Clue issuing method and device, electronic equipment and medium

Abstract

The invention relates to a clue issuing method and device, electronic equipment and a medium, and can be applied to the field of information distribution and sales, and the method comprises the steps: obtaining historical clue issuing data and historical clue processing data of each object in a to-be-issued group; according to the attribute information of each object, the historical clue issuing data and the historical clue processing data, generating ability portrait feature information and processing state information of each object; according to the ability portrait feature information and the processing state information, generating issuing matching information between a to-be-issued clue in a to-be-issued clue library and a to-be-issued group; and issuing the to-be-issued clue to the corresponding target object based on the issuing matching information. According to the method, the matching degree of the distributed clues and the distribution objects can be improved, clue distribution based on the supply-demand relation, the personal processing state and the personal ability is achieved, the follow-up efficiency and follow-up effect of the later clues can be improved, and the intelligent degree of clue distribution can also be effectively improved.

Description

Technical Field

[0001] The present disclosure relates to the fields of sales and information distribution technology, and in particular to a method, device, electronic device, and medium for issuing leads. Background Art

[0002] Across various industries, the distribution of leads and dedicated follow-up help achieve goals. In the sales field, for example, leads serve as an information source within the sales management system. These leads are primarily tracked and processed to reach potential customers, generating business volume, orders, or anticipated targets. Leads can generally be acquired through various means, including events, online information releases, interviews, personal follow-up, and outreach. Therefore, in the sales field, managing leads from potential customers and distributing them to relevant personnel for follow-up is crucial for achieving sales targets.

[0003] As the number of leads increases, the management of leads is gradually becoming more intelligent and refined. For example, refined management of leads can be performed through some enterprise resource planning (ERP) systems or customer relationship management (CRM) modules.

[0004] In the process of realizing the concept disclosed herein, the inventors discovered that there are at least the following technical problems in the related technology: in the process of distributing leads to relevant personnel, most of the time, managers distribute the leads to various responsible personnel (such as sales personnel or business personnel, etc.) based on their personal experience. This distribution method relies on personal experience, and there will be problems such as unreasonable distribution leading to poor efficiency or poor follow-up results in the subsequent lead follow-up. Summary of the Invention

[0005] In order to solve the above technical problems or at least partially solve the above technical problems, embodiments of the present disclosure provide a method, device, electronic device, and medium for issuing clues.

[0006] In a first aspect, embodiments of the present disclosure provide a method for thread dispatching. The method comprises: obtaining historical thread dispatching data and historical thread processing data for each object in a to-be-dispatched group; generating capability profile feature information and processing status information for each object based on attribute information of each object, the historical thread dispatching data, and the historical thread processing data; generating dispatch matching information between the to-be-dispatched thread in a to-be-dispatched thread library and the to-be-dispatched group based on the capability profile feature information and the processing status information; and dispatching the to-be-dispatched thread to the corresponding target object based on the dispatch matching information.

[0007] According to an embodiment of the present disclosure, the delivery matching information between the to-be-delivered clues in the to-be-delivered clue library and the to-be-delivered group is generated according to the capability profile feature information and the processing state information, including: determining the number of clues to be delivered and the delivery time of each object in the to-be-delivered group according to the processing state information of each object in the to-be-delivered group; determining the candidate object whose capability profile matches the to-be-delivered clue according to the capability profile feature information of each object in the to-be-delivered group; performing allocation processing on the amount of to-be-delivered clues matched by the candidate object according to the number of clues to be delivered, to obtain the delivery matching relationship between the to-be-delivered clues and the candidate object; and generating the delivery matching information according to the delivery matching relationship and the delivery time.

[0008] According to an embodiment of the present disclosure, the delivery matching relationship between the to-be-delivered clues and the candidate object is obtained by performing allocation processing on the amount of to-be-delivered clues matched by the candidate object according to the number of clues to be delivered, including: determining the relative size between the amount of to-be-delivered clues matched by the candidate object and the number of clues to be delivered; for a first candidate object whose amount of to-be-delivered clues is the same as the corresponding number of clues to be delivered, a first delivery matching relationship between the first candidate object and the matched to-be-delivered clue is constructed. For a second candidate object whose amount of to-be-delivered clues is less than the corresponding number of clues to be delivered, a second delivery matching relationship between the second candidate object and the matched to-be-delivered clue is constructed; or, the second candidate object is allocated with a first target to-be-delivered clue from other to-be-delivered clues, and a third delivery matching relationship between the second candidate object and the first target to-be-delivered clue is constructed; the first target to-be-delivered clue includes the matched to-be-delivered clue and a supplementary clue, and the supplementary clue is used to supplement the number of clues to be delivered of the second candidate object. For a third candidate object whose amount of to-be-delivered clues is greater than the corresponding number of clues to be delivered, the batch division of to-be-delivered clues is performed according to the tracking priority information of the matched to-be-delivered clues, to obtain a first batch of clues whose tracking priority is high and whose amount of clues meets the number of clues to be delivered, and a fourth delivery matching relationship between the third candidate object and the first batch of clues is constructed; the to-be-delivered clues in subsequent batches are to-be-delivered clues of the third candidate object in subsequent periods or are supplementary clues of the second candidate object.

[0009] According to an embodiment of the present disclosure, the processing state information includes: processing supply and demand state and efficient processing time; the efficient processing time of each object is obtained by counting the time sequence processing effect of the current object. The number of clues to be delivered matches the processing supply and demand state of the corresponding object; and the delivery time matches the efficient processing time of the corresponding object.

[0010] According to an embodiment of the present disclosure, the capability profile feature information comprises at least one of: lead signing feature information, lead supply feature information, and signing target achievement feature information. Based on the capability profile feature information of each object in the to-be-issued group, a candidate object that matches the capability profile and the to-be-issued lead is determined, comprising: generating a capability profile meta-label in a preset dimension according to the capability profile feature information; combining the capability profile meta-label according to the capability combination information required by the to-be-issued lead to obtain a composite label; and determining a candidate object that matches the composite label.

