and indispensable part of many complex product services, such as insurance services, financial loan services, wealth management and investment services, real estate services, automotive services, course services, travel route services, event invitations, etc. In the scenario of one-to-many between outbound call operators and outbound users, how to reasonably, orderly, and quantitatively allocate outbound users to outbound call operators to improve the efficiency of outbound calls and the order conversion rate has become a technical problem that needs to be solved urgently. Summary of the Invention

[0003] In view of the above problems, the present application is proposed to provide a method and device for recommending outbound users, an electronic device, and a computer-readable storage medium that can overcome or at least partially solve the above problems, and can improve the efficiency of outbound calls and the order conversion rate. The technical solutions are as follows:

[0004] In a first aspect, a method for recommending outbound users is provided, including:

[0005] Obtain the characteristic data of multiple outbound call operators and the characteristic data of multiple outbound users, and predict the value parameters of each outbound call operator corresponding to each outbound user according to the characteristic data of the multiple outbound call operators and the characteristic data of the multiple outbound users;

[0006] Sort the value parameters of each outbound call operator corresponding to each outbound user to obtain the sorted value parameters;

[0007] Obtain the outbound call working condition data of the multiple outbound call operators, and construct a cyclic allocation model according to the outbound call working condition data of the multiple outbound call operators;

[0008] Based on the sorted value parameters and the cyclic allocation model, cyclically allocate outbound users to each outbound call operator among the multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound call operators;

[0009] Recommend outbound users according to the ordered queue of pre-allocated outbound users for each of the multiple outbound call operators.

[0010] In a possible implementation manner, predicting the value parameters of each outbound call operator corresponding to each outbound user according to the characteristic data of the multiple outbound call operators and the characteristic data of the multiple outbound users includes:

[0011] According to the characteristic data of the multiple outbound operators and the characteristic data of the multiple outbound users, cross - associate the characteristic data of each outbound operator and the characteristic data of each outbound user to generate multiple candidate pairs of characteristics of the outbound operator and the outbound user;

[0012] Input the multiple candidate pairs of characteristics into a pre - trained value prediction model, and use the value prediction model to predict the value parameters of each candidate pair of characteristics, obtaining the predicted value parameters of each candidate pair of characteristics as the value parameters of each outbound operator corresponding to each outbound user.

[0013] In a possible implementation manner, construct a cyclic allocation model according to the outbound work condition data of the multiple outbound operators, including:

[0014] According to the outbound work condition data of the multiple outbound operators, count the outbound work duration of each outbound operator, the number of outbound calls in different time periods, and the order conversion rate;

[0015] According to the outbound work duration of each outbound operator, the number of outbound calls in different time periods, and the order conversion rate, construct a cyclic allocation model including the number of times each outbound operator participates in the cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation.

[0016] In a possible implementation manner, sort the value parameters of each outbound operator corresponding to each outbound user, obtaining the sorted value parameters, including:

[0017] Sort the value parameters of each outbound operator corresponding to each outbound user in descending order of the value parameters, obtaining the sorted value parameters.

[0018] In a possible implementation manner, based on the sorted value parameters and the cyclic allocation model, cyclically allocate outbound users to each outbound operator among the multiple outbound users, obtaining an ordered queue of pre - allocated outbound users for each of the multiple outbound operators, including:

[0019] Based on the sorted value parameters and the cyclic allocation model including the number of times each outbound operator participates in the cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation, cyclically allocate outbound users to each outbound operator among the multiple outbound users, obtaining an ordered queue of pre - allocated outbound users for each of the multiple outbound operators.

[0020] In a possible implementation, based on the sorted value parameters and the cyclic allocation model including the number of times each outbound operator participates in the cyclic allocation and the number of outbound users that can be selected in each round of cyclic allocation, cyclicly allocate outbound users to each outbound operator among the multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound operators, including:

[0021] Construct a two-dimensional table including the sorted value parameters, the outbound operators corresponding to the sorted value parameters, and the outbound users;

[0022] According to the number of times each outbound operator participates in the cyclic allocation and the number of outbound users that can be selected in each round of cyclic allocation, and according to the value parameters, the outbound operators corresponding to the value parameters, and the outbound users in each row of the two-dimensional table, cyclicly allocate outbound users to each outbound operator among the multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound operators; one outbound user can only be allocated to one outbound operator.

[0023] In a possible implementation, perform outbound user recommendation according to the ordered queues of pre-allocated outbound users for each of the multiple outbound operators, including:

[0024] According to the ordered queues of pre-allocated outbound users for each of the multiple outbound operators, allocate corresponding outbound users to each outbound operator; or

[0025] Provide the ordered queues of pre-allocated outbound users for each of the multiple outbound operators to the corresponding outbound operators, and the corresponding outbound operators perform outbound communication with the outbound users in sequence according to the ordered queues of pre-allocated outbound users.

