Method and device for determining key breakpoints in telemarketing

Abstract

The present invention discloses a method and device for determining a critical breakpoint of telephone marketing, which can be used in the financial field or other technical fields. The method comprises: obtaining case information of multiple marketing failure cases; establishing a marketing breakpoint knowledge graph according to the case information, wherein the marketing breakpoint knowledge graph comprises multiple nodes and edges for connecting nodes, and the nodes comprise: marketing scenario nodes, customer nodes, and speech template type nodes; obtaining the number of edges connected to each speech template type node from the marketing breakpoint knowledge graph; using speech template type nodes whose number is greater than a critical breakpoint determination threshold as marketing critical breakpoint nodes, and using the speech template type corresponding to the marketing critical breakpoint node as the marketing critical breakpoint speech template type. The present invention is helpful to improve speech templates for telephone marketing.

Description

Technical Field

[0001] The present invention relates to the field of knowledge graph technology, and in particular to a method and device for determining key breakpoints in telemarketing. Background Art

[0002] In recent years, with the development of Internet social media and mobile finance, people's demand for finance has also increased, showing diversified and personalized characteristics. Among them, telephone marketing has become a key marketing method for banks, such as manual outbound calls in marketing scenarios, and robot intelligent outbound calls realized by financial technology artificial intelligence technology to make up for the lack of manpower.

[0003] During telephone marketing, customer service selects appropriate speech templates to communicate with customers according to the marketing scenario, and these preset speech templates include various types, such as: opening remarks, product benefit introduction, handling method introduction, customer service first customer recovery, customer service second customer recovery, etc. When customer service communicates with customers, user rejection may occur in any speech template, for example: customer service opening introduction Hello! I am a customer service representative of XX Bank, the customer hangs up the phone, that is, the customer rejected the speech template of the opening remarks type, then the breakpoint of this marketing is the speech template of the "opening remarks" type.

[0004] Since user rejection may occur on any speech template, that is, any speech template may be a breakpoint in marketing, how to analyze the breakpoints, determine the key breakpoints in marketing, and improve the speech templates is a technical problem that urgently needs to be solved in this field. Summary of the invention

[0005] In order to solve the technical problems in the above-mentioned background technology, the present invention proposes a method and device for determining key breakpoints in telephone marketing.

[0006] In order to achieve the above object, according to one aspect of the present invention, a method for determining a key breakpoint in telemarketing is provided, the method comprising:

[0007] Obtain case information of multiple marketing failure cases, wherein the case information includes: marketing scenario information, customer information, and speech template types corresponding to marketing breakpoints;

[0008] A marketing breakpoint knowledge graph is established according to the case information, wherein the marketing breakpoint knowledge graph includes a plurality of nodes and edges for connecting the nodes, the nodes include: marketing scenario nodes, customer nodes, and speech template type nodes, and the edges are used to connect the marketing scenario nodes, customer nodes, and speech template type nodes belonging to the same case information;

[0009] Obtaining the number of edges connected to each speech template type node from the marketing breakpoint knowledge graph;

[0010] The speech template type nodes whose number is greater than the key breakpoint determination threshold are taken as marketing key breakpoint nodes, and the speech template type corresponding to the marketing key breakpoint nodes is taken as the marketing key breakpoint speech template type.

[0011] Optionally, the method for determining the key breakpoint of telemarketing further includes:

[0012] Obtaining customer service and customer dialogue text information corresponding to the marketing failure case, wherein the customer service and customer dialogue text information includes: customer service input text information and customer input text information, and the customer service input text information includes multiple types of speech templates;

[0013] Determine the speech template type corresponding to each customer service input segment in the customer service input text information;

[0014] The customer service input segment corresponding to the marketing breakpoint is determined according to the customer input text information, and then the speech template type corresponding to the marketing breakpoint is obtained.

[0015] Optionally, determining the speech template type corresponding to each customer service input segment in the customer service input text information specifically includes:

[0016] Input each customer service input segment in the customer service input text information into a preset machine learning model to obtain the speech template type corresponding to each customer service input segment, wherein the machine learning model is trained using speech templates marked with speech template types as training data.

[0017] Optionally, the method for determining the key breakpoint of telemarketing further includes:

[0018] The critical breakpoint determination threshold is determined according to the number of the marketing failure cases.