[0011] According to an embodiment of the present disclosure, the capability profile feature information comprises at least one of: lead signing feature information, lead supply feature information, and signing target achievement feature information. Based on the capability profile feature information of each object in the to-be-issued group, a candidate object that matches the capability profile and the to-be-issued lead is determined, comprising: generating a capability profile meta-label in a preset dimension according to the capability profile feature information; combining the capability profile meta-label according to the capability combination information required by the to-be-issued lead to obtain a composite label; and determining a candidate object that matches the composite label.

[0012] According to an embodiment of the present disclosure, the lead signing feature information comprises: signing rate data and processing rate data, and at least one of the signing rate data or the processing rate data is a result processed by Bayesian smoothing. The Bayesian smoothing processing comprises: calculating the overall mean and variance of initial data of a same object for multiple types of leads, and adjusting the initial data by using the overall mean and variance as prior probability to obtain adjusted data; and the initial data comprises at least one of the signing rate data or the processing rate data.

[0013] In a second aspect, an embodiment of the present disclosure provides a device for lead issuing. The device comprises: a data acquisition module, a feature information generation module, a matching module, and an issuing module. The data acquisition module is configured to acquire historical lead issuing data and historical lead processing data of each object in a to-be-issued group. The feature information generation module is configured to generate capability profile feature information and processing state information of each object according to attribute information of each object, the historical lead issuing data, and the historical lead processing data. The matching module is configured to generate issuing matching information between a to-be-issued lead in a to-be-issued lead library and the to-be-issued group according to the capability profile feature information and the processing state information. The issuing module is configured to issue the to-be-issued lead to a corresponding target object based on the issuing matching information.

[0014] In a third aspect, embodiments of the present disclosure provide an electronic device. The electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory complete communication with each other through the communication bus; the memory is configured to store a computer program; and the processor is configured to execute the program stored in the memory to implement the method for issuing a clue as described above.

[0015] In a fourth aspect, embodiments of the present disclosure provide a computer readable storage medium. The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method for issuing a clue as described above.

[0016] The above technical solutions provided by the embodiments of the present disclosure have at least some or all of the following advantages:

[0017] By obtaining the historical clue issuing data and the historical clue processing data of each object in the to-be-issued group, generating the ability portrait feature information and the processing state information of each object according to the attribute information of each object, the historical clue issuing data, and the historical clue processing data, since the ability portrait feature information can reflect the ability portrait features of each object in the to-be-issued group, and the processing state information can reflect the task saturation degree in the clue processing process, the processing efficiency and the processing effect in the specific task processing process, etc., the issuing matching information generated according to the ability portrait feature information and the processing state information is matched with the ability portrait and the processing state of each object, which can improve the matching degree of the distributed clues and the distributed objects, implement the clue distribution based on the supply-demand relationship, the individual processing state, and the individual ability, help to improve the follow-up efficiency and the follow-up effect of the clues, and effectively improve the intelligent degree of the clue distribution. BRIEF DESCRIPTION OF DRAWINGS

[0018] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate embodiments consistent with the present disclosure and, together with the description, further serve to explain the principles of the present disclosure.

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the prior art, brief descriptions will be given to the drawings needed to be used in the embodiments or the related technical descriptions. Obviously, for those skilled in the art, other drawings can also be obtained from these drawings without any creative effort.

[0020] Figure 1 The system architecture of the method and the device suitable for the clue distribution according to the embodiments of the present disclosure is schematically shown;

[0021] Figure 2 The flowchart of the method for issuing a clue according to an embodiment of the present disclosure is schematically shown.

[0022] Figure 3 A detailed implementation flowchart of step S230 according to an embodiment of the present disclosure is schematically shown;

[0023] Figure 4 A detailed implementation flowchart of step S320 according to an embodiment of the present disclosure is schematically shown;

[0024] Figure 5 A detailed implementation flowchart of step S330 according to an embodiment of the present disclosure is schematically shown;

[0025] Figure 6 A structural block diagram of a device for issuing a clue according to an embodiment of the present disclosure is schematically shown;

[0026] Figure 7 A structural block diagram of an electronic device provided by an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0027] To make the objectives, technical solutions and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments of the present disclosure. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present disclosure.

[0028] Figure 1 A system architecture of a method and device for clue distribution suitable for an embodiment of the present disclosure is schematically shown.

[0029] REFERENCE Figure 1 As shown in the figure, the system architecture 100 of the method and device for clue distribution suitable for an embodiment of the present disclosure includes terminal devices 101, 102, 103, a network 104 and a server 105. The network 104 is a medium providing a communication link between the terminal devices 101, 102, 103 and the server 105. The network 104 can include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0030] A user can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. The terminal devices 101, 102, 103 can be installed with a clue management type application, such as an enterprise resource planning (ERP) system, a customer relationship management (CRM) system, etc.; the above-mentioned clue management type application is used to realize the entry of clue information, the distribution of received clues, the state follow-up management of distributed clues, etc.

[0031] The terminal device can also be installed with other communication client applications, such as shopping applications, web browser applications, video playing applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0032] The terminal devices 101, 102, and 103 can be electronic devices with display screens, including but not limited to smartphones, tablet computers, notebook computers, smartwatches, desktop computers, vehicle-mounted intelligent terminals, etc.

[0033] The server 105 can be a background management server that provides data processing and data storage service support for the lead management application on the terminal devices 101, 102, and 103. The server 105 executes the method for distributing leads provided by the embodiments of the present disclosure.

[0034] It should be noted that the method for distributing leads provided by the embodiments of the present disclosure can generally be executed by the server 105 or a terminal device with certain computing capability. Accordingly, the apparatus for distributing leads provided by the embodiments of the present disclosure can generally be arranged in the server 105 or the terminal device with certain computing capability. The method for distributing leads provided by the embodiments of the present disclosure can also be executed by a server or a server cluster different from the server 105 and capable of communicating with at least one of the terminal devices 101, 102, and 103 or the server 105. Accordingly, the apparatus for distributing leads provided by the embodiments of the present disclosure can also be arranged in a server or a server cluster different from the server 105 and capable of communicating with at least one of the terminal devices 101, 102, and 103 or the server 105.

