Customer service reply method and device based on knowledge graph, electronic equipment and medium

Abstract

The embodiment of the invention provides a customer service reply method and device based on a knowledge graph, electronic equipment and a medium, and belongs to the technical field of artificial intelligence. The method comprises the following steps: acquiring customer service demand data; performing feature extraction on the customer service demand data to obtain customer service demand features; performing business operation retrieval on the customer service demand features from a pre-constructed business knowledge graph to obtain target business operation knowledge features; wherein the business knowledge graph is constructed based on customer service data; based on the customer service demand features and the target service operation knowledge features, constructing target service operation indication information; and inputting the target business operation indication information into a pre-trained business service operation model to generate target reply data matched with the customer service demand data through the business service operation model. Therefore, the customer service quality can be improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a customer service response method and device, electronic device and medium based on a knowledge graph. Background Art

[0002] At present, the customer service reply method usually uses a matching algorithm to match and calculate the business service content in the business service database with the customer service demand data, so as to determine the final customer service reply content. For example, in the body management service, the body management service database stores standardized customer body management plan content. If the customer service demand is to obtain an implementation plan to guide customers in body management, then the plan to guide customers in body management is found from the body management service database. However, this method can only simply match the service content stored in the database, which makes the accuracy of customer service replies low, and this method can only provide standardized service plans for different customers, and this method cannot provide continuous tracking of customer services, resulting in poor customer service quality. Therefore, how to improve the quality of customer service has become an urgent problem to be solved. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to propose a customer service response method and device, electronic device and medium based on knowledge graph, aiming to improve customer service quality.

[0004] To achieve the above objectives, a first aspect of an embodiment of the present application proposes a customer service reply method based on a knowledge graph, the method comprising:

[0005] Obtain customer service demand data;

[0006] Extracting features from the customer service demand data to obtain customer service demand features;

[0007] Performing a business operation search on the customer service demand feature from a pre-constructed business knowledge graph to obtain a target business operation knowledge feature; wherein the business knowledge graph is constructed based on customer service data;

[0008] Constructing target business operation instruction information based on the customer service demand characteristics and the target business operation knowledge characteristics;

[0009] The target business operation instruction information is input into a pre-trained business service operation model to generate target response data matching the customer service demand data through the business service operation model.

[0010] In some embodiments, the customer service demand feature is an ice-breaking service feature, the target business operation knowledge feature is an ice-breaking operation knowledge feature, and constructing the target business operation instruction information based on the customer service demand feature and the target business operation knowledge feature includes:

[0011] Acquire an ice-breaking reception service feature of the ice-breaking service feature, and acquire a reception operation knowledge feature from the ice-breaking operation knowledge feature based on the ice-breaking reception service feature;

[0012] Acquire an ice-breaking inquiry service feature of the ice-breaking service feature, and acquire an ice-breaking inquiry knowledge feature from the ice-breaking operation knowledge feature based on the ice-breaking inquiry service feature;

[0013] Acquire an ice-breaking consumption behavior service feature of the ice-breaking service feature, and acquire an ice-breaking consumption behavior knowledge feature from the ice-breaking operation knowledge feature based on the ice-breaking consumption behavior service feature;

[0014] The ice-breaking reception service feature, the reception operation knowledge feature, the ice-breaking inquiry service feature, the ice-breaking inquiry knowledge feature, the ice-breaking consumption behavior service feature and the ice-breaking consumption behavior knowledge feature are integrated to obtain an integrated ice-breaking service feature;

[0015] The fused ice-breaking service feature is input into a preset indication template to construct indication information, so as to generate ice-breaking service indication information as the target business operation indication information.

[0016] In some embodiments, the customer service demand feature is a customized service feature, the target business operation knowledge feature is a customized operation knowledge feature, and constructing the target business operation instruction information based on the customer service demand feature and the target business operation knowledge feature includes:

[0017] Acquire a customized customer guidance feature of the customized service feature, and acquire a customized guidance knowledge feature from the customized operation knowledge feature based on the customized customer guidance feature;

[0018] Acquire a customized service solution feature of the customized service feature, and acquire a customized solution knowledge feature from the customized operation knowledge feature based on the customized service solution feature;

[0019] Fusion of the customized customer guidance feature, the customized guidance knowledge feature, the customized service solution feature, and the customized solution knowledge feature to obtain a fused customized service feature;

[0020] The integrated customized service characteristics are input into a preset indication template to construct indication information, so as to generate customized service indication information as the target business operation indication information.

[0021] In some embodiments, the customer service demand feature is a remote supervision service feature, the target business operation knowledge feature is a remote supervision knowledge feature, and constructing the target business operation instruction information based on the customer service demand feature and the target business operation knowledge feature includes:

[0022] Acquire a remote guidance service feature of the remote supervision service feature, and acquire a remote guidance knowledge feature from the remote supervision knowledge feature based on the remote guidance service feature;

[0023] Acquire a remote supervision customer service feature of the remote supervision service feature, and acquire a remote supervision customer knowledge feature from the remote supervision knowledge feature based on the remote supervision customer service feature;

[0024] The remote guidance service feature, the remote guidance knowledge feature, the remote supervision customer service feature and the remote supervision customer knowledge feature are integrated to obtain an integrated remote supervision service feature;

[0025] The fused remote supervision service feature is input into a preset indication template to construct indication information, so as to generate remote supervision service indication information as the target business operation indication information.