[0026] In a second aspect, there is provided a device for recommending outbound users, including:

[0027] A value prediction module, configured to obtain feature data of multiple outbound operators and feature data of multiple outbound users, and predict value parameters of each outbound operator corresponding to each outbound user according to the feature data of the multiple outbound operators and the feature data of the multiple outbound users;

[0028] A sorting module, configured to sort the value parameters of each outbound operator corresponding to each outbound user to obtain sorted value parameters;

[0029] A construction module, configured to obtain outbound work condition data of the multiple outbound operators, and construct a cyclic allocation model according to the outbound work condition data of the multiple outbound operators;

[0030] A cyclic allocation module, configured to cyclically allocate outbound users among the multiple outbound users for each outbound operator based on the sorted value parameters and the cyclic allocation model, so as to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound operators;

[0031] A recommendation module, configured to recommend outbound users according to the ordered queues of pre-allocated outbound users for each of the multiple outbound operators.

[0032] In a possible implementation manner, the value prediction module is further configured to:

[0033] Cross-correlate the characteristic data of each outbound operator and the characteristic data of each outbound user according to the characteristic data of the multiple outbound operators and the characteristic data of the multiple outbound users, so as to generate multiple candidate pairs of characteristics of outbound operators and outbound users;

[0034] Input the multiple candidate pairs of characteristics into a pre-trained value prediction model, and use the value prediction model to predict the value parameters of each candidate pair of characteristics, so as to obtain the predicted value parameters of each candidate pair of characteristics, which are used as the value parameters of each outbound operator corresponding to each outbound user.

[0035] In a possible implementation manner, the construction module is further configured to:

[0036] According to the outbound work condition data of the multiple outbound operators, count the outbound work duration, the number of outbound calls in different time periods, and the order conversion rate of each outbound operator;

[0037] Construct a cyclic allocation model including the number of times each outbound operator participates in cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation according to the outbound work duration, the number of outbound calls in different time periods, and the order conversion rate of each outbound operator.

[0038] In a possible implementation manner, the sorting module is further configured to:

[0039] Sort the value parameters of each outbound operator corresponding to each outbound user in descending order of value parameters, so as to obtain the sorted value parameters.

[0040] In a possible implementation manner, the cyclic allocation module is further configured to:

[0041] Based on the sorted value parameters and the cyclic allocation model including the number of times each outbound operator participates in cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation, cyclically allocate outbound users among the multiple outbound users for each outbound operator, so as to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound operators.

[0042] In a possible implementation, the loop allocation module is further configured to:

[0043] Construct a two-dimensional table including the sorted value parameters, the outbound operators corresponding to the sorted value parameters, and the outbound users;

[0044] According to the number of times each outbound operator participates in the loop allocation and the number of outbound users allowed to be selected in each round of loop allocation, and according to the value parameters, the outbound operators corresponding to the value parameters, and the outbound users in each row of the two-dimensional table, loop allocate outbound users to each outbound operator among the multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each outbound operator; where one outbound user can only be allocated to one outbound operator.

[0045] In a possible implementation, the recommendation module is further configured to:

[0046] Allocate corresponding outbound users to each outbound operator according to the ordered queues of pre-allocated outbound users for each of the multiple outbound operators; or

[0047] Provide the ordered queues of pre-allocated outbound users for each of the multiple outbound operators to the corresponding outbound operators, and the corresponding outbound operators conduct outbound communications with the outbound users in sequence according to the ordered queues of pre-allocated outbound users.

[0048] In a third aspect, an electronic device is provided, which includes a processor and a memory. Wherein, a computer program is stored in the memory, and the processor is configured to run the computer program to execute the method for recommending outbound users according to any one of the above.

[0049] In a fourth aspect, a computer-readable storage medium is provided, and the computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the method for recommending outbound users according to any one of the above when running.

[0050] With the above technical solution, the method and apparatus for recommending outbound users, electronic device, and computer-readable storage medium provided by the embodiments of the present application can obtain the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users, and predict the value parameters of each outbound operator corresponding to each outbound user according to the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users; then sort the value parameters of each outbound operator corresponding to each outbound user to obtain the sorted value parameters; obtain the outbound work condition data of multiple outbound operators, and construct a cyclic allocation model according to the outbound work condition data of multiple outbound operators; then, based on the sorted value parameters and the cyclic allocation model, cyclically allocate outbound users to each outbound operator among multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each outbound operator; and recommend outbound users according to the ordered queue of pre-allocated outbound users for each outbound operator. It can be seen that the embodiments of the present application combine the characteristic data of outbound operators, the characteristic data of outbound users, and the outbound work condition data of outbound operators, and cyclically allocate outbound users to each outbound operator orderly and in quantity through cyclic allocation, which can realize the calling ability of outbound operators and improve the efficiency and order conversion rate of outbound calls. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments of the present application will be briefly introduced below.

[0052] Figure 1 The flowchart of the method for recommending outbound users provided by the embodiments of the present application is shown;

[0053] Figure 2 The business flowchart of the recommendation of outbound users provided by the embodiments of the present application is shown;

[0054] Figure 3 The structural diagram of the apparatus for recommending outbound users provided by the embodiments of the present application is shown;

[0055] Figure 4 The structural diagram of an electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0056] The exemplary embodiments of the present application will be described in more detail below with reference to the drawings. Although the exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present application can be more thoroughly understood and the scope of the present application can be completely conveyed to those skilled in the art.