[0035] It should be understood that the number of terminal devices, networks, and servers in the system architecture 100 is only illustrative. Any number of terminal devices, networks, and servers can be provided according to implementation needs. Figure 1

[0036] The first exemplary embodiment of the present disclosure provides a method for distributing leads. The method of the present embodiment can be applied to Figure 1 The server 105 in the exemplary system architecture 100.

[0037] The lead described in the embodiments of the present disclosure refers to a broad meaning of lead. The lead can be relevant information, relevant channels, relevant methods, etc. on the path to reach a certain target, which can be applied to the sales field but is not limited to this field.

[0038] Figure 2 A flowchart of the method for distributing leads according to an embodiment of the present disclosure is schematically shown.

[0039] Referring to Figure 2 ​As shown, the method for issuing a clue provided by the embodiments of the present disclosure includes the following steps: S210, S220, S230 and S240.

[0040] In step S210, the historical clue issuing data and the historical clue processing data of each object in the to-be-issued group are acquired.

[0041] The to-be-issued group refers to a group set of objects for issuing a clue, for example, a sales group of a certain logistics product, a business group corresponding to a certain target task, etc. The personnel in the above sales group or business group are the objects of this clue issuing, and the clue to be issued in the process of this clue issuing is described as a to-be-issued clue, and the object to be distributed is described as a target object. The target object can be part or all of the personnel in the to-be-issued group.

[0042] The historical clue issuing data and the historical clue processing data of each object in the above to-be-issued group can be acquired, so that the clue processing ability and the current processing state of each object can be evaluated and portrayed, and then the matching of the clue issuing is performed based on these information.

[0043] In a sales scenario, the historical clue issuing data of each object can be, for example, the historical clue information issued by the current object, including but not limited to: the clue issuing time of the historical clue, the clue type, the clue name, the business target of the historical issued clue, the total amount of clue issuing in the statistical period, the amount of clue issuing of each type of clue in the statistical period, the issuing frequency of the same type of clue, etc.

[0044] The historical clue processing data of each object can be, for example: the time of processing a certain type of clue by the current object, the processing period, the clue processing order, the processing preference information of different types of clues, the processing effect (such as the clue-to-business opportunity rate, the clue signing amount, the clue generated business amount, order amount or expected revenue, the clue signing rate, the clue signing target proportion, etc.), the efficient processing period or the efficient processing time, etc.

[0045] In step S220, the ability portrait feature information and the processing state information of each object are generated according to the attribute information of each object, the historical clue issuing data and the historical clue processing data.

[0046] The above attribute information can include attribute information associated with the clue target achievement. For example, in a sales scenario, the clue target achievement is to promote order generation, and the above object can be a salesperson, and the corresponding attribute information can be: the age, gender, work experience (or described as work proficiency), education, salesmanship level, basic knowledge level of a certain business, etc. of the salesperson.

[0047] In some embodiments, according to the attribute information of each object, the historical clue issuing data and the historical clue processing data, the ability portrait of each object can be constructed and the current processing state of each object can be determined, which can be specifically implemented by generating the ability portrait feature information and the processing state information of each object.

[0048] In some embodiments, the above-mentioned ability portrait feature information covers at least one of the following: clue signing feature information, clue supply feature information, signing target achievement feature information, etc. In a sales scenario, the above-mentioned clue signing feature information refers to the ability information for promoting sales business by using clues, for example, the signing amount and the signing rate achieved by using a certain type of clue for the current object; the above-mentioned clue supply feature information refers to the related ability information of the current object for providing clues, for example, the clue amount provided by the current object, the hot clue amount (the hot clue amount refers to the popular clue amount corresponding to the clue amount, which is opposite to the cold clue, and the cold clue refers to the relatively unpopular clue), etc. The above-mentioned signing target achievement feature information refers to the pre-set completion index for the result of signing, for example, the pre-set completion index for the signing business amount or the signing order amount, and the completion degree or the completion ratio of the actual business amount or the order amount completed by signing compared to the pre-set completion index can be used as the signing target achievement feature information.

[0049] In other scenarios, the above-mentioned clue signing feature information can be a phased result associated with the target achieved by the clue, for example, in the scenarios of character tracking, task tracking, video personalized recommendation, etc., the above-mentioned clue signing feature information can include: whether a certain clue can be used as an effective tracking clue, the proportion of effective tracking clues in all clues, etc. Similarly, in other scenarios, the above-mentioned signing target achievement feature information can refer to the completion degree or the completion ratio of the state (a phased result of achieving the target) of signing compared to the pre-set completion index, etc.

[0050] The specific content of the ability portrait feature information in the sales scenario is illustrated below with reference to Table 1.

[0051] Table 1: Ability portrait feature information example in a sales scenario

[0052]

[0053]

[0054]

[0055] The processing state information includes: processing supply and demand state and efficient processing time; and the efficient processing time of each object is obtained according to the time sequence processing effect of the current object. According to the processing state information of each object, the clue processing mode (such as what time period is efficient for processing different types of clues, how to arrange the processing order for different types of clues, etc.) of the current object and the saturation degree of the processing task can be obtained.

[0056] In step S230, according to the above-mentioned capability portrait feature information and the above-mentioned processing state information, the delivery matching information between the to-be-delivered clues in the to-be-delivered clue library and the above-mentioned to-be-delivered group is generated.

[0057] The to-be-delivered clue library refers to the database for storing the to-be-delivered clues in the current time period.