[0026] In some embodiments, the customer service demand feature is a plan modification service feature, the target business operation knowledge feature is a plan modification knowledge feature, and constructing the target business operation instruction information based on the customer service demand feature and the target business operation knowledge feature includes:

[0027] Acquire the customer return service feature of the plan revision service feature, and acquire the return service knowledge feature from the plan revision knowledge feature based on the customer return service feature;

[0028] Acquire a correction service solution feature of the plan correction service feature, and acquire a correction solution knowledge feature from the plan correction knowledge feature based on the correction service solution feature;

[0029] The customer return service feature, the return service knowledge feature, the correction service plan feature and the correction plan knowledge feature are integrated to obtain a fusion plan correction service feature;

[0030] The fusion plan modification service characteristics are input into a preset indication template to construct indication information, so as to generate plan modification service indication information as the target business operation indication information.

[0031] In some embodiments, inputting the target business operation indication information into a pre-trained business service operation model to generate target response data matching the customer service demand data through the business service operation model includes:

[0032] Using the business service operation model to identify the business operation intention of the target business operation indication information, to obtain a business operation indication diagram;

[0033] Based on the business operation indication diagram, a business operation indication search is performed on the target business operation indication information to obtain business operation indication search information;

[0034] The business operation instruction search information is inferred and linked to obtain target response data that matches the customer service demand data.

[0035] In some embodiments, performing business operation retrieval on the customer service demand feature from the pre-built business knowledge graph to obtain the target business operation knowledge feature includes:

[0036] Acquire the business knowledge features of the business knowledge graph, and perform feature matching between the customer service demand features and the business knowledge features to obtain candidate business knowledge features;

[0037] Obtaining similarity scores of the candidate business knowledge features, and selecting a candidate business knowledge feature with the highest similarity score from the candidate business knowledge features as a first business operation knowledge feature based on the similarity scores;

[0038] In response to a feature instruction selected by a preset terminal, acquiring a second business operation knowledge feature from the candidate business knowledge features;

[0039] The target business operation knowledge feature is selected from the candidate business knowledge features based on the first business operation knowledge feature and the second business operation knowledge feature.

[0040] To achieve the above purpose, the second aspect of the embodiment of the present application proposes a customer service reply device based on a knowledge graph, the device comprising:

[0041] Customer service demand data acquisition module, used to obtain customer service demand data;

[0042] A feature extraction module, used to extract features from the customer service demand data to obtain customer service demand features;

[0043] A business operation retrieval module is used to perform business operation retrieval on the customer service demand characteristics from a pre-built business knowledge graph to obtain target business operation knowledge characteristics; wherein the business knowledge graph is constructed based on customer service data;

[0044] An instruction information construction module, used to construct target business operation instruction information based on the customer service demand characteristics and the target business operation knowledge characteristics;

[0045] The customer service demand response generation module is used to input the target business operation instruction information into a pre-trained business service operation model to generate target response data matching the customer service demand data through the business service operation model.

[0046] To achieve the above objectives, a third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the method described in the first aspect is implemented.

[0047] To achieve the above-mentioned purpose, the fourth aspect of an embodiment of the present application proposes a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in the first aspect is implemented.

[0048] The customer service reply method and device, electronic device and medium based on knowledge graph proposed in this application obtain customer service demand data; extract features of customer service demand data to obtain customer service demand features; retrieve business operations from pre-built business knowledge graphs to obtain target business operation knowledge features; wherein the business knowledge graph is constructed based on customer service data; based on customer service demand features and target business operation knowledge features, target business operation instruction information is constructed; the target business operation instruction information is input into a pre-trained business service operation model to generate target reply data that matches the customer service demand data through the business service operation model. In this way, the quality of customer service can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a flow chart of a customer service reply method based on knowledge graph provided in an embodiment of the present application;

[0050] Figure 2 yes Figure 1 Flow chart of step S103 in FIG.

[0051] Figure 3 yes Figure 1 Flow chart of step S104 in FIG.

[0052] Figure 4 yes Figure 1 Another flow chart of step S104 in FIG. 1 ;

[0053] Figure 5 yes Figure 1 Another flow chart of step S104 in FIG. 1 ;

[0054] Figure 6 yes Figure 1 Another flowchart of step S104 in;

[0055] Figure 7 yes Figure 1 Flow chart of step S105 in FIG.