[0057] It should be noted that the terms "first," "second," etc. in the description, claims, and the above-mentioned drawings of this application are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such use can be interchanged under appropriate circumstances so that the embodiments of this application described herein can be implemented in an order other than those illustrated or described herein. In addition, the term "including" and its variants should be interpreted as open-ended terms meaning "including but not limited to."

[0058] An embodiment of this application provides a method for recommending outbound call users, which can be applied to electronic devices such as servers, personal computers, smartphones, tablets, smartwatches, etc. As Figure 1 shown, the method for recommending outbound call users can specifically include the following steps S101 to S105:

[0059] Step S101: Obtain the characteristic data of multiple outbound call operators and the characteristic data of multiple outbound call users, and predict the value parameters of each outbound call operator corresponding to each outbound call user according to the characteristic data of the multiple outbound call operators and the characteristic data of the multiple outbound call users;

[0060] Step S102: Sort the value parameters of each outbound call operator corresponding to each outbound call user to obtain the sorted value parameters;

[0061] Step S103: Obtain the outbound call working condition data of multiple outbound call operators, and construct a cyclic allocation model according to the outbound call working condition data of the multiple outbound call operators;

[0062] Step S104: Based on the sorted value parameters and the cyclic allocation model, cyclically allocate outbound call users to each outbound call operator among the multiple outbound call users to obtain an ordered queue of pre-allocated outbound call users for each outbound call operator;

[0063] Step S105: Recommend outbound call users according to the ordered queue of pre-allocated outbound call users for each outbound call operator.

[0064] Embodiments of the present application can obtain the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users, and predict the value parameters of each outbound operator corresponding to each outbound user according to the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users; subsequently, sort the value parameters of each outbound operator corresponding to each outbound user to obtain the sorted value parameters; obtain the outbound work condition data of multiple outbound operators, and construct a cyclic allocation model according to the outbound work condition data of multiple outbound operators; then, based on the sorted value parameters and the cyclic allocation model, cyclically allocate outbound users to each outbound operator among multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each outbound operator; recommend outbound users according to the ordered queue of pre-allocated outbound users for each outbound operator. It can be seen that the embodiments of the present application combine the characteristic data of outbound operators and the characteristic data of outbound users, as well as the outbound work condition data of outbound operators, and through cyclic allocation, pre-allocate outbound users to each outbound operator orderly and in accordance with the quantity, which can realize the calling ability of outbound operators and improve the efficiency and order conversion rate of outbound calls.

[0065] The characteristic data of the outbound operator mentioned in step S101 above can be some characteristic data of the outbound operator itself, such as age, work experience, educational experience, etc., and can also be, such as the number of outbound calls, the number of outbound users, the order completion data of outbound calls, etc. The embodiments of the present application do not limit this.

[0066] The outbound users here can be users who may have demands for insurance services, real estate services, automobile services, course services, etc. The characteristic data of the outbound users can be used to characterize the personal attributes and social attributes of the outbound users. Among them, the characteristic data of the outbound users can specifically include data related to personal attributes, such as gender, age, hobbies, family members, etc., and the characteristic data of the outbound users can specifically also include data related to social attributes, such as the city where they are located, occupation type, income status, etc. It should be noted that the examples here are only illustrative and do not limit the embodiments of the present application.

[0067] A possible implementation manner is provided in the embodiments of the present application. In step S101 above, according to the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users, predicting the value parameters of each outbound operator corresponding to each outbound user may specifically include the following steps A1 to A2:

[0068] Step A1, according to the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users, cross-correlate the characteristic data of each outbound operator and the characteristic data of each outbound user to generate multiple candidate pairs of characteristics of outbound operators and outbound users;

[0069] Step A2: Input multiple candidate feature pairs into a pre-trained value prediction model, and use the value prediction model to predict the value parameters of each candidate feature pair, obtaining the predicted value parameters of each candidate feature pair as the value parameters of each outbound operator corresponding to each outbound user.

[0070] In the embodiment of the present application, by combining the feature data of multiple outbound operators and the feature data of multiple outbound users, the feature data of each outbound operator and the feature data of each outbound user are cross-correlated to generate multiple candidate feature pairs of outbound operators and outbound users. Then, the multiple candidate feature pairs are input into a pre-trained value prediction model, and the value prediction model is used to predict the value parameters of each candidate feature pair, obtaining the predicted value parameters of each candidate feature pair as the value parameters of each outbound operator corresponding to each outbound user. This can recommend some outbound users suitable for the outbound operator in subsequent cyclic allocation by combining some features of the outbound operator itself, thereby forming personalized recommendations for each outbound operator, and improving the efficiency of outbound calls and the order conversion rate.

[0071] The value prediction model mentioned in Step A2 above is pre-trained. Specifically, an initial value prediction model can be constructed, multiple sample candidate feature pairs and historical sample order data are collected, and the initial value prediction model is trained based on the multiple sample candidate feature pairs and historical sample order data to obtain a trained value prediction model.