[0019] Optionally, the method for determining the key breakpoint of telemarketing further includes:

[0020] For each of the marketing key breakpoint nodes, the PageRank value of each customer node connected to the marketing key breakpoint node is calculated by using the PageRank algorithm;

[0021] The customer portrait corresponding to the customer node with the largest PageRank value is used as the customer portrait template of the marketing key breakpoint speech template type corresponding to the marketing key breakpoint node.

[0022] Optionally, the method for determining the key breakpoint of telemarketing further includes:

[0023] Obtain the marketing key breakpoint speech template type corresponding to the speech template in each marketing scenario;

[0024] Mark the key marketing breakpoint labels on the corresponding speech template;

[0025] The customer portrait template is marked on the corresponding speech template.

[0026] Optionally, the method for determining the key breakpoint of telemarketing further includes:

[0027] Obtaining audio data of the conversation between customer service and the customer corresponding to the marketing failure case;

[0028] Based on the preset speech recognition model and the audio data of the conversation between the customer service and the customer, the text information of the conversation between the customer service and the customer corresponding to the marketing failure case is generated.

[0029] In order to achieve the above object, according to another aspect of the present invention, a device for determining a key breakpoint in telemarketing is provided, the device comprising:

[0030] A case information acquisition module, used to acquire case information of multiple marketing failure cases, wherein the case information includes: marketing scenario information, customer information, and speech template types corresponding to marketing breakpoints;

[0031] A marketing breakpoint knowledge graph establishment module is used to establish a marketing breakpoint knowledge graph according to the case information, wherein the marketing breakpoint knowledge graph includes a plurality of nodes and edges for connecting the nodes, the nodes include: marketing scenario nodes, customer nodes and speech template type nodes, and the edges are used to connect the marketing scenario nodes, customer nodes and speech template type nodes belonging to the same case information;

[0032] A quantity statistics module, used to obtain the number of edges connected to each speech template type node from the marketing breakpoint knowledge graph;

[0033] The marketing key breakpoint speech template type determination module is used to take the speech template type nodes whose number is greater than the key breakpoint determination threshold as the marketing key breakpoint nodes, and take the speech template type corresponding to the marketing key breakpoint nodes as the marketing key breakpoint speech template type.

[0034] Optionally, the telephone marketing key breakpoint determination device further includes:

[0035] A customer service and customer dialogue text information acquisition module is used to acquire customer service and customer dialogue text information corresponding to the marketing failure case, wherein the customer service and customer dialogue text information includes: customer service input text information and customer input text information, and the customer service input text information includes multiple types of speech templates;

[0036] A speech template type identification module, used to determine the speech template type corresponding to each customer service input segment in the customer service input text information;

[0037] A module for determining the speech template type corresponding to a marketing breakpoint is used to determine the customer service input segment corresponding to the marketing breakpoint based on the customer input text information, and then obtain the speech template type corresponding to the marketing breakpoint.

[0038] Optionally, the speech template type identification module specifically includes:

[0039] A machine learning unit is used to input each customer service input segment in the customer service input text information into a preset machine learning model to obtain a speech template type corresponding to each customer service input segment, wherein the machine learning model is trained using speech templates marked with speech template types as training data.

[0040] In order to achieve the above-mentioned purpose, according to another aspect of the present invention, a computer device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps in the above-mentioned method for determining key breakpoints in telemarketing when executing the computer program.

[0041] In order to achieve the above-mentioned purpose, according to another aspect of the present invention, a computer-readable storage medium is further provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed in a computer processor, the steps in the above-mentioned method for determining key breakpoints in telemarketing are implemented.

[0042] The beneficial effects of the present invention are as follows: the present invention establishes a marketing breakpoint knowledge graph, wherein the marketing breakpoint knowledge graph includes marketing scenario nodes, customer nodes, and speech template type nodes, and determines the marketing key breakpoint speech template type according to the number of edges of each speech template type node in the marketing breakpoint knowledge graph, which helps backend operation personnel to modify or adjust the corresponding type of speech template according to the determined marketing key breakpoint speech template type, thereby helping to improve the quality of the speech template. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0044] Figure 1is a first flow chart of a method for determining a key breakpoint in telemarketing according to an embodiment of the present invention;

[0045] Figure 2 is a second flow chart of the method for determining key breakpoints in telemarketing according to an embodiment of the present invention;

[0046] Figure 3 is a third flow chart of the method for determining key breakpoints in telemarketing according to an embodiment of the present invention;

[0047] Figure 4 is a fourth flow chart of the method for determining key breakpoints in telemarketing according to an embodiment of the present invention;

[0048] Figure 5 This is a schematic diagram of a marketing breakpoint knowledge graph according to an embodiment of the present invention;

[0049] Figure 6 This is a first structural block diagram of a device for determining a key breakpoint in telemarketing according to an embodiment of the present invention;

[0050] Figure 7 This is a second structural block diagram of the apparatus for determining a critical breakpoint in telemarketing according to an embodiment of the present invention;

[0051] Figure 8 It is a schematic diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0053] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0054] It should be noted that the terms "including" and "having" and any variations thereof in the specification and claims of the present invention and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or apparatus.

