Method, device and equipment for matching agent based on knowledge graph and storage medium

Abstract

The application relates to a knowledge graph-based agent matching method and device, equipment and a storage medium, wherein the method comprises the following steps: obtaining incoming call information of a customer, and obtaining service category information determined by the customer through an interactive voice response; determining a current available agent in an agent list according to the service category information; the current available agent is online and in a ready state; determining a target agent matched with the customer from the current available agent according to the incoming call information and a preset knowledge graph; the knowledge graph at least comprises a corresponding relationship between the customer and the incoming call information and an association relationship between the customer and the agent; and establishing a call between the target agent and the customer. According to the method, the target agent matched with the customer can be determined from the current available agent determined based on the service category information through the incoming call information and the preset knowledge graph, so that the randomness and inaccuracy of agent matching can be solved, the matching degree of the agent and the customer is improved, and the satisfaction of the customer is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of customer service, and particularly relates to a method and device for matching an agent based on a knowledge graph, an apparatus and a storage medium. BACKGROUND

[0002] In power Internet technology (IT) operation and maintenance services, a voice call platform is an important channel. Users report daily problems, needs and faults through the voice call platform, and operation and maintenance personnel track and handle the problems.

[0003] In a traditional IT call system, after a client dials a phone number and selects a business skill group, the call platform selects an agent according to the agent queue, so as to realize the communication between the agent and the client. The traditional agent matching method only matches the agent condition, and the matching degree of the agent allocated by the system and the client is low, which affects the solution of the client problem, reduces the service quality, and affects the client satisfaction and experience. SUMMARY

[0004] To solve the technical problems of randomness and inaccuracy of agent matching in a power IT operation and maintenance call center, the present application provides a method and device for matching an agent based on a knowledge graph, an apparatus and a storage medium.

[0005] In a first aspect, the present application provides a method for matching an agent based on a knowledge graph, which comprises the following steps.

[0006] Obtaining the incoming call information of a client, and obtaining the service category information determined by the client through an interactive voice response;

[0007] According to the service category information, determining a current available agent in an agent list; the current available agent is online and in a ready state;

[0008] According to the incoming call information and a preset knowledge graph, determining a target agent matched with the client from the current available agents; the knowledge graph at least includes the corresponding relationship between the client and the incoming call information and the association relationship between the client and the agent;

[0009] Establishing a communication between the target agent and the client;

[0010] Optionally, before determining the target agent matched with the client from the current available agents according to the incoming call information and the preset knowledge graph, the method further comprises the following steps.

[0011] Obtaining the knowledge graph;

[0012] The construction process of the knowledge graph comprises the following steps.

[0013] acquire customer data and historical customer service records; the historical customer service records include at least one of historical traffic data, historical service work order data, agent data, agent service capability evaluation data and call service classification data;

[0014] extract at least one triple data of a customer according to the customer data and the historical customer service records respectively; the triple data includes data composed of the customer, a child node of the customer and an attribute of the child node; the child node at least includes an incoming information child node and a customer-agent association parameter child node; the incoming information child node represents the corresponding relationship between the customer and the incoming information; the customer-agent association parameter child node represents the association relationship between the customer and the agent;

[0015] construct the knowledge graph according to the triple data;

[0016] Optionally, according to the service category information, determining a current available agent in an agent list, comprising:

[0017] acquire an agent list, the agent list including at least one target classification agent list;

[0018] acquire a preset mapping relationship; the mapping relationship is a mapping relationship between a target classification agent list and service category information;

[0019] determine a target classification agent list from the agent list according to the service category information and the mapping relationship, and take an agent online and in a ready state in the target classification agent list as the current available agent;

[0020] Optionally, the incoming information at least includes an incoming number and an incoming area;

[0021] determine a target agent matching the customer from the current available agent according to the incoming information and a preset knowledge graph, comprising:

[0022] determine a target customer corresponding to the incoming number in the knowledge graph, and acquire a customer-agent association parameter of the target customer in the knowledge graph;

[0023] determine a list of agents corresponding to the incoming area from the current available agent;

[0024] determine an agent matching the target customer from the list of agents as the target agent according to the customer-agent association parameter;

[0025] Optionally, the customer-agent association parameter includes at least one of customer name, customer belonging unit, customer post, customer post level, customer sensitivity, customer complaint rate, customer satisfaction and customer preferred agent data.