[0072] In the embodiment of the present application, a possible implementation manner is provided. In Step S103 above, the outbound work condition data of multiple outbound operators is obtained, and a cyclic allocation model is constructed according to the outbound work condition data of the multiple outbound operators. Specifically, it may include the following Steps B1 to B2:

[0073] Step B1: According to the outbound work condition data of multiple outbound operators, count the outbound work duration, outbound call times in different time periods, and order conversion rate of each outbound operator.

[0074] Step B2: According to the outbound work duration, outbound call times in different time periods, and order conversion rate of each outbound operator, construct a cyclic allocation model including the number of times each outbound operator participates in cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation.

[0075] In the embodiment of the present application, by counting the call ability and order completion ability changes of each outbound operator during a day and using them as the subsequent cyclic allocation restriction conditions, each outbound operator has a high-quality candidate set when requesting outbound users at any time during a day, and each high-quality outbound user has a suitable outbound operator to seriously explore its order completion potential.

[0076] In an embodiment of the present application, a possible implementation manner is provided. In the above step S102, the value parameters of each outbound operator corresponding to each outbound user are sorted to obtain the sorted value parameters. Specifically, the value parameters of each outbound operator corresponding to each outbound user can be sorted in descending order of the value parameters to obtain the sorted value parameters. The embodiment of the present application can select high-quality outbound users for subsequent cyclic allocation, so that each outbound operator has a high-quality candidate set when requesting an outbound user at any time of the day, and each high-quality outbound user has a suitable outbound operator to seriously explore its potential for closing a deal.

[0077] In an embodiment of the present application, a possible implementation manner is provided. In the above step S104, based on the sorted value parameters and the cyclic allocation model, outbound users are cyclically allocated to each outbound operator among multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each outbound operator. Specifically, based on the sorted value parameters and the cyclic allocation model including the number of times each outbound operator participates in the cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation, outbound users are cyclically allocated to each outbound operator among multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each outbound operator.

[0078] Based on the sorted value parameters and the cyclic allocation model including the number of times each outbound operator participates in the cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation, the embodiment of the present application cyclically allocates outbound users to each outbound operator among multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each outbound operator, thereby maximizing the overall global value of the outbound user allocation.

[0079] In an embodiment of the present application, a possible implementation manner is provided. Based on the sorted value parameters and the cyclic allocation model including the number of times each outbound operator participates in the cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation, outbound users are cyclically allocated to each outbound operator among multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each outbound operator. Specifically, it may include the following steps C1 to C2:

[0080] Step C1, construct a two-dimensional table including the sorted value parameters, the outbound operators corresponding to the sorted value parameters, and the outbound users;

[0081] Step C2: According to the number of times each outbound operator participates in the cyclic allocation and the number of outbound users that can be selected in each round of cyclic allocation, and according to the value parameters of each row in the two-dimensional table, the outbound operators corresponding to the value parameters, and the outbound users, cyclicly allocate outbound users among multiple outbound users for each outbound operator to obtain an ordered queue of pre-allocated outbound users for each outbound operator; one outbound user can only be allocated to one outbound operator.

[0082] The embodiments of the present application can efficiently and accurately cyclicly allocate outbound users, so that each outbound operator has a high-quality waiting selection set when requesting outbound users at any time of the day, and each high-quality outbound user has a suitable outbound operator to seriously explore its potential for closing orders, improving the efficiency of outbound calls and the order conversion rate.

[0083] A possible implementation manner is provided in the embodiments of the present application. In the above step S105, outbound user recommendation is performed according to the ordered queues of pre-allocated outbound users for each outbound operator. Specifically, it can be to allocate corresponding outbound users to each outbound operator according to the ordered queues of pre-allocated outbound users for each outbound operator; or the ordered queues of pre-allocated outbound users for each outbound operator can also be provided to the corresponding outbound operator, and the corresponding outbound operator performs outbound communication with the outbound users in turn according to the ordered queue of pre-allocated outbound users. The embodiments of the present application can allocate outbound users to each outbound operator orderly and by quantity, improving the efficiency of outbound calls and the order conversion rate.

[0084] The above introduces Figure 1 multiple implementation manners of each link in the shown embodiments. Next, the outbound user recommendation method provided by the embodiments of the present application will be further described through specific embodiments.

[0085] The embodiments of the present application can be used in Internet sales (abbreviated as online sales). Under one allocation rule, combining the characteristic data of all candidate operators (i.e., outbound operators), the characteristic data of the potential conversion users to be called (i.e., outbound users), and the daily calling ability of the outbound operators, all existing outbound users are pre-allocated to each outbound operator orderly and by quantity, so as to maximize the exploration of the calling ability of the outbound operators and the potential for closing orders of each outbound user.

[0086] In the existing online sales recommendation system, after screening out high-quality potential conversion users, secondary value prediction and sorting are performed. Under the allocation rule, high-quality potential conversion users are issued to outbound operators in turn according to the request time of the outbound operators, or high-quality potential conversion users unique to them are screened out according to the characteristics of the outbound operators and distributed in turn. The potential conversion users are in a set and are obtained by the outbound operators on a first-come, first-served basis.