[0055] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0056] It should be noted that the method and device for determining the key breakpoints of telemarketing of the present invention can be used in the financial field, and can also be applied to other technical fields.

[0057] The purpose of this invention is to construct a marketing breakpoint knowledge graph, find key breakpoints and analyze them through the association relationship between nodes in the knowledge graph and use important node analysis methods, so as to improve the dialogue template and reduce the occurrence of marketing breakpoints.

[0058] Figure 1 is a first flow chart of the method for determining the key breakpoints of telemarketing according to an embodiment of the present invention. Figure 1 As shown, in one embodiment of the present invention, the method for determining key breakpoints in telemarketing of the present invention includes steps S101 to S104.

[0059] Step S101, obtaining case information of multiple marketing failure cases, wherein the case information includes: marketing scenario information, customer information, and speech template types corresponding to marketing breakpoints.

[0060] In one embodiment of the present invention, the present invention completes the construction of manual outbound calling and intelligent outbound calling scenarios and the design of scenario scripts through outbound calling scenario construction, and constructs a marketing scenario script library, the contents of which include: marketing scenarios and script templates corresponding to each marketing scenario.

[0061] In one embodiment of the present invention, after the customer service outbound call is completed, the marketing results are counted, and the marketing results include: success, failure, and pending. The present invention obtains the audio data of the conversation between the customer service and the customer whose outbound call result is marketing failure, and generates case information of the marketing failure case based on the conversation audio data.

[0062] In one embodiment of the present invention, the speech templates of the present invention include multiple types, such as: opening introduction, product benefit introduction, handling method introduction, customer service first customer recovery, customer service second customer recovery, and closing remarks, etc. When the customer service communicates with the customer, user rejection behavior may occur in any speech template, for example: customer service opening introduction Hello! I am a customer service representative of XX Bank, the customer hangs up the phone, that is, the customer rejects the speech template of the opening remarks type, and the breakpoint of this marketing is the speech template of the "opening remarks" type.

[0063] In one embodiment of the present invention, each marketing scenario corresponds to a plurality of speech templates, and the plurality of speech templates may be of various types, such as the following Table 1.

[0064]

[0065] Table 1

[0066] In one embodiment of the present invention, the speech template type corresponding to the marketing breakpoint in the case information of the marketing failure case is first determined by determining in which speech template the customer made the rejection action in the marketing failure case, and then obtaining the speech template type of the speech template in which the customer issued the rejection action, which is the speech template type corresponding to the breakpoint.

[0067] Step S102, establishing a marketing breakpoint knowledge graph based on the case information, wherein the marketing breakpoint knowledge graph includes multiple nodes and edges for connecting the nodes, the nodes include: marketing scenario nodes, customer nodes and speech template type nodes, and the edges are used to connect marketing scenario nodes, customer nodes and speech template type nodes belonging to the same case information.

[0068] It should be noted that the speech template type node in the marketing breakpoint knowledge graph in this step is the speech template type node corresponding to the marketing breakpoint in the above step S101.

[0069] Figure 5 is a schematic diagram of a marketing breakpoint knowledge graph according to an embodiment of the present invention. Figure 5 As shown, the marketing breakpoint knowledge graph includes multiple nodes, and the marketing scenario nodes, customer nodes, and speech template type nodes belonging to the same case information are connected through edges. Specifically, the marketing scenario nodes, customer nodes, and speech template type nodes belonging to the same case information are connected in pairs through edges. In one embodiment of the present invention, the marketing scenario nodes, customer nodes, and speech template type nodes can all belong to multiple different case information.

[0070] Step S103, obtaining the number of edges connected to each speech template type node from the marketing breakpoint knowledge graph.