[0026] According to the customer-agent association parameter, a customer agent in the agent list that matches the target customer is determined as the target agent, including:

[0027] A fitting degree weight corresponding to each customer-agent association parameter is obtained;

[0028] According to the customer-agent association parameter and the fitting degree weight, a score of each agent in the agent list is calculated;

[0029] According to the score of each agent, the agents in the agent list are sorted to obtain a sorting result, and an agent with the highest score in the sorting result is taken as the target agent;

[0030] Optionally, after establishing a call between the target agent and the customer, the method further includes:

[0031] A customer service record of the target agent for the customer is obtained;

[0032] Based on the customer service record, the knowledge graph is updated;

[0033] Optionally, according to the incoming information and a preset knowledge graph, a target agent that matches the customer is determined from the current available agents, including:

[0034] According to the incoming information and the preset knowledge graph, when the customer is determined to be a new customer, an agent satisfaction ranking table is obtained;

[0035] From the agent satisfaction ranking table, a current agent with the highest satisfaction in an online and ready state is determined, and the current agent with the highest satisfaction is taken as the target agent that matches the new customer.

[0036] In a second aspect, the present application provides a knowledge graph-based agent matching device, including:

[0037] An obtaining module is configured to obtain incoming information of a customer, and obtain service category information determined by the customer through an interactive voice response;

[0038] A determining module is configured to determine, according to the service category information, a current available agent in an agent list; the current available agent is online and in a ready state;

[0039] A matching module is configured to determine, according to the incoming information and a preset knowledge graph, a target agent that matches the customer from the current available agents; the knowledge graph at least includes a corresponding relationship between a customer and incoming information and an association relationship between a customer and an agent;

[0040] A call receiving module is configured to establish a call between the target agent and the customer.

[0041] In a third aspect, the present application provides an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus.

[0042] The memory is configured to store a computer program.

[0043] The processor is configured to execute the program stored on the memory, and implement the steps of the method for agent matching based on a knowledge graph according to any one of the first aspect.

[0044] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the method for agent matching based on a knowledge graph according to any one of the first aspect.

[0045] Compared with the prior art, the above technical solution provided by the embodiments of the present application has the following advantages:

[0046] The method provided by the embodiments of the present application can analyze the customer demand in depth, obtain the service category information determined by the customer through the interactive voice response, determine the current available agent in the agent list according to the service category information, and determine the target agent matched with the customer from the current available agent according to the incoming call information and the preset knowledge graph, so as to solve the randomness and inaccuracy of agent matching, improve the matching degree of the agent and the customer, improve the satisfaction and experience of the customer to the agent, and make the operation and maintenance service better serve the customer. BRIEF DESCRIPTION OF DRAWINGS

[0047] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present application and, together with the specification, serve to explain the principles of the application.

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0049] Figure 1 A system architecture diagram of the method for agent matching based on a knowledge graph provided by one embodiment of the present application;

[0050] Figure 2 A flowchart of the method for agent matching based on a knowledge graph provided by one embodiment of the present application;

[0051] Figure 3 A flowchart of a knowledge graph construction method provided by an embodiment of the present application is shown in the figure.

[0052] Figure 4 A structural diagram of a knowledge graph-based agent matching device provided by an embodiment of the present application is shown in the figure.

[0053] Figure 5 A structural diagram of an electronic device provided by an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0054] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in connection with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0055] The first embodiment of the present application provides a knowledge graph-based agent matching method, which can be applied to a system architecture as shown in the figure. Figure 1 The system architecture includes at least an agent matching system 101 and a customer 100. The customer 100 can establish a voice call with the agent matching system 101 through a cellular network. The number of customers 100 is not limited.