[0087] The existing technology has the following deficiencies in the distribution system:

[0088] 1) The quality of the users to be converted is often high for multiple outbound operators, but the conversion capabilities of outbound operators vary significantly. According to the existing logic, an outbound operator with the second-best ability is assigned to the top high-quality users to be converted before the best outbound operator, which obviously does not maximize the potential of the users to be converted.

[0089] 2) The best outbound operators still have work fatigue period, that is, assuming that the best users to be converted are distributed to the best outbound operators, when the outbound operators focus on some users to be converted with a higher tendency to convert orders, the remaining outbound users distributed to them cannot realize their maximum potential.

[0090] For the online sales recommendation system, maximizing the potential of a limited number of high-quality users to be converted and providing high-quality outbound call operators with a selection of users to be converted that can motivate them to maximize their efficiency in converting orders are two powerful ways to maximize overall efficiency. Based on the above two approaches, the embodiment of the present application comprehensively considers the allocation process of outbound call operators and outbound call users to be converted, so that all outbound call operators and outbound call users to be converted can achieve the maximum global overall value every day.

[0091] Figure 2 The following is a service flow chart showing the recommendation of outbound users provided by the embodiment of the present application. Figure 2 As shown, in the existing online sales recommendation system, under one allocation task, the feature data of all the outbound call operators LP to be selected are cross-correlated with the feature data of the user USER to be converted, so as to form a LP-USER pair to be selected. Here, when the LP-USER set is crossed, the deepFM (deep Factorization Machines) model can be used for feature cross-processing, or the XGBoost (eXtreme Gradient Boosting) model can be used for feature cross-processing; then, the pre-trained value prediction model is used to predict the score of the LP-USER pair to be selected, and the value parameters of each outbound call operator corresponding to each outbound call user are obtained and sorted, so as to obtain the sorted value parameters; combined with the daily dialing capacity of the LP to be selected, a restricted cyclic allocation is constructed, and the cyclic restriction may include the number of leads allowed to be selected by each LP in each round of the cycle, the number of times participating in the cycle, etc. After all the candidate outbound users are allocated, an ordered queue of LP-pre-allocated USER is formed and stored in the online sales recommendation system. When the LP sends a request, it is distributed in order in its own pre-allocated allocation set.

[0092] In a distribution task, there are outbound call operators Lp1 and Lp2 to be selected and users User1, User2, User3, and User4 to be converted. After the cross-features of the entire LP-USER set, the model predicts the full cross-pair score Lp-clue-score. The score distribution is shown in Table 1. Here, for comparison, the original logic is added to directly predict the value of the users to be converted, Clue (clue)-socre.

[0093] Under the original recommendation allocation system, if Lp2 arrives first, it will have priority in obtaining the users to be converted, that is, according to the Lp-clue-score sorting, activating and occupying User1 and User2, Lp1 can only obtain User3 and User4. That is, the average of the total Lp-clue-score is 0.7325. And according to the statistics of the outbound call operator's ability to make orders, when Lp1 has one order, the potential for exploring the order-making potential of the remaining outbound call users to be converted will decrease.

[0094] In the embodiment of the present application, the Lp-clue-score in Table 1 is sorted, and a circular allocation is performed according to the sorted Lp-clue-score. In each round, each LP is allocated an outbound user to be converted, and multiple rounds of circular allocation are continued until all candidate users to be converted are allocated. That is, Lp1 is allocated to User1 and User4, and Lp2 is allocated to User2 and User3. That is, the average value of the total Lp-clue-score is 0.7525. And each LP has obtained its own optimal users to be converted to the greatest extent, and at the same time, it has achieved the highest probability that the best user groups to be converted are carefully explored by the operators to have the potential to be converted into orders.

[0095] LP USER Lp-clue-score Clue-socre Lp1 User1 0.9 0.82 Lp1 User2 0.85 0.73 Lp1 User3 0.7 0.65 Lp1 User4 0.6 0.54 Lp2 User1 0.85 0.82 Lp2 User2 0.78 0.73 Lp2 User3 0.73 0.65 Lp2 User4 0.69 0.54

[0096] Table 1

[0097] In actual applications, in each round of allocation, the number of outbound users allocated to each LP can be set according to the capacity of the LP. For example, the number of outbound users allocated to LP1 in three rounds of allocation is 1, 2, and 1, and the number of outbound users allocated to LP2 in three rounds of allocation is 1, 1, and 2. It should be noted that the examples given here are only illustrative and do not limit the embodiments of the present application.

[0098] The embodiments of the present application strengthen the concept of globally optimal pre - allocation between outbound operators and outbound users to be converted in the online sales recommendation system, thereby achieving the maximization of benefits. By statistically analyzing the calling ability and order - closing ability changes of each outbound operator during a day, and using them as the scores for restricted full - volume cross - prediction for cyclic allocation, so that each LP has a high - quality candidate set when requesting users to be converted at any time during the day, and each high - quality user to be converted has a suitable LP to seriously explore its order - closing potential. After the model strategy is launched, not only the overall order - closing rate is effectively improved, but also some sub - excellent LPs can be motivated by high - quality leads, promoting their work enthusiasm and future order - closing ability, and realizing the sustainable growth of the system.