[0058] As an example, there are 50 to-be-delivered clues in the to-be-delivered clue library, and these 50 to-be-delivered clues are finely managed, for example, pre-classified or classified and labeled. Specifically, the to-be-delivered clues can be classified and managed finely according to the required comprehensive ability of the clues, the relative importance of the clues, the urgency of the follow-up required by the clues, the required follow-up time length of the clues, etc. There are five objects A, B, C, D and E in the to-be-delivered group. According to the capability portrait feature information and the processing state information of the objects A-E, the 500 to-be-delivered clues can be matched to the objects whose capability and task supply and demand state are matched. For example, the first to the tenth classified important to-be-delivered clues are matched to the object A whose large-disk achievement ability is strong and the supply and demand state is demand greater than supply state (which means that the current processing amount is not saturated); the eleventh to the twenty-fifth classified less important to-be-delivered hot clues are matched to the object B whose large-disk hot clue processing ability is strong and the supply and demand state is demand greater than supply state (which means that the current processing amount is not saturated); the twenty-sixth to the fortieth classified unimportant but short-time follow-up to-be-delivered hot clues are matched to the object C whose processing state is relatively idle in a short time, the object C is in the efficient processing period of the object C in the near future, and the clue follow-up processing efficiency is high; the forty-first to the forty-fourth classified important degree general to-be-delivered cold clues are matched to the object C whose cold clue achievement ability is strong and the task saturation amount is high; and the forty-fifth to the fiftieth classified long-time and non-urgent to-be-delivered clues are matched to the object E whose clue tracking ability is strong and there is no idle in the current processing period, but there is a processing gap in the subsequent period and the object E is in the efficient processing period.

[0059] Since the above-mentioned ability image feature information can reflect the ability image features of each object in the to-be-issued group, and the above-mentioned processing state information can reflect the task quantity saturation degree, processing efficiency and processing effect in the specific task processing process of each object in the clue processing process; therefore, the issued matching information generated according to the above-mentioned ability image feature information and the above-mentioned processing state information is matched with the ability image and processing state of each object, which can improve the matching degree of the distributed clues and the distributed objects, and realize the clue distribution based on the supply-demand relationship, the individual processing state and the individual ability.

[0060] In step S240, based on the above-mentioned issued matching information, the above-mentioned to-be-issued clues are issued to the corresponding target objects.

[0061] The above-mentioned issued matching information can include the issued matching relationship between the to-be-issued clues and the corresponding target objects, and the issued time at each moment. By issuing the to-be-issued clues to the corresponding target objects based on the above-mentioned issued matching information, the clue distribution based on the supply-demand relationship, the individual processing state and the individual ability is realized, which helps to improve the follow-up efficiency and follow-up effect of the clues in the later stage.

[0062] Based on the above-mentioned steps S210-S240, by obtaining the historical clue issuing data and the historical clue processing data of each object in the to-be-issued group; according to the attribute information of each object, the above-mentioned historical clue issuing data and the above-mentioned historical clue processing data, the ability image feature information and the processing state information of each object are generated; since the above-mentioned ability image feature information can reflect the ability image features of each object in the to-be-issued group, and the above-mentioned processing state information can reflect the task quantity saturation degree, processing efficiency and processing effect in the specific task processing process of each object in the clue processing process; therefore, the issued matching information generated according to the above-mentioned ability image feature information and the above-mentioned processing state information is matched with the ability image and processing state of each object, which can improve the matching degree of the distributed clues and the distributed objects, and realize the clue distribution based on the supply-demand relationship, the individual processing state and the individual ability, which helps to improve the follow-up efficiency and follow-up effect of the clues in the later stage, and also effectively improves the intelligent degree of the clue distribution.

[0063] Figure 3 A detailed implementation flowchart of step S230 according to an embodiment of the present disclosure is schematically shown.

[0064] Referring to Figure 3 In the above-mentioned step S230, according to the above-mentioned ability image feature information and the above-mentioned processing state information, the issued matching information between the to-be-issued clues in the to-be-issued clue library and the above-mentioned to-be-issued group is generated, including the following steps: S310, S320, S330 and S340.

[0065] At step S310, according to the processing state information of each object in the to-be-issued group, the number of clues to be issued and the issuing time of each object in the to-be-issued group are determined.

[0066] In some embodiments, the processing state information includes processing supply and demand state and efficient processing time. The efficient processing time of each object is obtained according to the timing processing effect of the current object.

[0067] The number of clues to be issued matches the processing supply and demand state of the corresponding object, and the issuing time matches the efficient processing time of the corresponding object.

[0068] For example, if the processing supply and demand state of an object is that the supply of clues is greater than the demand of clues, the number of clues to be issued should be reduced or even not issued. If the processing supply and demand state of an object is that the supply of clues is less than the demand of clues, the number of clues to be issued should be increased. In some embodiments, the efficient processing time can be a time period or a time point, and the issuing time matching the efficient processing time of the corresponding object includes the following cases: when the to-be-issued clues are issued to the target object, the issuing can be performed before the efficient processing time period or time point of the target object, so that the target object can process the clues at the efficient processing time, improving the processing efficiency of the clues or the target achievement effect.

[0069] In other embodiments, the issuing time matching the efficient processing time of the corresponding object also includes the following cases: according to the urgency of the processing required by the clues and the efficient processing time of the corresponding object, the issuing time of the clues is determined, for example, for some clues that need to be processed urgently, if the efficient processing time of the corresponding object coincides with the time period of urgent processing, the issuing time can be determined within or before this time period, for example, the starting time of the time period. If the efficient processing time of the corresponding object deviates from the time period of urgent processing, the issuing time can be selected as a time close to the efficient processing time and meeting the requirement of urgent processing, so as to realize the compromise between the requirement of urgent processing of the clues and the processing mode of the processing object.

[0070] At step S320, according to the ability portrait feature information of each object in the to-be-issued group, a candidate object whose ability portrait matches the to-be-issued clues is determined.

[0071] According to an embodiment of the present disclosure, the ability portrait feature information includes at least one of the following: clue signing feature information, clue supply feature information, and signing target achievement feature information.