[0056] The method can be applied to the agent matching system 101 in the system architecture, and is used for matching an agent for any customer 100 calling in the agent matching system 101. For example, the agent matching system 101 can be a power IT operation and maintenance call center, or can be configured in the power IT operation and maintenance call center to match an agent for a customer 100 calling in through a call platform.

[0057] Next, a knowledge graph-based agent matching method based on the system architecture is described in detail. A knowledge graph-based agent matching method, as shown in the figure. Figure 2 The method includes the following steps.

[0058] Step 201: Obtain the calling-in information of the customer, and obtain the service category information determined by the customer through an interactive voice response.

[0059] The calling-in information of the customer can include the calling-in number, the calling-in time, the calling-in area corresponding to the calling-in number, and the like.

[0060] Interactive Voice Response, IVR for short, also known as intelligent IVR service, can guide the customer to determine the service category information to be obtained through key voice navigation.

[0061] In step 202, according to the service category information, a currently available agent in the agent list is determined, which is online and in a ready state.

[0062] In one embodiment, according to the service category information, the currently available agent in the agent list is determined, which can specifically include: obtaining an agent list, the agent list including at least one target classification agent list; obtaining a preset mapping relationship; the mapping relationship is a mapping relationship between the target classification agent list and the service category information; according to the service category information and the mapping relationship, a target classification agent list is determined from the agent list, and the agents in the target classification agent list which are online and in a ready state are taken as the currently available agents.

[0063] In this embodiment, the agent list can be grouped according to the service category in advance, i.e., into one or more target classification agent lists, each target classification agent list corresponding to a service category information, and the agents in the target classification agent list are good at handling and answering various problems corresponding to the service category. Of course, part of the agents in the target classification agent list can be offline or in a listening state, or a certain agent can be grouped into multiple target classification agent lists while listening to the phone of other service categories, so when determining the currently available agents, the agents online and in a ready state need to be filtered from the determined target classification agent list.

[0064] In this embodiment, by grouping the agent list, the agents can be classified, the business they are good at can be handled, the service level can be improved, the operation and maintenance service can better serve the customers, and thus the satisfaction and experience of the customers to the agents can be improved.

[0065] In this embodiment, the IT service categories can be classified according to the power IT call center responsibilities and IT service range, and the agents can be grouped according to the service categories, and the intelligent IVR key voice navigation service can be built between the customer and the agent based on the intelligent IVR service. Through the intelligent IVR key voice navigation service, the mapping relationship between the customer key data and the IT service classification can be set, the customer can complete the distribution of the customer traffic after selecting the service category to be obtained through the intelligent IVR key interaction, and at the same time, the filtering of the agents according to the selected service category is completed, and the list of agents which are currently available to respond and handle the customer's appeal and in a ready state is generated.

[0066] In step 203, a target agent matching the customer is determined from the current available agents according to the incoming information and a preset knowledge graph, and the knowledge graph at least includes the corresponding relationship between the customer and the incoming information and the association relationship between the customer and the agent.

[0067] The customer corresponding to the incoming information can be determined through the corresponding relationship between the customer and the incoming information in the preset knowledge graph, and the target agent matching the customer can be determined according to the association relationship between the customer and the agent in the knowledge graph.

[0068] In step 204, a call between the target agent and the customer is established.

[0069] In the embodiment, the service category information determined by the customer through the interactive voice response can be obtained through the deep analysis of the customer demand, the current available agent in the agent list can be determined according to the service category information, and the target agent matching the customer can be determined from the current available agents according to the incoming information and the preset knowledge graph, so as to solve the randomness and inaccuracy of agent matching, improve the matching degree of the agent and the customer, and thus improve the satisfaction and experience of the customer to the agent, so that the operation and maintenance service can better serve the customer.

[0070] In one embodiment, before the target agent matching the customer is determined from the current available agents according to the incoming information and the preset knowledge graph, the method further includes obtaining the knowledge graph.

[0071] The construction process of the knowledge graph includes at least: Figure 3

[0072] In step 301, customer data and historical customer service records are obtained, and the historical customer service records include at least one of historical traffic data, historical service work order data, agent data, agent service capability evaluation data and call service classification data.