[0099] The embodiments of the present application analyze the daily order - closing ability curve of LPs in the online sales system, and then determine to allocate an appropriate number of users to be converted among high - quality users to be converted at different levels for each of them, achieving the optimal exploration of each high - quality user to be converted. Moreover, in the online sales system, based on the predicted scores of the full - volume cross - set of LP - users to be converted, restricted cyclic allocation is performed, and the restricted parameter control is carried out according to the online sales business scenario to maximize the global benefits of lead allocation.

[0100] It should be noted that in practical applications, all the above - mentioned possible implementation manners can be combined in any combination to form possible embodiments of the present application, which will not be elaborated one by one here.

[0101] Based on the outbound user recommendation method provided in the above - mentioned embodiments, based on the same inventive concept, the embodiments of the present application also provide an outbound user recommendation device.

[0102] Figure 3 The structure diagram of the outbound user recommendation device provided by the embodiments of the present application is shown. As Figure 3 shown, the outbound user recommendation device may include a value prediction module 310, a sorting module 320, a construction module 330, a cyclic allocation module 340, and a recommendation module 350.

[0103] The value prediction module 310 is used to obtain the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users, and predict the value parameters of each outbound operator corresponding to each outbound user according to the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users;

[0104] The sorting module 320 is used to sort the value parameters of each outbound operator corresponding to each outbound user to obtain the sorted value parameters;

[0105] The construction module 330 is used to obtain the outbound work condition data of multiple outbound operators and construct a cyclic allocation model according to the outbound work condition data of multiple outbound operators.

[0106] A cyclic allocation module 340, configured to cyclically allocate outbound users to each outbound operator among multiple outbound users based on the sorted value parameters and a cyclic allocation model, so as to obtain an ordered queue of pre-allocated outbound users for each outbound operator;

[0107] A recommendation module 350, configured to recommend outbound users according to the ordered queues of pre-allocated outbound users for each of the multiple outbound operators.

[0108] In an embodiment of the present application, a possible implementation manner is provided. The Figure 3 value prediction module 310 shown above is further configured to:

[0109] Cross-correlate the characteristic data of each outbound operator and the characteristic data of each outbound user according to the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users, so as to generate multiple candidate pairs of characteristics of outbound operators and outbound users;

[0110] Input the multiple candidate pairs of characteristics into a pre-trained value prediction model, and use the value prediction model to predict the value parameters of each candidate pair of characteristics, so as to obtain the predicted value parameters of each candidate pair of characteristics, which serve as the value parameters of each outbound operator corresponding to each outbound user.

[0111] In an embodiment of the present application, a possible implementation manner is provided. The Figure 3 construction module 330 shown above is further configured to:

[0112] According to the outbound work condition data of multiple outbound operators, count the outbound work duration, the number of outbound calls in different time periods, and the order conversion rate of each outbound operator;

[0113] Construct a cyclic allocation model including the number of times each outbound operator participates in cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation according to the outbound work duration, the number of outbound calls in different time periods, and the order conversion rate of each outbound operator.

[0114] In an embodiment of the present application, a possible implementation manner is provided. The Figure 3 sorting module 320 shown above is further configured to:

[0115] Sort the value parameters of each outbound operator corresponding to each outbound user in descending order of the value parameters, so as to obtain the sorted value parameters.

[0116] In an embodiment of the present application, a possible implementation manner is provided. The Figure 3 cyclic allocation module 340 shown above is further configured to:

[0117] Based on the sorted value parameters and a cyclic allocation model that includes the number of times each outbound operator participates in the cyclic allocation and the number of outbound users that can be selected in each round of cyclic allocation, outbound users are cyclically allocated to each outbound operator among multiple outbound users, and an ordered queue of pre-allocated outbound users for each outbound operator is obtained.

[0118] In an embodiment of the present application, a possible implementation is provided. As mentioned above Figure 3 The cyclic allocation module 340 shown above is further configured to:

[0119] Construct a two-dimensional table including the sorted value parameters, the outbound operators corresponding to the sorted value parameters, and the outbound users;

[0120] According to the number of times each outbound operator participates in the cyclic allocation and the number of outbound users that can be selected in each round of cyclic allocation, and according to the value parameters, the outbound operators corresponding to the value parameters, and the outbound users in each row of the two-dimensional table, outbound users are cyclically allocated to each outbound operator among multiple outbound users, and an ordered queue of pre-allocated outbound users for each outbound operator is obtained; where one outbound user can only be allocated to one outbound operator.

[0121] In an embodiment of the present application, a possible implementation is provided. As mentioned above Figure 3 The recommendation module 350 shown above is further configured to:

[0122] Allocate corresponding outbound users to each outbound operator according to the ordered queues of pre-allocated outbound users for each of the multiple outbound operators; or

[0123] Provide the ordered queues of pre-allocated outbound users for each of the multiple outbound operators to the corresponding outbound operators, and the corresponding outbound operators conduct outbound communications with the outbound users in sequence according to the ordered queues of pre-allocated outbound users.

