Customer interaction scheme generation method and device, electronic equipment and storage medium

Abstract

The invention discloses a customer interaction scheme generation method and device, electronic equipment and a storage medium. The method comprises the following steps: determining candidate customer interaction schemes corresponding to customers corresponding to a customer question text, and determining an expected value of each candidate customer interaction scheme by adopting a first type of atlas; determining a client intention corresponding to the client question text, determining multiple types of feasible client interaction schemes corresponding to the client intention from the candidate client interaction schemes according to the second type of map, and determining a target client interaction scheme from the feasible client interaction schemes according to the expected value; and obtaining a workflow corresponding to the target client interaction scheme in the third-class atlas, and sequentially displaying each step in the workflow in a front-end interaction interface. The technical problem of poor customer interaction scheme management capability in the related technology is solved.

Description

Technical Field

[0001] The present application relates to the technical field of knowledge graph data processing, and specifically, to a method, device, electronic device and storage medium for generating a customer interaction plan. Background Art

[0002] The current customer interaction solution management in related technologies has the following problems: separation of customer interaction solution services and marketing management, high coordination costs, training of marketing scripts from top to bottom, divorced from front-line reality, scattered storage of non-standard materials on marketing scripts, poor structural standardization, word-of-mouth transmission of marketing scripts between masters and apprentices, and difficulty in penetrating intensive management, which lead to technical problems such as poor customer interaction solution management capabilities.

[0003] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present application provide a method, device, electronic device and storage medium for generating a customer interaction plan, so as to at least solve the technical problem of poor customer interaction plan management capability in the related art.

[0005] According to one aspect of an embodiment of the present application, a method for generating a customer interaction solution is provided, comprising: determining a candidate customer interaction solution corresponding to a customer corresponding to a customer question text, and using a first type of map to determine an expected value of each candidate customer interaction solution, wherein the first type of map includes: customer characteristics of the customer, solution values ​​corresponding to the candidate customer interaction solutions, and probability weights corresponding to each customer characteristic and each candidate customer interaction solution, the probability weights being used to characterize the probability of successfully recommending a candidate customer interaction solution to a customer having the customer characteristics, and the expected value being determined by the probability weights and the solution values; determining a customer intention corresponding to the customer question text, and determining a feasible customer interaction solution corresponding to the customer intention from the candidate customer interaction solutions based on a second type of map, and determining a target customer interaction solution from the feasible customer interaction solutions based on the expected value, wherein the second type of map contains multiple customer intentions and candidate customer interaction solutions that can be executed to achieve each customer intention; obtaining a workflow corresponding to the target customer interaction solution in the third type of map, and displaying each step in the workflow in sequence in the front-end interaction interface.

[0006] Optionally, determining the customer intent corresponding to the customer question text includes: preprocessing the customer question text, and converting the preprocessed customer question text into a word vector, wherein the preprocessing includes: word segmentation, stop word removal, and stem extraction; determining the similarity between the word vector corresponding to the customer question text and the word vectors corresponding to each customer intent in a preset customer intent catalog; and determining the customer intent with the greatest similarity to the customer question text in the preset customer intent catalog as the customer intent corresponding to the customer question text.

[0007] Optionally, based on the second type of graph, determining a feasible customer interaction plan corresponding to the customer intention from the candidate customer interaction plans includes: determining the candidate customer interaction plans that can be executed in the second type of graph to achieve the customer intention as feasible customer interaction plans, wherein the feasible customer interaction plans include: service plans and marketing plans, and the marketing plans include: business plans and activity plans.

[0008] Optionally, determining the expected value of each candidate customer interaction plan includes: obtaining a customer portrait map of the customer corresponding to the customer question text, wherein the customer portrait map includes multiple customer characteristics of the customer; determining the probability weight of each customer characteristic for each candidate customer interaction plan, and determining the success probability of recommending the candidate customer interaction plan to the customer based on the probability weights corresponding to all the customer characteristics of the customer; obtaining the plan value corresponding to the candidate customer interaction plan in the first category of maps, and determining the expected value of the candidate customer interaction plan based on the plan value and the success probability, wherein the plan value includes: economic value and management value.

[0009] Optionally, determining the target customer interaction plan among feasible customer interaction plans based on the expected value includes: determining a first preset number and a second preset number corresponding to the customer intention; and determining the first preset number of service plans and the second preset number of marketing plans with the largest expected value among the feasible customer interaction plans as the target customer interaction plan.

[0010] Optionally, obtaining the workflow corresponding to the target customer interaction plan in the third type of graph includes: obtaining components corresponding to the target customer interaction plan, wherein the components include: atomic components and composite components, an atomic component corresponds to an atomic operation of executing the target customer interaction plan, and a composite component is encapsulated by multiple atomic components; determining the logical connection relationship between each component to obtain the workflow corresponding to the target customer interaction plan.

[0011] Optionally, the method also includes: sending the workflow corresponding to the target customer interaction plan to the front-end interaction interface for display; responding to the operation instructions of the front-end interaction interface, according to the preset rules of the workflow components corresponding to the operation instructions, and displaying the various steps in the workflows in sequence in the front-end interaction interface to guide the interaction process between employees and users.

[0012] According to another aspect of an embodiment of the present application, a customer interaction solution generating device is also provided, including: an expected value determination module, used to determine the candidate customer interaction solution corresponding to the customer corresponding to the customer problem text, and use the first type of map to determine the expected value of each candidate customer interaction solution, wherein the first type of map includes: the customer characteristics of the customer, the solution value corresponding to the candidate customer interaction solution, and the probability weight corresponding to each customer characteristic and each candidate customer interaction solution, the probability weight is used to characterize the probability of successfully recommending the candidate customer interaction solution to the customer with the customer characteristics, and the expected value is determined by the probability weight and the solution value; a target solution determination module, used to determine the customer intention corresponding to the customer problem text, and based on the second type of map, determine the feasible customer interaction solution corresponding to the customer intention from the candidate customer interaction solutions, and based on the expected value, determine the target customer interaction solution from the feasible customer interaction solutions, wherein the second type of map contains multiple customer intentions and the candidate customer interaction solutions that can be executed to realize each customer intention; a workflow guidance module, used to obtain the workflow corresponding to the target customer interaction solution in the third type of map, and display each step in the workflow in sequence in the front-end interactive interface.

[0013] According to another aspect of the embodiments of the present application, there is also provided an electronic device, including: a memory and a processor, the processor being configured to run a program stored in the memory, wherein the method for generating a customer interaction solution is executed when the program is run.