[0073] In step 302, at least one triple data of the customer is extracted according to the customer data and the historical customer service records; the triple data includes data composed of the customer, a child node of the customer and an attribute of the child node; the child node at least includes an incoming information child node and a customer-agent association parameter child node; the incoming information child node represents the corresponding relationship between the customer and the incoming information; and the customer-agent association parameter child node represents the association relationship between the customer and the agent.

[0074] In step 303, a knowledge graph is constructed according to the triple data.

[0075] In the embodiment, the knowledge graph generation model and algorithm are derived from the current mature knowledge graph generation model for application, and comprehensively cover the whole process of knowledge extraction, knowledge fusion, knowledge reasoning and knowledge storage.

[0076] ​The data extracted when generating the knowledge graph can include customer data, historical call data, historical service ticket data, agent data, agent service capability evaluation data, and call service classification data. Based on the acquired data, focusing on customers, deep learning methods are used to extract triples of data such as entities, relationships, and attributes (also known as nodes, child nodes, and attributes). An entity refers to the customer, also known as a node; a child node represents the relationship with the customer; and an attribute represents the attribute representing the relationship between the child node and the customer. An example is given below:

[0077] The ternary data format can be as follows: (Zhang San, Inbound Time, June 22, 2022, 22:23:23), (Zhang San, Position, Responsibility), (Zhang San, Business Group, Asset Group), (Zhang San, Position Level, VIP / Important / Ordinary), (Zhang San, Sensitivity, 10⁻¹), (Zhang San, Complaint Rate, 10⁻¹), (Zhang San, Satisfaction, 10⁻¹). In this format, Inbound Time, Business Group, Position Level, Complaint Rate, and Satisfaction are child nodes of the node "Zhang San," while Responsibility and Asset Group are attributes of their respective child nodes.

[0078] Based on the extracted triplet data, deep learning algorithms are used to complete the summation and reasoning of knowledge, forming the content, relationships, and attribute elements of each node in the knowledge graph. Finally, knowledge data storage and visualization of the knowledge graph can be completed using the Neo4j graph database.

[0079] The generated knowledge graph reflects the relationships and attribute information between customers, agents, and service categories.

[0080] Specifically, customer call information sub-nodes can be extracted from customer data, such as the customer's mobile phone number and the region where the customer is located.

[0081] Customer agent association parameters can be extracted from historical customer service records. These parameters can include one or more of the following: customer name, company / organization, job title, job level, customer sensitivity, complaint rate, customer satisfaction, and preferred agent data. These parameters establish a certain correlation between the customer and the agent. For example, a higher-level customer might be assigned to a more capable agent, or a higher complaint rate might require a highly skilled and service-oriented agent. To improve agent matching accuracy, all of these parameters can be included.

[0082] It should be noted that the child nodes and attributes in the specific triplet data can be set as needed, and other graph databases can also be selected as needed. The above are just examples and do not represent any restrictions.

[0083] It should be noted that this knowledge graph can also include a pre-defined mapping relationship between the target category agent list and service category information.

[0084] In one embodiment, the incoming call information includes at least the incoming number and the incoming region. Based on the incoming call information and a preset knowledge graph, the target agent matching the customer is determined from the currently available agents. This includes: determining the target customer corresponding to the incoming number in the knowledge graph, and obtaining the customer agent association parameters of the target customer in the knowledge graph; determining the agent list corresponding to the incoming region from the currently available agents; and determining the agent matching the target customer from the agent list as the target agent based on the customer agent association parameters.

[0085] In this embodiment, the target customer is the customer found in the knowledge graph based on the call-in information; that is, the customer corresponding to the call-in information. The difference between the target customer and the customer is merely a different name and has no substantial distinction. Different call-in regions may correspond to different service category information. The agent list can be categorized by call-in region to improve the accuracy of agent matching.

[0086] In one embodiment, determining the target agent from the agent list based on the customer agent association parameters includes: obtaining the matching weight corresponding to each customer agent association parameter; calculating the score of each agent in the agent list based on the customer agent association parameters and the matching weight; sorting the agents in the agent list according to the scores of each agent to obtain a sorting result, and selecting the agent with the highest score in the sorting result as the target agent.