[0124] Based on the same inventive concept, an embodiment of the present application further provides an electronic device, including a processor and a memory. A computer program is stored in the memory, and the processor is configured to run the computer program to execute the method for recommending outbound users in any one of the above embodiments.

[0125] In an exemplary embodiment, an electronic device is provided, as Figure 4 shown Figure 4 The electronic device 400 shown includes: a processor 401 and a memory 403. Among them, the processor 401 and the memory 403 are connected, such as connected through a bus 402. Optionally, the electronic device 400 may further include a transceiver 404. It should be noted that in actual applications, the transceiver 404 is not limited to one, and the structure of the electronic device 400 does not constitute a limitation to the embodiments of the present application.

[0126] The processor 401 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of the present application. The processor 401 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0127] The bus 402 may include a path for transmitting information between the above components. The bus 402 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus 402 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 only a thick line is shown in the figure, but it does not mean that there is only one bus or one type of bus.

[0128] The memory 403 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, or it may also be an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0129] The memory 403 is used to store the application program code for executing the solution of this application, and is controlled by the processor 401 to execute. The processor 401 is used to execute the application program code stored in the memory 403 to implement the content shown in the foregoing method embodiments.

[0130] Among them, the electronic device includes but is not limited to: mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The illustrated electronic device is only an example and should not impose any restrictions on the functions and usage scope of the embodiments of this application.

[0131] Based on the same inventive concept, the embodiments of this application also provide a computer-readable storage medium, in which a computer program is stored. Among them, the computer program is set to execute the recommended method for calling external users in any of the foregoing embodiments when running.

[0132] Those skilled in the art can clearly understand the specific working processes of the above-described systems, devices, and modules, and can refer to the corresponding processes in the foregoing method embodiments. For the sake of brevity, they will not be described in detail here.

[0133] Those of ordinary skill in the art can understand that the technical solution of this application can essentially or all or part of the technical solution be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several program instructions, so that an electronic device (such as a personal computer, a server, or a network device, etc.) executes all or part of the steps of the methods described in the embodiments of this application when running the program instructions. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0134] Alternatively, all or part of the steps of implementing the foregoing method embodiments can be completed by hardware related to program instructions (such as an electronic device such as a personal computer, a server, or a network device), and the program instructions can be stored in a computer-readable storage medium. When the program instructions are executed by the processor of the electronic device, the electronic device executes all or part of the steps of the methods described in the embodiments of this application.

[0135] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that within the spirit and principle of the present application, it is still possible to modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the corresponding technical solutions to deviate from the protection scope of the present application.

Claims

1. A method for recommending to outbound users, characterized in that Including: Obtain the characteristic data of multiple outbound operators and the characteristic data of multiple outbound users, and predict the value parameters of each outbound operator corresponding to each outbound user according to the characteristic data of the multiple outbound operators and the characteristic data of the multiple outbound users; Sort the value parameters of each outbound operator corresponding to each outbound user from large to small to obtain the sorted value parameters; Obtain the outbound work condition data of the multiple outbound operators, and construct a cyclic allocation model according to the outbound work condition data of the multiple outbound operators; the cyclic allocation model includes the number of times each outbound operator participates in cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation; Based on the sorted value parameters and the cyclic allocation model, cyclically allocate outbound users to each outbound operator among the multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound operators; Recommend outbound users according to the ordered queue of pre-allocated outbound users for each of the multiple outbound operators.

2. The recommended method for calling external users according to claim 1, characterized in that, Predict the value parameters of each outbound operator corresponding to each outbound user according to the characteristic data of the multiple outbound operators and the characteristic data of the multiple outbound users, including: According to the characteristic data of the multiple outbound operators and the characteristic data of the multiple outbound users, cross-correlate the characteristic data of each outbound operator and the characteristic data of each outbound user to generate multiple candidate pairs of characteristics of outbound operators and outbound users; Input the multiple candidate pairs of characteristics into a pre-trained value prediction model, and use the value prediction model to predict the value parameters of each candidate pair of characteristics to obtain the predicted value parameters of each candidate pair of characteristics, which are used as the value parameters of each outbound operator corresponding to each outbound user.

3. The recommended method for outbound calling users according to claim 1, wherein Construct a cyclic allocation model according to the outbound work condition data of the multiple outbound operators, including: According to the outbound work condition data of the multiple outbound operators, count the outbound work duration, the number of outbound calls in different time periods, and the order conversion rate of each outbound operator; Construct a cyclic allocation model according to the outbound work duration, the number of outbound calls in different time periods, and the order conversion rate of each outbound operator.

4. The recommended method for outbound calling users according to claim 1, wherein Based on the sorted value parameters and the cyclic allocation model, cyclically allocate outbound users to each outbound operator among the multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound operators, including: Based on the sorted value parameters and the cyclic allocation model including the number of times each outbound operator participates in cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation, cyclically allocate outbound users to each outbound operator among the multiple outbound users to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound operators.