[0072] Figure 4 A detailed implementation flowchart of step S320 according to an embodiment of the present disclosure is schematically shown.

[0073] Referring to Figure 4As shown, in step S320, according to the ability portrait feature information of each object in the to-be-issued group, a candidate object whose ability portrait matches the to-be-issued clue is determined, including the following steps: S410, S420 and S430.

[0074] In step S410, an ability portrait meta-label in a preset dimension is generated according to the ability portrait feature information.

[0075] In some embodiments, in step S410, the ability portrait meta-label in the preset dimension is generated according to the ability portrait feature information, including: inputting the ability portrait feature information of each object into a pre-trained classification model, and outputting the corresponding ability portrait meta-label in the preset dimension of each object.

[0076] The preset dimension is greater than or equal to two dimensions. In the training phase, the training input of the classification model is the ability portrait feature information of the training sample object, and the training label is the real label of the ability portrait corresponding to the preset dimension of the training sample object.

[0077] The classification model is a deep learning model, which includes a feature embedding layer, a fully connected network layer and an activation layer. For example, in an embodiment, the ability portrait feature information is encoded by one-hot coding. In a specific embodiment, the vectorization of the ability portrait feature information includes the following processing example: for the actual hot clue amount feature of the sales, first, the data is grouped according to the distribution (grouping according to the distribution of the data), and then onehot coding is used to obtain the corresponding embedding vector in the vector pool according to the number.

[0078] The fully connected network layer is a 5-layer fully connected hidden layer network, which is used for nonlinear activation and feature cross. The activation layer uses a sigmoid function.

[0079] As an example, after the ability portrait feature information of object X is input into the classification model, a 12*1-dimensional ability portrait meta-label vector [1, 0, 0, 1, 1, 0, 1, 0, 1, 0, 0, 1] is output, and the elements in each dimension of the vector correspond to the following meta-label content in order: strong ability to sign large plates, weak ability to sign large plates, weak hot clue ability, strong hot clue ability, strong ability to achieve large plates, weak ability to achieve large plates, strong hot clue achievement ability, weak hot clue achievement ability, strong ability to supply large plates, weak ability to supply large plates, weak hot clue supply ability, and strong hot clue supply ability; the element value corresponding to the dimension is 1, indicating that the label in this dimension is hit. For example, the first element value is 1, indicating that the object has strong ability to sign large plates; the fourth element value is 1, indicating that the object has strong hot clue ability.

[0080] Each element in the ability portrait meta-label vector corresponds to one dimension of ability meta-label data. Whether the label in this dimension is hit or not, if hit, the value is 1, and if not hit, the value is 0. The composite label can be obtained by combining the ability meta-label data of each dimension. It can be understood that the dimensions of the corresponding ability portrait meta-label vector, the content of each element, etc. can be adjusted and changed according to the differences of the scene and the required ability of the clue.

[0081] In some embodiments, in order to improve the classification accuracy of the classification model, the above classification model and input are optimized, and the above optimization process includes three parts.

[0082] The first part is to optimize the model parameters, mainly including the selection of the maximum number of iterations, the selection of the activation function, the selection of the convergence method, the setting of the convergence threshold, etc.

[0083] The second part is to pre-optimize the input ability portrait feature information, which can mainly be optimized for processing rate, signing rate, etc. Since the signing rate = the number of personal subscriptions / the number of issued quantities ( / represents the mathematical operation of division), too small or 0 value of the issued quantity will reduce the confidence, therefore, the Bayesian smoothing is introduced to adjust the signing rate.

[0084] According to the embodiments of the present disclosure, the above clue signing feature information includes: signing rate data, processing rate data, and at least one of the above signing rate data and processing rate data is a result processed by Bayesian smoothing. Taking the signing rate data as an example, the Bayesian smoothing processing includes: calculating the overall mean and variance of the initial signing rate data (which can be replaced by the initial processing rate data) of a same object for multiple types of clues, and adjusting the initial signing rate data by taking the overall mean and variance as the prior probability to obtain the adjusted signing rate data. Similarly, the processing rate can also be optimized, and the processing method is the same as that of the signing rate.

[0085] The third part is that the deep learning model needs a large amount of data in the training stage. If the data amount is insufficient, the feature is complex, and the overfitting problem may occur. The overfitting phenomenon on the training set will reduce the generalization ability of the model. When the test set tests the model, the experimental results are very different from the training, and the model has no practical application value. In order to avoid the overfitting of the model, in the process of actually training the above classification model, the data amount of the training sample is expanded, the data of the classification model is made more complete, and the generalization ability of the model is enhanced.

[0086] In step S420, the ability portrait meta-label is combined to obtain a composite label according to the ability combination information required by the to-be-issued clue.

[0087] For example, the capability combination information required for the clue M to be issued is: strong market contracting capability and strong market supply capability, then the composite label obtained by the combination is: [1,0,0,0,0,0,0,0,1,0,0,0].

[0088] In step S430 , candidate objects matching the composite tag are determined.

[0089] Based on each object's capability profile feature information, the capability profile meta-tag vector corresponding to the capability profile meta-tag under the preset dimension is obtained. By matching the capability profile meta-tag vector with the composite tag, candidate objects whose capabilities meet the requirements of the lead to be issued are obtained. For example, the capability profile meta-tag vector of object X described above, [1,0,0,1,1,0,1,0,1,0,0,1], matches the composite tag [1,0,0,0,0,0,0,0,1,0,0,0], making object X a candidate for lead M to be issued.

[0090] In step S330, based on the number of leads to be issued, the number of leads to be issued that are matched by the candidate objects is adjusted to obtain an issuing matching relationship between the leads to be issued and the candidate objects.

[0091] Figure 5 The detailed implementation flow chart of step S330 according to an embodiment of the present disclosure is schematically shown.