[0014] According to another aspect of the embodiments of the present application, a non-volatile storage medium is provided, the non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the method for generating a customer interaction solution by running the computer program.

[0015] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, which implements the steps of the customer interaction solution generating method when the computer program is executed by a processor.

[0016] In an embodiment of the present application, a candidate customer interaction solution corresponding to a customer corresponding to a customer question text is determined, and a first type of graph is used to determine an expected value of each candidate customer interaction solution, wherein the first type of graph includes: customer characteristics of the customer, solution values ​​corresponding to the candidate customer interaction solutions, and probability weights corresponding to each customer characteristic and each candidate customer interaction solution, the probability weights are used to characterize the probability of successfully recommending a candidate customer interaction solution to a customer having the customer characteristics, and the expected value is determined by the probability weights and the solution values; the customer intention corresponding to the customer question text is determined, and based on the second type of graph, a feasible customer interaction solution corresponding to the customer intention is determined from the candidate customer interaction solutions, and based on the expected value, a target customer interaction solution is determined from the feasible customer interaction solutions, wherein the second type of graph contains multiple customer intentions and candidate customer interaction solutions that can be executed to achieve each customer intention; a workflow corresponding to the target customer interaction solution in the third type of graph is obtained, and each step in the workflow is sequentially displayed in the front-end interactive interface. By using the technology and method of generating customer interaction solutions for employees based on the three types of knowledge graphs, the purpose of improving the customer interaction solution management capability is achieved, thereby solving the technical problem of poor customer interaction solution management capability in the related technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0018] Figure 1 It is a hardware structure block diagram of a computer terminal (or electronic device) for implementing a method for generating a customer interaction solution according to an embodiment of the present application;

[0019] Figure 2 It is a schematic diagram of a method flow for generating a customer interaction solution provided in an embodiment of the present application;

[0020] Figure 3 It is a schematic diagram of the current status of marketing speech management provided according to an embodiment of the present application;

[0021] Figure 4 It is a schematic diagram of a method for generating a customer interaction solution based on a knowledge graph according to an embodiment of the present application;

[0022] Figure 5 It is a schematic diagram of a process step for generating a customer interaction solution based on a knowledge graph according to an embodiment of the present application;

[0023] Figure 6 is a schematic diagram of a first type of spectrum provided according to an embodiment of the present application;

[0024] Figure 7 is a schematic diagram of a second type of spectrum provided according to an embodiment of the present application;

[0025] Figure 8 It is a schematic diagram of a FTTR marketing solution workflow provided according to an embodiment of the present application;

[0026] Fig. 9 It is a schematic diagram of a map factory design provided according to an embodiment of the present application;

[0027] Fig.10 It is a structural schematic diagram of a customer interaction solution generating device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device 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 devices.

[0030] In order to facilitate those skilled in the art to better understand the embodiments of the present application, some technical terms or nouns involved in the embodiments of the present application are explained as follows:

[0031] Knowledge graph: A knowledge graph is a semantic network, a graph-based data structure consisting of nodes and edges. Nodes correspond to entities in the real world, and edges represent the relationships between entities. Both nodes and edges have attributes such as identifiers, names, and types. Knowledge graphs have been widely used in various fields, including search engines, smart question and answer, customer profiling, workflow management, marketing program management, etc.

[0032] Customer interaction plan: Customer interaction refers to the action plan for two-way communication between employees and customers at various touchpoints (including: customer service, home maintenance, face-to-face store, street vendors, etc.). From the perspective of employees, the customer interaction process includes: 1) identifying customer intent; 2) selecting appropriate service plans and marketing plans based on customer intent; 3) executing service and marketing plans, introducing and displaying business-related information to customers; during the execution process, it is necessary to intersperse customer dialogues and inquiries, and complete necessary operations such as inquiries and processing on various systems.

[0033] The customer interaction plan can be understood as a container. 1) The service plan includes the solutions that employees use to solve customer complaints, such as how to order services, how to cancel services, how to guide customers to check their equipment or arrange for maintenance technicians to come to their homes, etc.; 2) The marketing plan can be either a business marketing plan or an activity marketing plan, which includes how employees introduce services to customers, how to present results, how to pay for orders, etc.

[0034] Talking skills: The words spoken by employees to customers or the text output on the screen during the introduction and presentation process are called talking skills. Talking skills are not a stereotyped string of characters, but are organically embedded in the workflow. Excellent talking skills need to be flexibly and strategically organized and generated in the process of employee-customer interaction based on customer characteristics, contact scenarios, and communication situations. Excellent talking skills can capture customer interests, fully demonstrate product advantages, and significantly improve service satisfaction and business marketing success rate.

[0035] Workflow map: The specific interaction process between employees and customers who implement service and marketing plans is not a simple conversation or typing, but a set of designated workflows that require employees to make selections, inquiries, and queries based on many variables in the interaction process.

[0036] Cross-demand and cross-marketing: Customers contact telecom employees for a variety of reasons. Some customers are mainly service-oriented, such as solving problems with unstable Wi-Fi and insufficient data traffic; some customers are mainly business-oriented, such as broadband speed-up and applying for secondary cards for family members; and some customers appear to have service demands, but in fact they are cross-demands of business needs, such as complaining that Wi-Fi can be connected in some places at home but not in others, and that coverage is poor. This is a complaint about service issues on the surface, but in fact it is a marketing opportunity for upgrading FTTR (Fiber to the Room) or whole-house Wi-Fi to a higher-end service. Identifying and utilizing cross-demands between services and marketing and conducting cross-marketing of services is an important means to improve business development capabilities.

[0037] In the related art, the current customer interaction solution management has the following problems:

[0038] 1) Customer interaction solution services and marketing management are separated, and cross-marketing is a weakness;

[0039] Under the current organizational structure, the management of customer interaction plans is seriously separated, and in fact two separate service knowledge bases and marketing knowledge bases have been built: (a) the service plan is the responsibility of the service management department, and the marketing plan is the responsibility of the marketing development department. The two departments are highly professional, independent of each other, and have difficulty in collaborative communication; (b) the service plan is basically the same across the province and can be formulated uniformly by the service management department, but the marketing plan is highly local and time-sensitive. The policy formulation, key business selection, regional competition situation, preferential subsidies, regional organizational structure and operating habits of various cities, district bureaus and branches vary greatly, making it difficult to form a province-wide intensive marketing plan; and the market changes rapidly, the life cycle of the marketing plan is short, and changes and adjustments are frequent, resulting in extremely high costs for cross-collaboration between marketing and service. Therefore, the service management department is unwilling to cooperate with the marketing development department, and the marketing development department has no time to pay attention to collaboration with services.