[0071] In one embodiment of the present invention, the present invention sequentially analyzes the speech template type nodes in the marketing breakpoint knowledge graph, and the specific steps are as follows: using the network key node calculation method, the speech template type nodes in the knowledge graph are analyzed, first the node degree is calculated according to the edge of the node in the knowledge graph, and the speech template type nodes are sorted according to the node degree. For example: the template type opening introduction has the most edges connected as a speech template type node, indicating that this speech template type has the highest frequency of breakpoints and is a key speech template type that affects marketing.

[0072] Step S104, taking the speech template type nodes whose quantity is greater than the critical breakpoint determination threshold as marketing critical breakpoint nodes, and taking the speech template type corresponding to the marketing critical breakpoint nodes as the marketing critical breakpoint speech template type.

[0073] In one embodiment of the present invention, the critical breakpoint determination threshold may be determined based on the number of marketing failure cases.

[0074] Figure 2 is a second flow chart of the method for determining key breakpoints in telemarketing according to an embodiment of the present invention. Figure 2 As shown, in one embodiment of the present invention, the speech template type corresponding to the marketing breakpoint in the case information of the above step S101 is specifically obtained through the following steps S201 to S203.

[0075] Step S201, obtaining the text information of the conversation between the customer service and the customer corresponding to the marketing failure case, wherein the text information of the conversation between the customer service and the customer includes: customer service input text information and customer input text information, and the customer service input text information includes multiple types of speech templates.

[0076] In one embodiment of the present invention, the text information of the conversation between customer service and the customer in this step is specifically obtained by the following steps: obtaining the audio data of the conversation between customer service and the customer corresponding to the marketing failure case; generating the text information of the conversation between customer service and the customer corresponding to the marketing failure case based on a preset speech recognition model and the audio data of the conversation between customer service and the customer.

[0077] Step S202: determine the speech template type corresponding to each customer service input segment in the customer service input text information.

[0078] In one embodiment of the present invention, this step may specifically be inputting each customer service input segment in the customer service input text information into a preset machine learning model to obtain a speech template type corresponding to each customer service input segment, wherein the machine learning model is trained using speech templates labeled with speech template types as training data.

[0079] In one embodiment of the present invention, each customer service input segment in the customer service input text information can be regarded as an entity, and each customer service input segment is classified through natural language processing text classification technology. The specific steps are as follows: Text classification is divided into two stages, the first stage is machine learning model training, and the second stage is the prediction stage. The first stage is the training stage: the speech template that has been labeled with the speech template type label is used as the training data, and the text content is first segmented, and then the feature extraction of the segmented data is performed and converted into the corresponding feature vector. Finally, the training data is trained through the machine learning supervised learning algorithm to obtain the machine learning model. The second stage is the prediction stage: each customer service input segment in the customer service input text information is successively segmented, feature extracted and converted into a feature vector, and finally the machine learning model is used for prediction to obtain the speech template type of each customer service input segment.

[0080] Step S203, determining the customer service input segment corresponding to the marketing breakpoint based on the customer input text information, and then obtaining the speech template type corresponding to the marketing breakpoint.

[0081] In an embodiment of the present invention, this step may first determine the breakpoint position in the conversation according to the customer input text information, and then determine the customer service input segment corresponding to the breakpoint position in the customer service input text information.

[0082] Figure 3 is a third flow chart of the method for determining the key breakpoint of telemarketing according to an embodiment of the present invention. Figure 3 As shown, in one embodiment of the present invention, the method for determining key breakpoints in telemarketing of the present invention further includes step S301 and step S302.

[0083] Step S301, for each of the marketing key breakpoint nodes, the PageRank value of each customer node connected to the marketing key breakpoint node is calculated by using the PageRank algorithm.

[0084] Step S302, using the customer portrait corresponding to the customer node with the largest PageRank value as the customer portrait template of the marketing key breakpoint speech template type corresponding to the marketing key breakpoint node.

[0085] In one embodiment of the present invention, the customer portrait includes information such as age, educational background, income level, customer asset registration, customer work unit and place of origin.

[0086] In one embodiment of the present invention, the present invention analyzes each customer node connected to the key marketing breakpoint node, calculates the PR value of each customer node through the PageRank algorithm, and finds the customer node with the greatest contribution to the key marketing breakpoint node as the key customer node. For example: the customer who refuses marketing in the speech template type "opening introduction" is the customer node of the key marketing breakpoint node "opening introduction". The PageRank value of this customer node is the largest through PR value calculation, and the contribution of this customer node is the largest. It can be considered that this customer is a representative of a group of customers, and the characteristics of this user need to be deeply analyzed. Analysis points: According to the entity attribute similarity analysis, the common characteristics of the customer and the related customers, such as: the same asset level, the same work unit, etc. Therefore, the cause of the breakpoint can be analyzed from the perspective of customer characteristics.