[0087] In this embodiment, the matching weight corresponding to the customer agent association parameter is related to the attribute of the sub-node of the customer agent association parameter. For example, when the customer agent association parameter is the customer complaint rate, the matching weight is related to the specific value of the customer complaint rate. If the complaint rate of agent A in the agent list is low, then agent A's score for this item will be high. Alternatively, the agent's score is related to the agent's satisfaction and complaint rate data.

[0088] Specifically, a customer intention prediction model and algorithm can be pre-set in the knowledge graph. The customer intention prediction model is built based on an attention mechanism, and the service capability matching weight of the agents is analyzed and calculated in conjunction with the attention mechanism. The customer's name, company, job position, job level, sensitivity, complaint rate, and satisfaction rate are used as inputs, as are the service category, satisfaction rate, complaint rate, and agent data associated with the customer. The service matching weight of each agent in the list of agents who are ready to respond to and handle customer requests is calculated. Based on the weights, the expected service requirements of the customer are predicted and analyzed. After the weights are calculated, the agents are sorted according to the weight from largest to smallest.

[0089] It should be noted that the attention mechanism refers to the ability of the customer intention prediction model to focus on its input parameters, i.e., to select specific inputs. In the case of limited computing power, the attention mechanism is a resource allocation scheme that allocates computing resources to more important tasks, which is a main means to solve the problem of information overload.

[0090] In one embodiment, after establishing the call between the target agent and the customer, the method further comprises: obtaining the customer service record of the target agent to the customer; and updating the knowledge graph based on the customer service record.

[0091] In this embodiment, the knowledge graph further comprises a knowledge graph completion model and algorithm. After the customer service is completed, the new customer service record is generated, and the knowledge graph completion algorithm is triggered to start, and the update and completion of the knowledge graph are completed based on the newly generated customer service record data, further improving the knowledge graph and supporting the next application.

[0092] In this step, the knowledge graph completion model and algorithm can be built based on a recurrent neural network (RNN) relationship reasoning model. After the new customer service record is generated, the update and completion of the knowledge graph data related to the customer are completed based on the new customer service record, further improving the knowledge graph content, and completing the continuous optimization of the knowledge graph.

[0093] It should be noted that in the above embodiment, the main purpose is to save the information of the customer corresponding to the incoming number in the knowledge graph, that is, the incoming customer is an old customer. When the incoming customer is a new customer, the target agent matching the customer is determined from the current available agents according to the incoming information and the preset knowledge graph, including: when it is determined that the customer is a new customer according to the incoming information and the preset knowledge graph, obtaining the agent satisfaction ranking table; determining the current highest satisfaction agent from the agent satisfaction ranking table, and taking the current highest satisfaction agent as the target agent matching the new customer.

[0094] In this embodiment, for a new customer, the agents are sorted from high to low according to the agent satisfaction, and the agent with the highest agent satisfaction is preferentially recommended to the customer to provide service to the customer and complete the service record. To ensure the service quality for new customers, and after completing the service record, the node, subnode and attribute information of the customer are established in the knowledge graph according to the customer service record.

[0095] Based on the same technical concept, the second embodiment of the present application provides a knowledge graph-based agent matching device, as shown in Figure 4 The device comprises:

[0096] The acquisition module 401 is configured to acquire incoming call information of a customer and acquire service category information determined by the customer through interactive voice response;

[0097] The determination module 402 is configured to determine a current available agent in an agent list according to the service category information; the current available agent is online and in a ready state.

[0098] The matching module 403 is configured to determine a target agent matched with the customer from the current available agents according to the incoming call information and a preset knowledge graph; the knowledge graph at least includes a corresponding relationship between a customer and incoming call information and an association relationship between a customer and an agent.

[0099] The answering module 404 is configured to establish a call between the target agent and the customer.