[0040] 2) Marketing techniques are transferred from top to bottom and are divorced from the actual situation on the front line;

[0041] The current source of marketing script production is the provincial and municipal administrators of various business lines, such as Figure 3 As shown, they compiled a number of simple "question and answer texts" based on their own understanding of the business and the "imaginary work scenarios" of the front line, distributed them to internal trainers, district bureau or branch administrators in various regions, and then forwarded them to the backbones of the front line (including store star employees, service team leaders, installation and maintenance team leaders, etc.). These backbones embedded these scripts into the process based on practical experience, and formed practical scripts according to the scene contacts, customer portraits, preferential policies, regional situations, business characteristics, etc. of the specific process, realizing "a thousand faces and a thousand scripts for a thousand people."

[0042] From the above process, it can be seen that there are huge problems in the management system of marketing scripts: (a) The source of the scripts is business managers who are separated from the front line, rather than elite employees on the front line. The scripts written by business managers lack the essence of the "process orchestration" scripts that are adapted to local conditions on the front line and are not very effective. Elite employees on the front line are at a low level and cannot provide feedback upward; (b) A backbone employee has limited energy and can only be familiar with certain businesses and have special marketing tricks. He spends time and energy to come up with efficient scripts, but may know nothing about other businesses. Therefore, the efficient marketing scripts summarized by individuals for certain businesses can only be used in a small local area. The efficient scripts summarized by many backbone employees are difficult to inspire and learn from each other across regions, and knowledge is difficult to spread, resulting in waste.

[0043] 3) Non-standard materials on marketing techniques are stored in a scattered manner, with poor structural standardization;

[0044] Marketing techniques, especially the knowledge of "process arrangement" embedded in local conditions, currently lack a good storage method and are still scattered in the hands of administrators at all levels. The data formats are also varied, including pictures, word, excel, pdf, and ppt. Everyone has different working habits, and many data are poorly structured and standardized, with a low level of digitization, making them difficult to be used by the system.

[0045] 4) Marketing techniques are passed down by word of mouth between teachers and apprentices, which makes it difficult to penetrate intensive management;

[0046] The production and training of marketing techniques all rely on the experience of key employees. There is a lack of effective system management, and they can only be passed on by word of mouth between master and apprentice. The master does not know how well the apprentice has learned and how well he has used it, making it difficult to ensure the consistency, accuracy, and effectiveness of employee-customer interaction. The learning effect of the apprentice mainly depends on his personal experience and quality, and the effect of improving the marketing technique ability is difficult to guarantee.

[0047] In order to solve the above problems, relevant solutions are provided in the embodiments of the present application, which are described in detail below.

[0048] According to an embodiment of the present application, a method embodiment for generating a customer interaction solution is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0049] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computer terminal (or electronic device) for implementing a method for generating a customer interaction solution. Figure 1 As shown, the computer terminal 10 (or electronic device) may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0050] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or electronic device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0051] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the customer interaction solution generation method in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, the above-mentioned customer interaction solution generation method is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 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 a combination thereof.

[0052] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0053] The display may be, for example, a touch screen liquid crystal display (LCD) that enables a user to interact with a user interface of the computer terminal 10 (or electronic device).

[0054] In the above operating environment, the embodiment of the present application provides a method for generating a customer interaction solution. Figure 2 is a schematic diagram of a method flow for generating a customer interaction solution according to an embodiment of the present application, such as Figure 2 As shown, the method comprises the following steps:

[0055] Step S202, determining the candidate customer interaction solutions corresponding to the customer corresponding to the customer question text, and using the first type of graph to determine the expected value of each candidate customer interaction solution, wherein the first type of graph includes: customer characteristics of the customer, solution values ​​corresponding to the candidate customer interaction solutions, and probability weights corresponding to each customer characteristic and each candidate customer interaction solution, the probability weights are used to represent the probability of successfully recommending the candidate customer interaction solutions to the customer with the customer characteristics, and the expected value is determined by the probability weights and solution values;

[0056] Step S204, determining the customer intention corresponding to the customer question text, and determining a feasible customer interaction solution corresponding to the customer intention from the candidate customer interaction solutions based on the second type of graph, and determining a target customer interaction solution from the feasible customer interaction solutions based on the expected value, wherein the second type of graph includes multiple customer intentions and candidate customer interaction solutions that can be executed to achieve each customer intention;

[0057] Step S206, obtaining the workflow corresponding to the target customer interaction solution in the third type of graph, and displaying each step in the workflow in sequence in the front-end interactive interface.

[0058] Through the above steps, by generating technologies and methods for employees to manage customer interaction plans based on three types of knowledge graphs, the purpose of improving the customer interaction plan management capabilities is achieved, thereby solving the technical problem of poor customer interaction plan management capabilities in related technologies.

[0059] The following further introduces the method for generating a customer interaction solution in steps S202 to S206 of the embodiment of the present application.

[0060] Figure 4 is a schematic diagram of a method for generating a customer interaction solution based on a knowledge graph according to an embodiment of the present application, such as Figure 4 As shown, by comprehensively based on the three types of knowledge graphs, the technology and methods for generating employee-customer interaction solutions can be used. Specifically, the value of all candidate customer interaction solutions can be estimated based on the customer portrait graph, service solution, enterprise business, marketing activity graph, etc. (the first type of graph) (solving the question "How valuable is the solution?"); according to the "customer intention-interaction solution" block matrix graph (the second type of graph), feasible service and marketing solutions suitable for customers can be matched, that is, the target customer interaction solution can be determined (solving the question "What solutions can be made?"); according to the workflow graph (the third type of graph), nanny-style interaction operation instructions can be generated in the employee-customer interaction work interface (solving the question "How to implement the solution specifically?"), and multi-modality can be used to guide average employees at each touch point to perform customer interaction in accordance with the efficient practices of excellent employees.

[0061] The following is a detailed introduction to the construction and use of the three types of maps. Figure 5 As shown, this application connects three types of knowledge graphs throughout to generate a directly usable multimodal employee-customer interaction solution, forming a complete workflow based on the knowledge graph, as follows.

[0062] After obtaining the customer question text provided by a user, the customer characteristics can be obtained (based on the customer number, etc.), and the first type of graph can be used to calculate the value of all candidate customer interaction plans for this customer, that is, to determine the expected value of each candidate customer interaction plan. The specific steps are as follows.