[0087] Figure 4 4 is a flow chart of the method for determining the key breakpoint of telemarketing according to an embodiment of the present invention. Figure 4 As shown, in one embodiment of the present invention, after the above step S302, the method for determining the key breakpoints of telemarketing of the present invention further includes steps S401 and S403.

[0088] Step S401, obtaining the marketing key breakpoint speech template type corresponding to the speech template in each marketing scenario.

[0089] Step S402, marking the marketing key breakpoint label on the corresponding speech template.

[0090] Step S403, marking the customer portrait template on the corresponding speech template.

[0091] The present invention identifies the marketing key breakpoint labels and customer portrait templates on the corresponding speech templates in each marketing scenario for the marketing key breakpoint speech template type. When selecting speech templates, subsequent customer service personnel can clearly see which speech templates belong to the marketing key breakpoint speech template type and carefully select them. In addition, customer service personnel can understand the characteristics of customers who may make rejection actions in the speech template based on the customer portrait template, and then decide whether to adopt the speech template based on the characteristics of the current customer, thereby helping to improve the marketing success rate of customer service personnel.

[0092] In one embodiment of the present invention, the present invention forms a mapping between the marketing key breakpoint speech template type and the speech template corresponding to each marketing scenario, optimizes the speech template and marketing strategy, and reduces the occurrence of breakpoints. The specific steps are as follows: a mapping relationship is formed between the marketing key breakpoint speech template type and the speech template in each marketing scenario. According to the matching of the marketing key breakpoint speech template type, the corresponding speech template and marketing scenario are located, so that the background staff can make corresponding adjustments to the specific speech template to reduce the number of occurrences of this breakpoint. For example: if the marketing key breakpoint speech template type is "handling operation introduction", it is necessary to find the speech template with the speech template type of "handling operation introduction" in each marketing scenario, and then the background staff can adjust the speech template to improve the quality of the speech template and the marketing effect.

[0093] It can be seen from the above embodiments that this method uses knowledge graph technology to establish relationships between the types of speech templates corresponding to customers, marketing scenarios, and marketing breakpoints, and uses the network key node algorithm to develop an analysis method for key breakpoints. The method of the present invention can adjust the speech templates corresponding to the marketing scenarios according to the breakpoint links of different marketing scenarios, which helps to improve the quality of the speech templates and reduce the occurrence of breakpoints. The present invention reduces the derailment links of marketing through breakpoint analysis, helps to improve the success rate of telephone marketing, and enhances the intelligence of digital marketing of commercial banks.

[0094] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0095] Based on the same inventive concept, an embodiment of the present invention also provides a device for determining a critical breakpoint in telemarketing, which can be used to implement the method for determining a critical breakpoint in telemarketing described in the above embodiment, as described in the following embodiment. Since the principle of solving the problem by the device for determining a critical breakpoint in telemarketing is similar to that of the method for determining a critical breakpoint in telemarketing, the embodiment of the device for determining a critical breakpoint in telemarketing can refer to the embodiment of the method for determining a critical breakpoint in telemarketing, and the repeated parts will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiment is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0096] Figure 6 is a first structural block diagram of a device for determining a key breakpoint in telemarketing according to an embodiment of the present invention. Figure 6 As shown, the apparatus for determining the key breakpoints of telemarketing according to the embodiment of the present invention includes:

[0097] A case information acquisition module 1 is used to acquire case information of multiple marketing failure cases, wherein the case information includes: marketing scenario information, customer information, and speech template types corresponding to marketing breakpoints;

[0098] A marketing breakpoint knowledge graph establishment module 2 is used to establish a marketing breakpoint knowledge graph according to the case information, wherein the marketing breakpoint knowledge graph includes a plurality of nodes and edges for connecting the nodes, the nodes include: marketing scenario nodes, customer nodes and speech template type nodes, and the edges are used to connect the marketing scenario nodes, customer nodes and speech template type nodes belonging to the same case information;

[0099] A quantity statistics module 3 is used to obtain the number of edges connected to each speech template type node from the marketing breakpoint knowledge graph;

[0100] The marketing key breakpoint speech template type determination module 4 is used to take the speech template type nodes whose number is greater than the key breakpoint determination threshold as marketing key breakpoint nodes, and take the speech template type corresponding to the marketing key breakpoint nodes as the marketing key breakpoint speech template type.