[0100] In the embodiment, the agent matching device can perform deep analysis in combination with customer demand, acquire service category information determined by the customer through interactive voice response, determine a current available agent in an agent list according to the service category information, and determine a target agent matched with the customer from the current available agents according to incoming call information and a preset knowledge graph, so as to solve the randomness and inaccuracy of agent matching, improve the matching degree of agents and customers, thereby improving customer satisfaction and experience degree for agents, and enabling operation and maintenance services to better serve customers.

[0101] As shown in Figure 5 The third embodiment of the present application provides an electronic device, which comprises a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112 and the memory 113 complete communication with each other through the communication bus 114,

[0102] The memory 113 is configured to store a computer program.

[0103] In one embodiment, the processor 111 is configured to execute the program stored in the memory 113 to implement the agent matching method based on the knowledge graph provided in any one of the preceding method embodiments, which comprises:

[0104] Acquiring incoming call information of a customer and acquiring service category information determined by the customer through interactive voice response;

[0105] Determining a current available agent in an agent list according to the service category information; the current available agent is online and in a ready state.

[0106] According to the incoming information and a preset knowledge graph, a target agent matched with the customer is determined from the current available agents; the knowledge graph at least includes a corresponding relationship between the customer and the incoming information and an association relationship between the customer and the agent;

[0107] A call between the target agent and the customer is established.

[0108] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0109] The communication interface is used for communication between the terminal and other devices.

[0110] The memory can include a Random Access Memory (RAM) and can also include a non-volatile memory, for example, at least one disk memory. Optionally, the memory can also be at least one storage device located away from the aforementioned processor.

[0111] The processor mentioned above can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; can also be a Digital Signal Processing (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0112] The fourth embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of a kind of agent matching method based on knowledge graph provided by any one of the preceding method embodiments:

[0113] Incoming information of a customer is acquired, and service category information determined by the customer through interactive voice response is acquired;

[0114] According to the service category information, a current available agent in the agent list is determined; the current available agent is online and in a ready state;

[0115] According to the incoming information and a preset knowledge graph, a target agent matching the customer is determined from the current available agents; the knowledge graph at least includes a corresponding relationship between the customer and the incoming information and an association relationship between the customer and the agent;

[0116] A call between the target agent and the customer is established.

[0117] In the above embodiments, all or part of them can be realized by software, hardware, firmware or any combination thereof. When realized by software, all or part of them can be realized in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, all or part of them produce the processes or functions described in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another, for example, the computer instructions can be transferred from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media. The available media can be magnetic media (for example, floppy disk, hard disk, magnetic tape), optical media (for example, DVD), or semiconductor media (for example, solid state disk (SSD)) and the like.

[0118] It should be noted that in this paper, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0119] It should be understood that the specific embodiments described herein merely exemplify the application and should not be considered limiting. In the description, the suffixes "module", "part" or "unit" used for components are merely intended for facilitation of explanation of the present application and by themselves do not have any specific meaning. Therefore, "module", "part" or "unit" can be mixedly used.

[0120] The above descriptions are merely specific embodiments of the present application to enable those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the generic principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the application. Accordingly, the present application is not to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A knowledge graph-based agent matching method, characterized in that, The method includes: Obtain customer call information, and obtain service category information determined by the customer through interactive voice response; Based on the service category information, determine the currently available agents in the agent list; the currently available agents are online and in a ready state; Based on the incoming call information and a preset knowledge graph, a target agent matching the customer is determined from the currently available agents; the knowledge graph includes at least the correspondence between the customer and the incoming call information and the association between the customer and the agent. Establish a call between the target agent and the customer; Before determining the target agent matching the customer from the currently available agents based on the incoming call information and a preset knowledge graph, the method further includes: Obtain the knowledge graph; The construction process of the knowledge graph includes: Obtain customer data and historical customer service records; the historical customer service records include at least one of historical call data, historical service ticket data, agent data, agent service capability evaluation data, and call service classification data; Based on the customer data and the historical customer service records, at least one triplet data for each customer is extracted; the triplet data includes data composed of the customer, the customer's child nodes, and the attributes of the child nodes; the child nodes include at least an inbound call information child node and a customer agent association parameter child node; the inbound call information child node represents the correspondence between the customer and the inbound call information; the customer agent association parameter child node represents the association between the customer and the agent. The knowledge graph is constructed based on the triplet data; The determination of currently available agents in the agent list based on the service category information includes: Obtain a list of agents, the list of agents including at least one target category of agent list; Obtain a preset mapping relationship; the mapping relationship is the mapping relationship between the target category seat list and the service category information; Based on the service category information and the mapping relationship, a target category seat list is determined from the seat list, and the online and ready seats in the target category seat list are taken as the currently available seats.