[0063] In some embodiments of the present application, determining the expected value of each candidate customer interaction plan includes: obtaining a customer portrait map of the customer corresponding to the customer question text, wherein the customer portrait map includes multiple customer characteristics of the customer; determining the probability weight of each customer characteristic for each candidate customer interaction plan, and determining the success probability of recommending the candidate customer interaction plan to the customer based on the probability weights corresponding to all the customer characteristics of the customer; obtaining the plan value corresponding to the candidate customer interaction plan in the first category of maps, and determining the expected value of the candidate customer interaction plan based on the plan value and the success probability, wherein the plan value includes: economic value and management value.

[0064] The following is an introduction to the construction process of the first type of atlas.

[0065] First, a customer interaction plan catalog is established. According to the nature of the plan content, it can be divided into two categories: service plans and marketing plans. Marketing plans can include subcategories such as business plans and activity plans. In this embodiment, service plans and marketing plans are established in two categories according to needs. The purpose is to facilitate the solution of the functional problems of service cross-marketing in the next step. In practice, more types of plans can be established as needed.

[0066] After that, you can build the first type of map, including customer portrait map and service solution map, enterprise business, and marketing activity map, and estimate the value, success rate, and expected value of successfully recommending a solution to a customer, such as Figure 6 As shown, the customer portrait map records the characteristics of each customer, represented by a 0-1 variable. If customer C1 has characteristic T1, the variable: characteristic has C1T1 = 1; if customer C1 does not have characteristic T1, the variable: characteristic has C1T1 = 0. At the same time, based on historical data, the probability weight of each characteristic T for business S and activity A can be calculated using parameter estimation methods such as neural networks or regression, and the success probability of a customer recommending a successful business S and activity A can be obtained, as shown in the following formula:

[0067]

[0068] In this embodiment, it can be set that the service plan cannot generate economic value and has no management value, so the expected total value of a service plan is zero. Therefore, the matching of the service plan is determined by the nature of the customer's problem, not by its value. The matching of the marketing plan needs to be judged based on the expected value. The marketing specialist can evaluate the economic value that a customer can bring by adopting the marketing plan, ordering the recommended business or participating in the recommended activities. In some cases, a marketing plan also has management value. For example, the probability of success of promoting business A to a customer is higher and the value is greater, but due to strategic considerations, it will tend to develop business B. The marketing specialist will add a management value to business B (the default management value of all plans is 0). Finally, the expected total value of recommending a marketing plan to a customer can be calculated = (economic value + management value) * success probability, that is, the expected value of the candidate customer interaction plan is equal to the product of the plan value and the success probability.

[0069] On the other hand, after obtaining a customer question text provided by a user, it is also necessary to determine the customer intent corresponding to the customer question text. For example, algorithms such as similarity can be used to match the customer question text to a preset customer intent directory to identify the customer intent. The specific steps are as follows.

[0070] In some embodiments of the present application, determining the customer intent corresponding to a customer question text includes the following steps: preprocessing the customer question text, and converting the preprocessed customer question text into a word vector, wherein the preprocessing includes: word segmentation, stop word removal, and stem extraction; determining the similarity between the word vector corresponding to the customer question text and the word vectors corresponding to each customer intent in a preset customer intent catalog; and determining the customer intent in the preset customer intent catalog that has the greatest similarity with the customer question text as the customer intent corresponding to the customer question text.

[0071] In this embodiment, it is necessary to establish a customer intent catalog to cover various customer demand situations that need to be handled. Specifically, the preset customer intent catalog needs to exhaustively list the possible intentions of the customer at each touch point and express them in a standard text. If it is found that a customer problem cannot be covered by the existing intent catalog, the gap needs to be filled immediately.

[0072] After preprocessing, the text of the customer's question (i.e., the customer's question text) can be matched with the standard question text representing the customer's intention using similarity and other algorithms to identify the customer's intention. Specifically, if it is an online customer question-and-answer format, the FAISS library can be used to calculate the similarity between the customer's question text and the standard customer intention in the customer intention catalog in the word vector embedding space to match the customer's intention, such as "complaining about wifi coverage"; if it is a door-to-door installation and maintenance, or face-to-face store, employees can also directly select the customer's intention.

[0073] After determining the current customer intention, you can obtain feasible service plans and marketing plans based on the "customer intention-interaction plan" matrix of the second type of map. That is, based on the second type of map, determine the feasible customer interaction plan corresponding to the customer intention. The specific steps are as follows.

[0074] In some embodiments of the present application, based on the second type of graph, determining a feasible customer interaction plan corresponding to the customer intention from the candidate customer interaction plans includes: determining the candidate customer interaction plans that can be executed in the second type of graph to achieve the customer intention as feasible customer interaction plans, wherein the feasible customer interaction plans include: service plans and marketing plans, and the marketing plans include: business-type plans and activity-type plans.

[0075] The following is an introduction to the process of establishing the second type of atlas.

[0076] In this embodiment, a second type of graph can be established to form a "customer intention-interaction solution" block matrix graph to construct a feasible relationship between customer intention and interaction solution, such as Figure 7 As shown, the connection forms a "feasible" relationship between customer intention and interaction plan, which means: after identifying the customer's intention, find a feasible plan under the intention based on the connection. If there is no connection between the intention and the plan, it means that this plan cannot and should not be implemented for the customer.

[0077] In addition, as an optional implementation, the second type of map can also be displayed in the form of a "customer intention-interaction plan" block matrix as shown in the following table, with "customer intention" as the row, service plan and marketing plan as the column, and the necessary control parameters can be configured. Import the block matrix file to generate the second type of map.

[0078]

[0079] As shown in the table above, the matrix columns are divided into blocks, one for service-related interaction solutions and the other for marketing-related interaction solutions. If the matrix element is "Y", it means that if the above steps identify a row of customer intent, the employee should execute the solution to interact with the customer. For example, if the customer's intent is to "report wifi coverage", the system can present the service solution of "unstable network connection" and the marketing solutions of "recommending whole-house wifi" and "recommending FTTR" on the work interface. That is, employees need to first help customers answer questions according to the service solution, and find opportunities to cross-market whole-house wifi or FTTR products.

[0080] It should be noted that, generally speaking, only one service plan needs to be configured for a customer intent, but multiple marketing plans may need to be configured. The "Y" here means a switch, and there is a corresponding line on the graph, indicating that under this intent, a certain business can be marketed. Because not all intents can be marketed for a business, for example, if a customer is very angry and complains about overcharges and asks for an explanation, it is not suitable for recommending a business. Therefore, for this type of intent to complain about overcharges, only a service plan needs to be configured, and no marketing plan can be configured.