[0092] According to the embodiment of the present disclosure, referring to Figure 5 As shown, in step S330, based on the number of leads issued, the number of leads matched to the candidate objects to be issued is adjusted to obtain an issuance matching relationship between the leads to be issued and the candidate objects, including the following steps: S510 and S521. In other embodiments, at least one of the following steps: {S522a or S522b} and S523 may also be included. It is understood that in an embodiment, only one of S521, {S522a or S522b}, and S523 may exist, or multiple of these may exist simultaneously.

[0093] In step S510, the relative size between the number of leads to be sent matched by the candidate object and the number of leads sent is determined.

[0094] The above-mentioned number of leads issued serves as indication information of the supply and demand status of the candidate objects. Based on this indication information as a reference for the number of leads issued, there will be a relative size between the number of leads to be issued and the corresponding number of leads to be issued obtained by matching the required capabilities of the leads to be issued and the above-mentioned number of leads issued. Accordingly, different strategies are adopted to construct issuance matching relationships under different size relationships.

[0095] In step S521, for the first candidate object whose to-be-delivered lead quantity is the same as the corresponding lead delivery quantity, a first delivery matching relationship between the first candidate object and the matched to-be-delivered lead is constructed.

[0096] After the first delivery matching relationship is constructed and the lead is successfully delivered, the processing state information of the first candidate object is dynamically updated.

[0097] In order to simplify the description subsequently, Y is used to represent the to-be-delivered lead quantity, and Z is used to represent the lead delivery quantity.

[0098] For the second candidate object whose to-be-delivered lead quantity is less than the corresponding lead delivery quantity, there are two construction strategies.

[0099] In step S522a, for the second candidate object whose to-be-delivered lead quantity Y is less than the corresponding lead delivery quantity Z, a second delivery matching relationship between the second candidate object and the matched to-be-delivered lead is constructed.

[0100] In the embodiment containing step S522a, for the case that the number Y of the matched to-be-delivered lead is less than the reference quantity Z of the supply-demand relationship, the second delivery matching relationship can be constructed according to the actual matched quantity, for example, the second candidate object matches 5 to-be-delivered leads, the lead delivery quantity of the second candidate object is 10, and there are 5 vacancies, in which case the second delivery matching relationship between the second candidate object and the 5 to-be-delivered leads is first constructed, and the processing state information of the second candidate object is dynamically updated after the lead is successfully delivered. After subsequent new matching, the corresponding delivery matching relationship is generated according to the newly matched to-be-delivered lead.

[0101] Alternatively, in step S522b, for the second candidate object whose to-be-delivered lead quantity is less than the corresponding lead delivery quantity, a third delivery matching relationship between the second candidate object and a first target to-be-delivered lead is constructed by allocating a supplementary lead to the second candidate object from other to-be-delivered leads. The first target to-be-delivered lead includes the matched to-be-delivered lead and the supplementary lead, and the supplementary lead is used to supplement the lead delivery quantity of the second candidate object.

[0102] In the embodiment containing step S522b, for the case that the number Y of the matched to-be-delivered lead is less than the reference quantity Z of the supply-demand relationship, the supplementary lead can be allocated to the second candidate object, so that the actual to-be-delivered lead quantity of the second candidate object is equal to the reference quantity Z, that is, by increasing the number δ of the supplementary lead on the basis of Y, Y+δ=Z, in which case the third delivery matching relationship between the second candidate object and the matched to-be-delivered lead and the supplementary lead is constructed, and the processing state information of the second candidate object is dynamically updated after the lead is successfully delivered.

[0103] In step S523, for the third candidate object whose to-be-delivered clue amount is greater than the corresponding clue delivery quantity, the batch division of the to-be-delivered clues is performed according to the matched tracking priority information of the to-be-delivered clues, to obtain a first batch of clues with higher tracking priority and whose clue amount meets the above-mentioned clue delivery quantity, and a fourth delivery matching relationship between the third candidate object and the first batch of clues is constructed.

[0104] The to-be-delivered clues of the subsequent batch are to-be-delivered clues of the subsequent period as the third candidate object or are complementary clues of the second candidate object.

[0105] The tracking priority information includes, for example, tracking urgency, relative importance of clues, and required ability of clues. For example, the higher the tracking urgency, the higher the corresponding priority; the more important the relative importance of clues, the higher the corresponding priority; the higher the required ability of clues, the higher the corresponding priority. The priority can also be divided by comprehensively considering the above-mentioned multiple dimensions, for example, the higher the tracking urgency and the higher the relative importance of clues, the higher the priority. The higher the required ability of clues and the higher the tracking urgency, the higher the corresponding priority.

[0106] In the embodiment including step S523, for the case that the number Y of the matched to-be-delivered clues is greater than the reference number Z of the supply-demand relationship, the batch division of the to-be-delivered clues is performed according to the tracking priority information of the to-be-delivered clues, for example, to divide into a first batch of clues with a number of Z, and the total number of the subsequent batches is Y-Z.

[0107] In step S340, the delivery matching information is generated according to the above-mentioned delivery matching relationship and the above-mentioned delivery time.

[0108] The delivery matching information includes the delivery matching relationship and the delivery time.

[0109] In the embodiment including steps S310-S340, since the clue delivery quantity is used as the indication information of the supply-demand state of the candidate object, the to-be-delivered clue amount matched by the candidate object is adjusted based on the indication information as a reference of the delivery quantity, to obtain the delivery matching relationship between the to-be-delivered clues and the candidate object, which can dynamically adapt to the supply-demand state of each candidate object and can take into account the processing progress of the to-be-delivered clues.

[0110] A second exemplary embodiment of the present disclosure provides a clue delivery device.

[0111] Figure 6 The structure block diagram of the clue delivery device according to the embodiment of the present disclosure is schematically shown.

[0112] Reference Figure 6As shown, the device 600 for issuing a clue includes a data acquisition module 601, a feature information generation module 602, a matching module 603, and an issuing module 604.

[0113] The data acquisition module 601 is configured to acquire historical clue issuing data and historical clue processing data of each object in a to-be-issued group.