[0101] Figure 7 is a second structural block diagram of the apparatus for determining a critical breakpoint in telemarketing according to an embodiment of the present invention. Figure 7 As shown, the apparatus for determining key breakpoints of telemarketing in the embodiment of the present invention further includes:

[0102] A customer service and customer dialogue text information acquisition module 5 is used to acquire the customer service and customer dialogue text information corresponding to the marketing failure case, wherein the customer service and customer dialogue text information includes: customer service input text information and customer input text information, and the customer service input text information includes multiple types of speech templates;

[0103] A speech template type identification module 6 is used to determine the speech template type corresponding to each customer service input segment in the customer service input text information;

[0104] The speech template type determination module 7 corresponding to the marketing breakpoint is used to determine the customer service input segment corresponding to the marketing breakpoint according to the customer input text information, and then obtain the speech template type corresponding to the marketing breakpoint.

[0105] In one embodiment of the present invention, the speech template type identification module 6 specifically includes:

[0106] A machine learning unit is used to input each customer service input segment in the customer service input text information into a preset machine learning model to obtain a speech template type corresponding to each customer service input segment, wherein the machine learning model is trained using speech templates marked with speech template types as training data.

[0107] In order to achieve the above object, according to another aspect of the present application, a computer device is also provided. Figure 8 As shown, the computer device includes a memory, a processor, a communication interface and a communication bus. A computer program that can be executed on the processor is stored in the memory. When the processor executes the computer program, the steps in the above-mentioned embodiment method are implemented.

[0108] The processor may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0109] The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer executable programs and units, such as the corresponding program units in the above method embodiments of the present invention. The processor executes various functional applications of the processor and works data processing by running the non-transitory software programs, instructions and modules stored in the memory, that is, implementing the method in the above method embodiments.

[0110] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0111] The one or more units are stored in the memory, and when executed by the processor, the method in the above embodiment is performed.

[0112] The specific details of the above-mentioned computer device can be understood by referring to the corresponding related descriptions and effects in the above-mentioned embodiments, and will not be repeated here.

[0113] In order to achieve the above purpose, according to another aspect of the present application, a computer-readable storage medium is also provided, wherein the computer-readable storage medium stores a computer program, and the computer program implements the steps in the above-mentioned method for determining the key breakpoints of telemarketing when executed in a computer processor. Those skilled in the art can understand that the implementation of all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and the program can include the processes of the embodiments of the above-mentioned methods when executed. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (Flash Memory), a hard disk (Hard Disk Drive, abbreviated as: HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above-mentioned types of memory.

[0114] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0115] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for determining key breakpoints in telemarketing, characterized in that: include: Obtain case information of multiple marketing failure cases, wherein the case information includes: marketing scenario information, customer information, and speech template types corresponding to marketing breakpoints; A marketing breakpoint knowledge graph is established according to the case information, wherein the marketing breakpoint knowledge graph includes a plurality of nodes and edges for connecting the nodes, the nodes include: marketing scenario nodes, customer nodes, and speech template type nodes, and the edges are used to connect the marketing scenario nodes, customer nodes, and speech template type nodes belonging to the same case information; Obtaining the number of edges connected to each speech template type node from the marketing breakpoint knowledge graph; The speech template type nodes whose number is greater than the critical breakpoint determination threshold are used as marketing critical breakpoint nodes, and the speech template type corresponding to the marketing critical breakpoint node is used as the marketing critical breakpoint speech template type; For each of the marketing key breakpoint nodes, the PageRank value of each customer node connected to the marketing key breakpoint node is calculated by using the PageRank algorithm; The customer portrait corresponding to the customer node with the largest PageRank value is used as the customer portrait template of the marketing key breakpoint speech template type corresponding to the marketing key breakpoint node; Obtain the marketing key breakpoint speech template type corresponding to the speech template in each marketing scenario; The marketing key breakpoint label is marked on the corresponding speech template, and the customer portrait template is marked on the corresponding speech template, so that the subsequent customer service personnel can clearly see which speech templates belong to the marketing key breakpoint speech template type when selecting speech templates, and the customer service personnel can understand the characteristics of customers who may reject the speech template based on the customer portrait template, and then decide whether to adopt the speech template based on the characteristics of the current customer.