[0081] The second type of map plays the role of a connecting hub. On the one hand, it carries the feasible relationship information between "customer intention-interaction plan", and on the other hand, it connects the first and third types of maps. Moreover, its matrix map block structure, by supporting service-marketing intersection, can provide a unified customer interaction plan management system, which can support service-type interaction plans and marketing-type interaction plans, and provide a unified interface for front-line employees, effectively solving the problem of coordinated intersection among multiple types of plans such as marketing and service.

[0082] It should be noted that in the embodiments of the present application, only an exemplary method for unified management of two types of interactive schemes, service schemes and marketing schemes, is given. In fact, the present application does not limit the number of categories of interactive schemes managed. Based on the above ideas, unified management of multiple types of interactive schemes can be achieved. For example, the above scheme can be expanded by adding new interactive scheme categories in the block matrix (for example, adding a category of public welfare messages).

[0083] After obtaining the expected value of each feasible customer interaction plan, you can further select the appropriate plan with the highest value as the target customer interaction plan based on the needs. The specific steps are as follows.

[0084] In some embodiments of the present application, determining a target customer interaction plan among feasible customer interaction plans based on expected value includes: determining a first preset number and a second preset number corresponding to customer intentions; and determining a first preset number of service plans among the feasible customer interaction plans and a second preset number of marketing plans with the greatest expected value as target customer interaction plans.

[0085] Specifically, an appropriate plan with the highest value can be selected according to the control parameters corresponding to the customer's intention (including but not limited to: the first preset number, the second preset number, etc.). For example, it can be stipulated that a customer's intention can only match one service plan, but multiple marketing plans may be feasible. For example, for customers who complain about wifi coverage, the only service plan requires employees to tell him how to troubleshoot by himself and can help arrange for installation and maintenance technicians to come to the door, and there are two feasible marketing plans: "FTTR marketing" and "whole-house wifi marketing". If the number of marketing plan recommendations = 1 in the control parameters of the customer's intention, the one with the largest expected value is selected from the two plans of "FTTR marketing" and "whole-house wifi marketing", and the employee is asked to execute the marketing. If the number of marketing plan recommendations = 2 in the control parameters of the customer's intention, the employee is asked to execute the two marketing plans in order of expected value.

[0086] The first type of map and the second type of map are connected together, which solves the ranking problem of recommended business and which business can be promoted in the scenario in a serial manner. These two marketing decision-making problems can recommend business to customers in a targeted manner and improve the success rate of business development. For example, in the table above, for the customer intention of "querying the package optimization application method", the second type of map sets 3 feasible solutions, but due to the limited time of the customer, only 2 can be selected. Therefore, according to the control parameters, the two with the largest expected value in the first map are selected and let the employees execute them in order. Some customers received two recommendations of "applying for a secondary card" and "applying for a comprehensive video package", while some customers received two recommendations of "applying for a traffic package" and "applying for a comprehensive video package". In this way, based on the second type of map, the business is recommended within the business scope set by the marketing personnel, and based on the first type of map, the interests of the enterprise are maximized, which is in line with the management norms and has the flexibility of business development. At the same time, the division of labor interface within the marketing development department is also relatively clear. The setting of the first type of map can be completed by marketing data experts, while the second type of map can be set by marketing business experts and managers.

[0087] After determining the target customer interaction plan, you can use the third type of map to present and guide employees to perform interaction tasks in the employee-customer interaction work interface according to the workflow associated with the plan. The specific steps are as follows.

[0088] In some embodiments of the present application, obtaining a workflow corresponding to a target customer interaction solution in a third type of graph includes the following steps: obtaining components corresponding to the target customer interaction solution, wherein the components include: atomic components and composite components, an atomic component corresponds to an atomic operation of executing the target customer interaction solution, and a composite component is encapsulated by multiple atomic components; determining the logical connection relationship between each component to obtain the workflow corresponding to the target customer interaction solution.

[0089] Specifically, in this embodiment, a third type of map can be built using a canvas drag-and-drop construction tool called "map factory" to produce workflows and fill in customer interaction plans. A map component in the "map factory" represents an action of an employee interacting with a customer. Knowledge specialists can use map components to assemble interactive workflows with customers based on the experience of outstanding employees and associate them with designated customer interaction plans.

[0090] The above target customer interaction plan is a container that loads specific interactive operation guidance content. The content loading can be completed through the third type of graph. The following introduces the process of determining the corresponding workflow by taking the execution of the FTTR business marketing plan as an example.

[0091] The implementation of the FTTR business marketing plan generally requires the following: 1) Communicate and inquire about the customer's address and apartment type; 2) After obtaining the customer's address and residence, enter the customer's address and apartment type on the community apartment type query webpage to find out whether the customer's residence is suitable for installing FTTR. 3.1) If the apartment is small and not suitable for FTTR, it is necessary to promote the whole-house wifi product; 3.2) If the apartment is suitable, select the device networking solution on the page, generate a heat map, send it to or present it to the customer, and use the language of speech to fully introduce and display the product advantages and features of FTTR. The corresponding workflow is as follows: Figure 8 As shown, a detailed introduction is given below.

[0092] First, obtain the components corresponding to the target customer interaction solution, such as Fig. 9 As shown in the figure, a component in the graph factory represents an atomic operation of an employee to perform a customer interaction solution, and the type includes a process selection class. It supports selection, branching, looping, etc. Specifically, it includes but is not limited to:

[0093] Information query components, supporting database query, interface query, etc.;

[0094] Parsing and processing components support conversion, regular expressions, storage of intermediate variables, variable substitution, custom functions, etc.

[0095] The solution jump component supports jumping to another solution after executing this one. For example, if the query finds that the customer's apartment is small and not suitable for FTTR, it will jump from this solution to the marketing solution of whole-house WiFi.

[0096] Customer interaction components include page forms (html, css, control logic js and other page components), support multimodal interaction, and can generate customized pictures, web pages, AIGC videos and other content for employees to present to customers. When the workflow executes to this component, multimodal content will be presented on the employee-customer interaction interface. This component can not only support presenting information to employees, but also accept employee input (text input, clicks, etc.) for feedback.