[0114] The feature information generation module 602 is configured to generate capability portrait feature information and processing state information of each object according to attribute information of each object, the historical clue issuing data, and the historical clue processing data.

[0115] The matching module 603 is configured to generate issuing matching information between a to-be-issued clue in a to-be-issued clue library and the to-be-issued group according to the capability portrait feature information and the processing state information.

[0116] The issuing module 604 is configured to issue the to-be-issued clue to a corresponding target object based on the issuing matching information.

[0117] According to an embodiment of the present disclosure, the generating of the issuing matching information between the to-be-issued clue in the to-be-issued clue library and the to-be-issued group according to the capability portrait feature information and the processing state information includes: determining a clue issuing quantity and an issuing time of each object in the to-be-issued group according to the processing state information of each object in the to-be-issued group; determining a candidate object whose capability portrait matches the to-be-issued clue according to the capability portrait feature information of each object in the to-be-issued group; performing an adjustment processing on a to-be-issued clue quantity matched by the candidate object according to the clue issuing quantity, to obtain an issuing matching relationship between the to-be-issued clue and the candidate object; and generating the issuing matching information according to the issuing matching relationship and the issuing time.

[0118] According to an embodiment of the present disclosure, the quantity of the to-be-issued clues matched by the candidate object is adjusted according to the number of the clues, to obtain an issuing matching relationship between the to-be-issued clues and the candidate object, including: determining the relative size between the quantity of the to-be-issued clues matched by the candidate object and the number of the clues; for a first candidate object with the same quantity of to-be-issued clues and corresponding number of clues, a first issuing matching relationship between the first candidate object and the matched to-be-issued clues is constructed. For a second candidate object with a quantity of to-be-issued clues less than the corresponding number of clues, a second issuing matching relationship between the second candidate object and the matched to-be-issued clues is constructed; or, the second candidate object is supplemented with clues from other to-be-issued clues, to construct a third issuing matching relationship between the second candidate object and a first target to-be-issued clue; the first target to-be-issued clue includes the matched to-be-issued clue and the supplemented clue, and the supplemented clue is used to supplement the number of clues of the second candidate object. For a third candidate object with a quantity of to-be-issued clues greater than the corresponding number of clues, the to-be-issued clues are batched according to the tracking priority information of the matched to-be-issued clues, to obtain a first batch of clues with higher tracking priority and a quantity of clues satisfying the number of clues, and a fourth issuing matching relationship between the third candidate object and the first batch of clues is constructed; the to-be-issued clues of the subsequent batch are used as the to-be-issued clues of the third candidate object in the subsequent period or as the supplemented clues of the second candidate object.

[0119] According to an embodiment of the present disclosure, the processing state information includes: processing supply and demand state and efficient processing time; the efficient processing time of each object is obtained by counting the time sequence processing effect of the current object. Wherein, the number of clues matches the processing supply and demand state of the corresponding object; the issuing time matches the efficient processing time of the corresponding object.

[0120] According to an embodiment of the present disclosure, the capability portrait feature information covers at least one of: clue signing feature information, clue supply feature information, and signing target achievement feature information. According to the capability portrait feature information of each object in the to-be-issued group, the candidate object whose capability portrait matches the to-be-issued clue is determined, including: generating a capability portrait meta-label in a preset dimension according to the capability portrait feature information; combining the capability portrait meta-label to obtain a composite label according to the capability combination information required by the to-be-issued clue; determining the candidate object matched with the composite label.

[0121] According to an embodiment of the present disclosure, the ability portrait feature information of each object is input into a pre-trained classification model to obtain the ability portrait meta-label of each object in the preset dimension.

[0122] According to an embodiment of the present disclosure, the clue signing feature information includes signing rate data and processing rate data, and at least one of the signing rate data or the processing rate data is a result processed by Bayesian smoothing. The processing by Bayesian smoothing includes: calculating the overall mean and variance of initial data of a same object for multiple types of clues, and adjusting the initial data by using the overall mean and variance as prior probability to obtain adjusted data; and the initial data includes at least one of the signing rate data or the processing rate data.

[0123] More details and benefits of the present embodiment can be referred to the description of the first embodiment, which will not be repeated here.

[0124] Any of the plurality of functional modules included in the apparatus 600 can be combined in one module, or any of the modules can be split into a plurality of modules. Alternatively, at least part of the function of one or more of the modules can be combined with at least part of the function of another module, and implemented in one module. At least one of the functional modules included in the apparatus 600 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on board, a system in package, an application specific integrated circuit (ASIC), or any other reasonable hardware or firmware that can be integrated or packaged, or implemented in any one of software, hardware and firmware or in any appropriate combination of any of them. Alternatively, at least one of the functional modules included in the apparatus 600 can be at least partially implemented as a computer program module that can perform corresponding functions when executed.

[0125] A third exemplary embodiment of the present disclosure provides an electronic device.

[0126] Figure 7 A structural block diagram of an electronic device provided by an embodiment of the present disclosure is schematically shown.

[0127] Reference Figure 7As shown, the electronic device 700 provided by the embodiment of the present disclosure includes a processor 701, a communication interface 702, a memory 703 and a communication bus 704, wherein the processor 701, the communication interface 702 and the memory 703 complete the communication among each other through the communication bus 704; the memory 703 is used for storing a computer program; and the processor 701 is used for executing the program stored on the memory to implement the method for distributing clues as described above.

[0128] The fourth exemplary embodiment of the present disclosure also provides a computer readable storage medium. The computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method for distributing clues as described above.

[0129] The computer readable storage medium can be included in the device or apparatus described in the above embodiments; or can exist independently without being assembled into the device or apparatus. The computer readable storage medium carries one or more programs, and the one or more programs are executed to implement the method according to the embodiments of the present disclosure.

[0130] According to the embodiments of the present disclosure, the computer readable storage medium can be a non-volatile computer readable storage medium, which can include but is not limited to: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any appropriate combination thereof. In the present disclosure, the computer readable storage medium can be any tangible medium that contains or stores a program, which can be used by or in connection with an instruction execution system, apparatus or device.