2. The method for determining the key breakpoints of telemarketing according to claim 1, characterized in that: Also includes: Obtaining customer service and customer dialogue text information corresponding to the marketing failure case, wherein the customer service and customer dialogue text information includes: customer service input text information and customer input text information, and the customer service input text information includes multiple types of speech templates; Determine the speech template type corresponding to each customer service input segment in the customer service input text information; The customer service input segment corresponding to the marketing breakpoint is determined according to the customer input text information, and then the speech template type corresponding to the marketing breakpoint is obtained.

3. The method for determining the key breakpoints of telemarketing according to claim 2, characterized in that: The determining of the speech template type corresponding to each customer service input segment in the customer service input text information specifically includes: Input each customer service input segment in the customer service input text information into a preset machine learning model to obtain the speech template type corresponding to each customer service input segment, wherein the machine learning model is trained using speech templates marked with speech template types as training data.

4. The method for determining the key breakpoints of telemarketing according to claim 2, characterized in that: Also includes: The critical breakpoint determination threshold is determined according to the number of the marketing failure cases.

5. The method for determining the key breakpoints of telemarketing according to claim 2, characterized in that: Also includes: Obtaining audio data of the conversation between customer service and the customer corresponding to the marketing failure case; Based on the preset speech recognition model and the audio data of the conversation between the customer service and the customer, the text information of the conversation between the customer service and the customer corresponding to the marketing failure case is generated.

6. A device for determining key breakpoints in telemarketing, characterized in that: include: A case information acquisition module, used to acquire case information of multiple marketing failure cases, wherein the case information includes: marketing scenario information, customer information, and speech template types corresponding to marketing breakpoints; A marketing breakpoint knowledge graph establishment module is used to establish a marketing breakpoint knowledge graph according to the case information, wherein the marketing breakpoint knowledge graph includes a plurality of nodes and edges for connecting the nodes, the nodes include: marketing scenario nodes, customer nodes and speech template type nodes, and the edges are used to connect the marketing scenario nodes, customer nodes and speech template type nodes belonging to the same case information; A quantity statistics module, used to obtain the number of edges connected to each speech template type node from the marketing breakpoint knowledge graph; A marketing key breakpoint speech template type determination module is used to take the speech template type nodes whose number is greater than the key breakpoint determination threshold as marketing key breakpoint nodes, and take the speech template type corresponding to the marketing key breakpoint node as the marketing key breakpoint speech template type; for each of the marketing key breakpoint nodes, calculate the PageRank value of each customer node connected to the marketing key breakpoint node through the PageRank algorithm; take the customer portrait corresponding to the customer node with the largest PageRank value as the customer portrait template of the marketing key breakpoint speech template type corresponding to the marketing key breakpoint node; obtain the speech template corresponding to the marketing key breakpoint speech template type in each marketing scenario; mark the marketing key breakpoint label on the corresponding speech template, and mark the customer portrait template on the corresponding speech template, so that subsequent customer service personnel can clearly see which speech templates belong to the marketing key breakpoint speech template type when selecting the speech template, and enable the customer service personnel to understand the characteristics of customers who may make a rejection action in the speech template based on the customer portrait template, and then decide whether to adopt the speech template based on the characteristics of the current customer.

7. The device for determining key breakpoints in telemarketing according to claim 6, characterized in that: Also includes: A customer service and customer dialogue text information acquisition module is used to acquire customer service and customer dialogue text information corresponding to the marketing failure case, wherein the customer service and customer dialogue text information includes: customer service input text information and customer input text information, and the customer service input text information includes multiple types of speech templates; A speech template type identification module, used to determine the speech template type corresponding to each customer service input segment in the customer service input text information; A module for determining the speech template type corresponding to a marketing breakpoint is used to determine the customer service input segment corresponding to the marketing breakpoint based on the customer input text information, and then obtain the speech template type corresponding to the marketing breakpoint.

8. The device for determining key breakpoints of telemarketing according to claim 7, characterized in that: The speech template type identification module specifically includes: A machine learning unit is used to input each customer service input segment in the customer service input text information into a preset machine learning model to obtain a speech template type corresponding to each customer service input segment, wherein the machine learning model is trained using speech templates marked with speech template types as training data.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed in a computer processor, the computer program implements the method according to any one of claims 1 to 5.

Images