[0097] Composite component class, in order to facilitate the operation of knowledge specialists, composite components can be encapsulated based on atomic operation components, that is, some atomic operation components can be packaged and used according to common operations. For example, the commonly used composite component of "determining FTTR apartment type" includes (a) interactive components: how to present the page on the employee-customer interactive work interface; (b) interactive components: how to accept employee feedback input; (c) information query components: how to call the information interface for judgment based on employee input; (d) parsing|processing components: how to parse the interface results into "yes / no" standard results.

[0098] After the graph factory assembles components to form a workflow graph, it can be published as an interface and associated with the corresponding solution. After that, an employee-customer interaction work interface (i.e., the above-mentioned front-end interaction interface) can be established. In the work interface, employees can view the workflow graph and interact with the system. For example, they can execute the workflow in the selected solution, present and guide the interaction process between employees and customers, including but not limited to: responding to and processing employee selection, input, click and other operations. For example, employees can enter the customer's specific address based on the customer's answer, query results, etc. In addition, it should be noted that although only the solution for interacting with employees is shown in the embodiment of the present application, in fact, this solution can also be applied to scenarios where interaction with customers is required, and there is no limitation on this.

[0099] In this embodiment, the interactive work interface can execute the workflow in the selected solution and present it step by step according to the employee's operation. The workflow in the solution is determined by repeated discussions between experts and excellent employees (doing homework), which can guide employees to complete their work like a nanny. Ordinary employees only need to "copy homework" according to the instructions, which reduces the training cost of employees and improves their capabilities.

[0100] For example, if the FTTR marketing plan workflow is executed: Determine the FTTR apartment type composite component: the interactive work interface generates a page, prompting the employee to inquire about the customer's residence and enter text; the employee selects the apartment type of the customer on the page; queries the relevant interface and determines whether the apartment type is suitable for the FTTR business; if "no", jump to another "whole house wifi" plan; if "yes", display the marketing script text: "FTTR extends the fiber optic network to the room, which can provide full coverage and no dead angle wifi access for large-area rooms, which can completely solve your pain points, and more and more people around you have chosen it." The employee clicks to send the text. According to the customer's apartment type, adjust the dedicated interface to generate a thermal effect map of the customer's apartment type. The employee clicks send and pushes it to the customer.

[0101] In addition, as an optional implementation, the method may further include the following steps: sending the workflow corresponding to the target customer interaction plan to the front-end interaction interface for display; responding to the operation instructions of the front-end interaction interface, and displaying the various steps in the various workflows in sequence in the front-end interaction interface according to the preset rules of the workflow components corresponding to the operation instructions, so as to guide the interaction process between employees and users.

[0102] In addition, in the embodiment of the present application, RPA (Robotic Process Automation) traversal tasks can be compiled according to needs to simulate the real employee-customer interaction process. Starting from the customer's intention, three types of inspection maps are traversed, including every branch of the workflow, to find bugs, problems that are about to fail, and problems that need to be changed, saving manpower.

[0103] The embodiments of the present application involve graphs of three types of different functions and involve many branch processes of employee-customer interaction. Manual inspection of each process is extremely costly. RPA traversal tasks can be compiled as needed to simulate the employee-customer interaction process, so as to identify and correct problems in a timely manner and save manpower for testing and inspection.

[0104] This application scheme involves a technology and method that can generate customer interaction plans for employees based on three types of knowledge graphs, build a unified knowledge base for services and marketing, solve the problems of production, storage and collaborative use of multiple types of knowledge, and focus on improving service cross-marketing capabilities. It can be applied to service scenarios, marketing scenarios, and service cross-marketing scenarios at the same time.

[0105] The entire process of this application plan is completely generated by graph configuration, with high visualization and readability, and is compatible with multiple types of service, marketing knowledge and their execution plans. The knowledge graph is used to standardize the preservation of the entire process of enterprise services and marketing, which improves readability, usability and interpretability, and solves the problem of scattered storage of existing non-standard knowledge structures.

[0106] Experts and excellent employees can do their homework, while novices and ordinary employees can copy it. Ordinary employees can easily copy the skills of experts and excellent employees in the whole process of estimating the value of solutions, matching and selecting solutions, and executing operation solutions, which can improve the quality of employee-customer interaction, reduce the difficulty of employee work, and reduce training costs.

[0107] A unified service + marketing knowledge base can be constructed using a unified standard method such as knowledge graph, which changes the problem of building the service knowledge base and the marketing knowledge base separately. Employees can be guided in a multi-modal manner to complete service and marketing interaction tasks with customers, accommodating and solving the problem of service cross-marketing. In this way, the service management department and the marketing development department can jointly formulate corresponding service plans and marketing plans based on customer intentions and characteristics in a scenario. The goals and boundaries are clear in the collaborative process. There is no longer a situation where one department directs another department to do things. Instead, the two departments solve the same problem in their respective fields. The weight parameters of the first type of graph, the feasible relationships and control parameters of the second type of graph, and the interaction process of the third type of graph can be set independently. The coupling between the two departments is reduced, and coordination becomes easier.

[0108] Applications can be directly published to various touchpoints, including: Wanhao Telecom hotline, IM customer service (instant message customer service, such as QQ customer service), enterprise micro-blog and other touchpoints can be automatically presented on the employee's work interface screen, providing nanny-style guidance; for touchpoints such as face-to-face store and street vendors that cannot be viewed and used immediately on the spot, a simulation training interface is provided so that employees can prepare for training for each scenario.

[0109] It can greatly enhance the micro-level penetration capability of marketing management, allowing marketing management work to penetrate deeply from the original macro-level areas such as policy formulation, product design, price control, and business analysis to the micro-level employee-customer interaction process in the terminal touchpoints.

[0110] This application solution can specifically meet the following management requirements for customer interaction solutions:

[0111] 1) Unified customer interaction solution management capabilities, supporting service-marketing crossover. A unified customer interaction solution management system can be provided to support both service-related and marketing-related interaction solutions, providing a unified interface for front-line employees. It can provide both service-related and marketing-related solutions for employee-customer interaction, and can also organically combine the two to form an efficient collaborative crossover, allowing service employees to do business marketing while serving customers, and marketing employees to solve customer service problems while marketing the business.

[0112] 2) Marketing language should be extracted from the front line rather than from senior business managers. Marketing language should be refined and extracted to be close to the first-line star employees, rather than marketing language written by senior business managers on the spur of the moment. Moreover, marketing language should be organically embedded in the process. It should not be just text spoken to customers, but a "workflow" that choreographs the overall process of the language.

[0113] 3) Customer interaction plans can be managed and stored in a unified, convenient, and readable manner. Through this application plan, a system can be established that allows knowledge specialists in various regions and departments to compile and store knowledge, and can be compatible with service plans and marketing plans. At the same time, the plans should be easy to operate and the generated plans should be easy to understand, and no longer be a variety of scattered documents without a unified structure.