[0131] It should be noted that, in the technical solutions provided by the embodiments of the present disclosure, the collection, collection, update, analysis, processing, use, transmission and storage of user personal information are in line with the relevant legal regulations, are used for legal purposes, and do not violate public order and good customs. Necessary measures are taken for user personal information to prevent illegal access to user personal information data, and to maintain user personal information security, network security and national security.

[0132] It has to be noted that, in the present document, relational terms are intended only to convey a possible relationship between elements or

[0133] The above description is merely that of the specific embodiments of the present disclosure and therefore is not intended to limit the present disclosure. Various modifications made to the embodiments of the present disclosure will be apparent to those skilled in the art to which the present disclosure pertains, and such modifications are not to be interpreted within the scope or spirit of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments shown herein but will be construed to include all the modifications, equivalents, and alternatives falling within the scope and spirit of the present disclosure.

Claims

1. A method for lead distribution, characterized in that, The method comprises the following steps: acquiring historical clue distribution data and historical clue processing data of each object in a to-be-distributed group; generating capability portrait feature information and processing state information of each object according to attribute information of each object, the historical clue distribution data and the historical clue processing data; generating distribution matching information between a to-be-distributed clue in a to-be-distributed clue library and the to-be-distributed group according to the capability portrait feature information and the processing state information; distributing the to-be-distributed clue to a corresponding target object based on the distribution matching information.

2. The method of claim 1, wherein, The method comprises the following steps: determining a clue distribution quantity and a distribution time of each object in the to-be-distributed group according to processing state information of each object in the to-be-distributed group; determining a candidate object whose capability portrait matches the to-be-distributed clue according to capability portrait feature information of each object in the to-be-distributed group; performing adjustment processing on a to-be-distributed clue quantity matched by the candidate object according to the clue distribution quantity, to obtain a distribution matching relationship between the to-be-distributed clue and the candidate object; generating the distribution matching information according to the distribution matching relationship and the distribution time.

3. The method of claim 2, wherein, The method comprises the following steps: determining a relative size between the to-be-distributed clue quantity matched by the candidate object and the clue distribution quantity; constructing a first distribution matching relationship between the first candidate object and the matched to-be-distributed clue for a first candidate object whose to-be-distributed clue quantity is the same as the corresponding clue distribution quantity; constructing a second distribution matching relationship between the second candidate object and the matched to-be-distributed clue for a second candidate object whose to-be-distributed clue quantity is less than the corresponding clue distribution quantity, or adjusting a supplementary clue from other to-be-distributed clues to the second candidate object to construct a third distribution matching relationship between the second candidate object and a first target to-be-distributed clue; the first target to-be-distributed clue comprises the matched to-be-distributed clue and the supplementary clue, and the supplementary clue is used to supplement the clue distribution quantity of the second candidate object; for a third candidate object whose to-be-distributed clue quantity is greater than the corresponding clue distribution quantity, performing batch division on the to-be-distributed clue according to tracking priority information of the matched to-be-distributed clue to obtain a first batch clue whose tracking priority is high and whose clue quantity meets the clue distribution quantity, and constructing a fourth distribution matching relationship between the third candidate object and the first batch clue; subsequent batch to-be-distributed clues are to-be-distributed clues of the third candidate object in a subsequent period or are supplementary clues of the second candidate object.

4. The method of claim 2, wherein, The processing state information comprises a processing supply-demand state and an efficient processing time; the efficient processing time of each object is obtained by statistical processing according to a time sequence processing effect of the current object. The number of the clues is matched with the processing supply-demand state of the corresponding object. The delivery time is matched with the efficient processing time of the corresponding object.

5. The method of claim 2, wherein, The capability profile feature information covers at least one of the following: clue signing feature information, clue supply feature information, and signing target achievement feature information. According to the capability profile feature information of each object in the to-be-delivered group, a candidate object whose capability profile matches the to-be-delivered clue is determined, including: According to the capability profile feature information, a capability profile meta-label under a preset dimension is generated. According to the capability combination information required by the to-be-delivered clue, the capability profile meta-label is combined to obtain a composite label. A candidate object matching the composite label is determined.

6. The method of claim 5, wherein, According to the capability profile feature information, a capability profile meta-label under a preset dimension is generated, including: The capability profile feature information of each object is input into a pre-trained classification model, and a capability profile meta-label under a preset dimension corresponding to each object is output; the preset dimension is greater than or equal to two dimensions, and in the training stage, the training input of the classification model is the capability profile feature information of the training sample object, and the training label is the capability profile real label corresponding to the training sample object under the preset dimension.

7. The method of claim 5, wherein, The clue signing feature information includes signing rate data and processing rate data, and at least one of the signing rate data or the processing rate data is a result processed by Bayesian smoothing. The Bayesian smoothing processing includes: calculating the overall mean and variance of the initial data of the same object for multiple types of clues, and adjusting the initial data by taking the overall mean and variance as the prior probability to obtain adjusted data; the initial data includes at least one of the signing rate data or the processing rate data.

8. An apparatus for issuing a clue, the apparatus comprising: It includes: A data acquisition module is configured to acquire historical clue delivery data and historical clue processing data of each object in a to-be-delivered group; A feature information generation module is configured to generate capability profile feature information and processing state information of each object according to attribute information, the historical clue delivery data, and the historical clue processing data of each object; A matching module is configured to generate delivery matching information between a to-be-delivered clue in a to-be-delivered clue library and the to-be-delivered group according to the capability profile feature information and the processing state information; A delivery module is configured to deliver the to-be-delivered clue to a corresponding target object based on the delivery matching information.

9. An electronic device, comprising: It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory complete communication with each other through the communication bus; The memory is used to store a computer program; The processor is configured to execute the program stored on the memory to implement the method in any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the method in any one of claims 1-7.

Images