5. The recommended method for outbound users according to claim 4, wherein Based on the sorted value parameters and the loop allocation model including the number of times each outbound operator participates in the loop allocation and the number of outbound users allowed to be selected in each round of loop allocation, loop allocate outbound users to each outbound operator among the multiple outbound users to obtain the pre-allocated outbound user ordered queues for each of the multiple outbound operators, including: Construct a two-dimensional table including the sorted value parameters, the outbound operators corresponding to the sorted value parameters, and the outbound users; According to the number of times each outbound operator participates in the loop allocation and the number of outbound users allowed to be selected in each round of loop allocation, and according to the value parameters, the outbound operators corresponding to the value parameters, and the outbound users in each row of the two-dimensional table, loop allocate outbound users to each outbound operator among the multiple outbound users to obtain the pre-allocated outbound user ordered queues for each of the multiple outbound operators; one outbound user can only be allocated to one outbound operator.

6. The recommended method for outbound users according to claim 1, wherein Perform recommendation of outbound users according to the pre-allocated outbound user ordered queues for each of the multiple outbound operators, including: According to the pre-allocated outbound user ordered queues for each of the multiple outbound operators, allocate the corresponding outbound users to each outbound operator; or Provide the pre-allocated outbound user ordered queues for each of the multiple outbound operators to the corresponding outbound operators, and the corresponding outbound operators conduct outbound communications with the outbound users in sequence according to the pre-allocated outbound user ordered queues.

7. A recommendation device for outbound calling users, characterized in that, Including: A value prediction module, configured to obtain the feature data of multiple outbound operators and the feature data of multiple outbound users, and predict the value parameters of each outbound operator corresponding to each outbound user according to the feature data of the multiple outbound operators and the feature data of the multiple outbound users; A sorting module, configured to sort the value parameters of each outbound operator corresponding to each outbound user from large to small to obtain the sorted value parameters; A construction module, configured to obtain the outbound work condition data of the multiple outbound operators, and construct a loop allocation model according to the outbound work condition data of the multiple outbound operators; the loop allocation model includes the number of times each outbound operator participates in the loop allocation and the number of outbound users allowed to be selected in each round of loop allocation; A loop allocation module, configured to loop allocate outbound users to each outbound operator among the multiple outbound users based on the sorted value parameters and the loop allocation model to obtain the pre-allocated outbound user ordered queues for each of the multiple outbound operators; A recommendation module, configured to perform recommendation of outbound users according to the pre-allocated outbound user ordered queues for each of the multiple outbound operators.

8. The recommended device for outbound calling users according to claim 7, wherein, The value prediction module is further configured to: According to the feature data of the multiple outbound operators and the feature data of the multiple outbound users, cross-correlate the feature data of each outbound operator and the feature data of each outbound user to generate multiple candidate pairs of features of outbound operators and outbound users; Input the multiple pairs of candidate features into a pre-trained value prediction model, and use the value prediction model to predict the value parameters of each pair of candidate features, so as to obtain the predicted value parameters of each pair of candidate features, which are used as the value parameters of each outbound operator corresponding to each outbound user.

9. The recommendation device for outbound users according to claim 7, wherein The construction module is further configured to: According to the outbound work condition data of the multiple outbound operators, count the outbound work duration, the number of outbound calls in different time periods, and the order conversion rate of each outbound operator; Construct a cyclic allocation model according to the outbound work duration, the number of outbound calls in different time periods, and the order conversion rate of each outbound operator.

10. The recommended device for outbound calling users according to claim 7, characterized in that, The cyclic allocation module is further configured to: Based on the sorted value parameters and the cyclic allocation model including the number of times each outbound operator participates in the cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation, cyclically allocate outbound users to each outbound operator among the multiple outbound users, so as to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound operators.

11. The recommended device for outbound calling users according to claim 10, characterized in that, The cyclic allocation module is further configured to: Construct a two-dimensional table including the sorted value parameters, the outbound operators corresponding to the sorted value parameters, and the outbound users; According to the number of times each outbound operator participates in the cyclic allocation and the number of outbound users allowed to be selected in each round of cyclic allocation, and according to the value parameters, the outbound operators corresponding to the value parameters, and the outbound users in each row of the two-dimensional table, cyclically allocate outbound users to each outbound operator among the multiple outbound users, so as to obtain an ordered queue of pre-allocated outbound users for each of the multiple outbound operators; One outbound user can only be allocated to one outbound operator.

12. The recommended device for outbound users according to claim 7, wherein The recommendation module is further configured to: Allocate corresponding outbound users to each outbound operator according to the ordered queues of pre-allocated outbound users for each of the multiple outbound operators; or Provide the ordered queues of pre-allocated outbound users for each of the multiple outbound operators to the corresponding outbound operators, and the corresponding outbound operators conduct outbound communications with the outbound users in sequence according to the ordered queues of pre-allocated outbound users.

13. An electronic device, characterized in that, It includes a processor and a memory. Among them, a computer program is stored in the memory, and the processor is configured to run the computer program to execute the method for recommending outbound users according to any one of claims 1 to 6.

14. A computer-readable storage medium, characterized in that, A computer program is stored in the computer-readable storage medium, and the computer program is configured to execute the method for recommending outbound users according to any one of claims 1 to 6 when running.

Images