[0114] 4) The customer interaction plan is directly used in production, which is convenient for evaluating and understanding the learning and usage of employees. After the interaction plan is generated, it can be directly published and applied to various touch points. Among them: Wanhao Telecom Hotline, IM customer service (such as QQ customer service), enterprise WeChat and other touch points can be automatically presented on the employee's work interface screen for nanny-style guidance; for touch points such as face-to-face meetings in stores and street vendors that cannot be viewed and used on the spot, a simulated training interface is provided so that employees can prepare for training for each scenario.

[0115] According to an embodiment of the present application, an embodiment of a device for generating a customer interaction plan is also provided. Fig.10 Schematic diagram of a structure of a client interaction solution generating device provided according to an embodiment of the present application. Fig.10 As shown, the device comprises:

[0116] The expected value determination module 1000 is used to determine the candidate customer interaction solutions corresponding to the customer corresponding to the customer question text, and use the first type of map to determine the expected value of each candidate customer interaction solution, wherein the first type of map includes: customer characteristics of the customer, solution values ​​corresponding to the candidate customer interaction solutions, and probability weights corresponding to each customer characteristic and each candidate customer interaction solution, the probability weights are used to represent the probability of successfully recommending the candidate customer interaction solution to the customer with the customer characteristics, and the expected value is determined by the probability weights and solution values;

[0117] The target solution determination module 1002 is used to determine the customer intention corresponding to the customer question text, and determine the feasible customer interaction solution corresponding to the customer intention from the candidate customer interaction solutions according to the second type of map, and determine the target customer interaction solution from the feasible customer interaction solutions according to the expected value, wherein the second type of map includes multiple customer intentions and the candidate customer interaction solutions that can be executed to realize each customer intention;

[0118] The workflow guidance module 1004 is used to obtain the workflow corresponding to the target customer interaction solution in the third type of graph, and to display each step in the workflow in sequence in the front-end interactive interface.

[0119] Optionally, determining the customer intent corresponding to the customer question text includes: preprocessing the customer question text, and converting the preprocessed customer question text into a word vector, wherein the preprocessing includes: word segmentation, stop word removal, and stem extraction; determining the similarity between the word vector corresponding to the customer question text and the word vectors corresponding to each customer intent in a preset customer intent catalog; and determining the customer intent with the greatest similarity to the customer question text in the preset customer intent catalog as the customer intent corresponding to the customer question text.

[0120] Optionally, based on the second type of graph, determining a feasible customer interaction plan corresponding to the customer intention from the candidate customer interaction plans includes: determining the candidate customer interaction plans that can be executed in the second type of graph to achieve the customer intention as feasible customer interaction plans, wherein the feasible customer interaction plans include: service plans and marketing plans, and the marketing plans include: business plans and activity plans.

[0121] Optionally, determining the expected value of each candidate customer interaction plan includes: obtaining a customer portrait map of the customer corresponding to the customer question text, wherein the customer portrait map includes multiple customer characteristics of the customer; determining the probability weight of each customer characteristic for each candidate customer interaction plan, and determining the success probability of recommending the candidate customer interaction plan to the customer based on the probability weights corresponding to all the customer characteristics of the customer; obtaining the plan value corresponding to the candidate customer interaction plan in the first category of maps, and determining the expected value of the candidate customer interaction plan based on the plan value and the success probability, wherein the plan value includes: economic value and management value.

[0122] Optionally, determining the target customer interaction plan among feasible customer interaction plans based on the expected value includes: determining a first preset number and a second preset number corresponding to the customer intention; and determining the first preset number of service plans and the second preset number of marketing plans with the largest expected value among the feasible customer interaction plans as the target customer interaction plan.

[0123] Optionally, obtaining the workflow corresponding to the target customer interaction plan in the third type of graph includes: obtaining components corresponding to the target customer interaction plan, wherein the components include: atomic components and composite components, an atomic component corresponds to an atomic operation of executing the target customer interaction plan, and a composite component is encapsulated by multiple atomic components; determining the logical connection relationship between each component to obtain the workflow corresponding to the target customer interaction plan.

[0124] Optionally, the workflow guidance module 1004 is also used to: send the workflow corresponding to the target customer interaction plan to the front-end interaction interface for display; respond to the operation instructions of the front-end interaction interface, and display the various steps in the various workflows in sequence in the front-end interaction interface according to the preset rules of the workflow components corresponding to the operation instructions, so as to guide the interaction process between employees and users.

[0125] It should be noted that the various modules in the above-mentioned customer interaction solution generating device can be program modules (for example, a set of program instructions that implement a certain specific function) or hardware modules. For the latter, it can be expressed in the following forms, but is not limited to this: the expression form of each of the above-mentioned modules is a processor, or the functions of each of the above-mentioned modules are implemented by a processor.

[0126] It should be noted that the client interaction solution generation device provided in this embodiment can be used to execute Figure 2 The customer interaction plan generating method shown, therefore, the relevant explanations and instructions on the above-mentioned customer interaction plan generating method are also applicable to the embodiments of the present application and will not be repeated here.

[0127] The embodiment of the present application also provides a non-volatile storage medium, the non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the following customer interaction solution generation method by running the computer program: determining the candidate customer interaction solution corresponding to the customer corresponding to the customer problem text, and using the first type of map to determine the expected value of each candidate customer interaction solution, wherein the first type of map includes: the customer characteristics of the customer, the solution value corresponding to the candidate customer interaction solution, and the probability weight corresponding to each customer characteristic and each candidate customer interaction solution, the probability weight is used to characterize the probability of successfully recommending the candidate customer interaction solution to the customer with the customer characteristics, and the expected value is determined by the probability weight and the solution value; determining the customer intention corresponding to the customer problem text, and based on the second type of map, determining the feasible customer interaction solution corresponding to the customer intention from the candidate customer interaction solutions, and based on the expected value, determining the target customer interaction solution from the feasible customer interaction solutions, wherein the second type of map contains multiple customer intentions and the candidate customer interaction solutions that can be executed to realize each customer intention; obtaining the workflow corresponding to the target customer interaction solution in the third type of map, and displaying each step in the workflow in sequence in the front-end interactive interface.