2. The method according to claim 1, characterized in that, The incoming call information includes at least the incoming number and the incoming region; Based on the incoming call information and a preset knowledge graph, a target agent matching the customer is determined from the currently available agents, including: Determine the target customer corresponding to the incoming number in the knowledge graph, and obtain the customer agent association parameters of the target customer in the knowledge graph; Determine the list of agents corresponding to the incoming call region from the currently available agents; Based on the customer seat association parameters, the seat that matches the target customer is determined from the seat list as the target seat.

3. The method according to claim 2, characterized in that, The customer agent association parameters include at least one of the following: customer name, customer's company, customer's position, customer's position level, customer sensitivity, customer complaint rate, customer satisfaction, and customer preferred agent data. Based on the customer seat association parameters, determining the seat in the seat list that matches the target customer as the target seat includes: Obtain the matching weight corresponding to each of the customer agent association parameters; Based on the customer agent association parameters and the matching weight, calculate the score for each agent in the agent list; The seats in the seat list are sorted according to the score of each seat to obtain a sorting result, and the seat with the highest score in the sorting result is selected as the target seat.

4. The method according to claim 1, characterized in that, After establishing a call between the target agent and the customer, the method further includes: Obtain the customer service records of the target agent for the customer; The knowledge graph is updated based on the customer service records.

5. The method according to claim 1, characterized in that, Based on the incoming call information and a preset knowledge graph, a target agent matching the customer is determined from the currently available agents, including: Based on the incoming call information and the preset knowledge graph, when the customer is determined to be a new customer, a customer service satisfaction ranking table is obtained. From the agent satisfaction ranking table, determine the agent with the highest current satisfaction who is online and ready, and select the agent with the highest current satisfaction as the target agent to match the new customer.

6. A knowledge graph-based agent matching device, characterized in that, The device includes: The acquisition module is used to acquire customer call information and service category information determined by the customer through interactive voice response. The determining module is used to determine the currently available seats in the agent list based on the service category information; the currently available seats are online and in a ready state; wherein, determining the currently available seats in the agent list based on the service category information includes: obtaining an agent list, the agent list including at least one target category agent list; obtaining a preset mapping relationship; the mapping relationship is a mapping relationship between the target category agent list and the service category information; determining a target category agent list from the agent list based on the service category information and the mapping relationship, and taking the online and ready seats in the target category agent list as the currently available seats; A matching module is used to determine a target agent matching the customer from the currently available agents based on the incoming call information and a preset knowledge graph. The knowledge graph includes at least the correspondence between the customer and the incoming call information and the association between the customer and the agent. The matching module is also used to acquire the knowledge graph. The construction process of the knowledge graph includes: acquiring customer data and historical customer service records; the historical customer service records include at least one of historical call data, historical service ticket data, agent data, agent service capability evaluation data, and call service classification data; extracting at least one triplet data for the customer based on the customer data and the historical customer service records; the triplet data includes data composed of the customer, the customer's child nodes, and the attributes of the child nodes; the child nodes include at least an incoming call information child node and a customer-agent association parameter child node; the incoming call information child node represents the correspondence between the customer and the incoming call information; the customer-agent association parameter child node represents the association between the customer and the agent; and constructing the knowledge graph based on the triplet data. The call answering module is used to establish a call between the target agent and the customer.

7. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the steps of the knowledge graph-based agent matching method according to any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the knowledge graph-based agent matching method as described in any one of claims 1-5.

Images