[0128] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the customer interaction solution generation method described in each embodiment of the present application: determining the candidate customer interaction solution corresponding to the customer corresponding to the customer question text, and using the first type of map to determine the expected value of each candidate customer interaction solution, wherein the first type of map includes: the customer characteristics of the customer, the solution value corresponding to the candidate customer interaction solution, and the probability weight corresponding to each customer characteristic and each candidate customer interaction solution, the probability weight is used to characterize the probability of successfully recommending the candidate customer interaction solution to the customer with the customer characteristics, and the expected value is determined by the probability weight and the solution value; determining the customer intention corresponding to the customer question text, and determining the feasible customer interaction solution corresponding to the customer intention from the candidate customer interaction solutions based on the second type of map, and determining the target customer interaction solution from the feasible customer interaction solutions based on the expected value, wherein the second type of map contains multiple customer intentions and the candidate customer interaction solutions that can be executed to realize each customer intention; obtaining the workflow corresponding to the target customer interaction solution in the third type of map, and displaying each step in the workflow in sequence in the front-end interactive interface.

[0129] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0130] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0131] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0132] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0133] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0134] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0135] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for generating a customer interaction plan, characterized in that: include: Determine the candidate customer interaction scheme corresponding to the customer corresponding to the customer question text, and use the first type of graph to determine the expected value of each of the candidate customer interaction schemes, wherein the first type of graph includes: customer characteristics of the customer, scheme values ​​corresponding to the candidate customer interaction schemes, and probability weights corresponding to each customer characteristic and each candidate customer interaction scheme, the probability weights are used to represent the probability of successfully recommending the candidate customer interaction scheme to the customer with the customer characteristics, and the expected value is determined by the probability weights and the scheme values; Determine the customer intention corresponding to the customer question text, and determine the feasible customer interaction solution corresponding to the customer intention from the candidate customer interaction solutions according to the second type of graph, and determine the target customer interaction solution from the feasible customer interaction solutions according to the expected value, wherein the second type of graph includes multiple customer intentions and the candidate customer interaction solutions that can be executed to realize each customer intention; Obtain the workflow corresponding to the target customer interaction solution in the third type of graph, and display the steps in the workflow in sequence in the front-end interaction interface.

2. The method for generating a customer interaction plan according to claim 1, characterized in that: Determining the customer intent corresponding to the customer question text includes: Preprocessing the customer question text and converting the preprocessed customer question text into a word vector, wherein the preprocessing includes: word segmentation, stop word removal, and stem extraction; Determine the similarity between the word vector corresponding to the customer question text and the word vector corresponding to each customer intent in the preset customer intent catalog; The customer intention with the greatest similarity to the customer question text in the preset customer intention catalog is determined as the customer intention corresponding to the customer question text.

3. The method for generating a customer interaction plan according to claim 2, characterized in that: According to the second type of graph, determining a feasible customer interaction solution corresponding to the customer intention from the candidate customer interaction solutions includes: The candidate customer interaction plans that can be executed in the second type of graph to realize the customer intention are determined as the feasible customer interaction plans, wherein the feasible customer interaction plans include: service plans and marketing plans, and the marketing plans include: business plans and activity plans.

4. The method for generating a customer interaction plan according to claim 3, characterized in that: Determining the expected value of each of the candidate customer interaction solutions includes: Obtaining a customer portrait map of the customer corresponding to the customer question text, wherein the customer portrait map includes a plurality of customer features of the customer; Determine the probability weight of each of the customer characteristics for each of the candidate customer interaction solutions, and determine the success probability of recommending the candidate customer interaction solution to the customer based on the probability weights corresponding to all of the customer characteristics of the customer; The solution value corresponding to the candidate customer interaction solution in the first type of graph is obtained, and the expected value of the candidate customer interaction solution is determined based on the solution value and the success probability, wherein the solution value includes: economic value and management value.

5. The method for generating a customer interaction plan according to claim 4, characterized in that: Determining a target customer interaction solution from the feasible customer interaction solutions based on the expected value includes: Determine a first preset quantity and a second preset quantity corresponding to the customer intention; A first preset number of the service solutions among the feasible customer interaction solutions and a second preset number of the marketing solutions with the greatest expected value are determined as the target customer interaction solutions.

6. The method for generating a customer interaction plan according to claim 1, characterized in that: The workflow for obtaining the third type of graph corresponding to the target customer interaction solution includes: Obtaining a component corresponding to the target customer interaction solution, wherein the component includes: an atomic component and a composite component, one atomic component corresponding to executing an atomic operation of the target customer interaction solution, and one composite component encapsulated by a plurality of the atomic components; The logical connection relationship between each of the components is determined to obtain the workflow corresponding to the target customer interaction solution.

7. The method for generating a customer interaction plan according to claim 1, characterized in that: The method further comprises: Sending the workflow corresponding to the target customer interaction solution to the front-end interaction interface for display; In response to the operation instructions of the front-end interactive interface, according to the preset rules of the workflow component corresponding to the operation instructions, the various steps in the workflow are displayed in sequence in the front-end interactive interface to guide the interaction process between employees and users.

8. A customer interaction plan generating device, characterized in that: include: An expected value determination module is used to determine the candidate customer interaction scheme corresponding to the customer corresponding to the customer question text, and use the first type of map to determine the expected value of each candidate customer interaction scheme, wherein the first type of map includes: customer characteristics of the customer, the scheme value corresponding to the candidate customer interaction scheme, and the probability weight corresponding to each customer characteristic and each candidate customer interaction scheme, the probability weight is used to represent the probability of successfully recommending the candidate customer interaction scheme to the customer with the customer characteristics, and the expected value is determined by the probability weight and the scheme value; a target solution determination module, configured to determine the customer intention corresponding to the customer question text, and determine a feasible customer interaction solution corresponding to the customer intention from the candidate customer interaction solutions according to the second type of graph, and determine a target customer interaction solution from the feasible customer interaction solutions according to the expected value, wherein the second type of graph includes multiple customer intentions and candidate customer interaction solutions that can be executed to realize each customer intention; The workflow guidance module is used to obtain the workflow corresponding to the target customer interaction solution in the third type of graph, and to display the various steps in the workflow in sequence in the front-end interactive interface.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the processor is used to run a program stored in the memory, wherein the program, when running, executes the method for generating a customer interaction solution according to any one of claims 1 to 7.

10. A non-volatile storage medium, characterized in that: The non-volatile storage medium includes a stored computer program, wherein the device where the non-volatile storage medium is located executes the method for generating a customer interaction solution according to any one of claims 1 to 7 by running the computer program.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method for generating a customer interaction solution according to any one of claims 1 to 7 are implemented.

Images