[0056] Figure 8 It is a structural schematic diagram of a customer service reply device based on a knowledge graph provided in an embodiment of the present application;

[0057] Fig. 9 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0059] It should be noted that, although the functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first", "second", etc. in the specification, claims and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0061] First, some nouns involved in this application are analyzed:

[0062] Artificial intelligence (AI) is a new technical science that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. AI is a branch of computer science that attempts to understand the essence of intelligence and produce a new type of intelligent machine that can respond in a similar way to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing and expert systems. AI can simulate the information process of human consciousness and thinking. AI is also a theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0063] Based on this, the embodiments of the present application provide a customer service response method and device, electronic device and medium based on a knowledge graph, aiming to improve the quality of customer service.

[0064] The customer service response method and device based on knowledge graph, electronic device and medium provided in the embodiments of the present application are specifically illustrated through the following embodiments. First, the customer service response method based on knowledge graph in the embodiments of the present application is described.

[0065] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0066] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0067] The customer service reply method based on knowledge graph provided in the embodiment of the present application relates to the field of artificial intelligence technology. The customer service reply method based on knowledge graph provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or it can be configured as a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the customer service reply method based on knowledge graph, etc., but is not limited to the above forms.

[0068] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0069] Figure 1 is an optional flowchart of a customer service reply method based on a knowledge graph provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S105.

[0070] Step S101, obtaining customer service demand data.

[0071] Step S102: extract features from the customer service demand data to obtain customer service demand features.

[0072] Step S103, performing business operation retrieval on customer service demand features from a pre-constructed business knowledge graph to obtain target business operation knowledge features; wherein the business knowledge graph is constructed based on customer service data.

[0073] Step S104: construct target business operation instruction information based on customer service demand characteristics and target business operation knowledge characteristics.

[0074] Step S105 , inputting the target business operation instruction information into the pre-trained business service operation model, so as to generate target response data matching the customer service demand data through the business service operation model.

[0075] Steps S101 to S105 shown in the embodiment of the present application are as follows: obtaining customer service demand data; extracting features from the customer service demand data to obtain customer service demand features; performing business operation retrieval on the customer service demand features from the pre-constructed business knowledge graph to obtain target business operation knowledge features; wherein the business knowledge graph is constructed based on the customer service data; constructing target business operation instruction information based on the customer service demand features and the target business operation knowledge features; inputting the target business operation instruction information into the pre-trained business service operation model to generate target response data that matches the customer service demand data through the business service operation model. In this way, the quality of customer service can be improved.

[0076] In step S101 of some embodiments, specifically, customer service demand data refers to customer service's business service operation demand data, which may be at least one of ice-breaking service business operation demand data, customized service business operation demand data, remote supervision business operation demand data, and plan correction business operation demand data.

[0077] For example, in the field of body management services, the business operation demand data for ice-breaking services can be what ice-breaking words are used when receiving customers for the first time and understanding their weight and background information; the business operation demand data for customized services can be how to customize corresponding customer fat loss plans based on the input customer identity information; the business operation demand data for remote supervision can be what group messages are sent when remotely supervising customers' body management; the business operation demand data for plan revision can be how to update and customize fat loss plans based on the input customer weight and body fat changes; it can also be how to conduct ice-breaking consultations on customers' physical health status and customize corresponding customer fat loss plans based on the consultation results.

[0078] In step S102 of some embodiments, specifically, the customer service demand feature refers to the vector information of the customer service demand data in the semantic space. The customer service demand feature is the key information extracted from the customer service demand data. Therefore, the customer service demand feature can also be at least one of the customer service demand features including ice-breaking service features, customized service features, remote supervision service features and plan correction service features.

[0079] Specifically, first, the customer service demand data is preprocessed (including removing stop words, stemming and word form restoration, etc.), which helps to eliminate the noise of the customer service demand data and improve the data quality; secondly, the preprocessed customer service demand data is Tokenized to obtain multiple tokenized customer service demand data; finally, the segmented customer service demand data is Embedded to generate a vector representation for each token. These vectors are the customer service demand features.

[0080] For example, in the field of body management services, if the business operation demand data of the ice-breaking service is what is the ice-breaking words for understanding the basic information of the customer (such as age, height, weight, weight loss history, etc.) when first receiving the customer, then through Embedding, the quantized ice-breaking service characteristics of age, height, weight, weight loss history, ice-breaking, and words can be extracted; if the business operation demand data of the customization service is how to customize a personalized fat-reduction plan based on the input basic information of the customer, then through Embedding, the quantized customer basic information, personalization, and customization service characteristics of the fat-reduction plan can be extracted; if the business operation demand data of the remote supervision service is what is the group message about the dietary composition during the weight loss period, then through Embedding, the quantized remote supervision, weight loss period, dietary composition, and remote supervision service characteristics of group messages can be extracted; if the business operation demand data of the plan correction service is how to update and customize the fat-reduction plan according to the input customer weight and body fat changes, then through Embedding, the quantized customer weight change, body fat change, update, customization, fat loss, and plan correction service characteristics of the plan can be extracted.

[0081] In this embodiment, feature extraction of customer service demand data is a multi-dimensional and multi-level process, which involves extracting key demand information from customer service demand data and converting the key demand information into structured data that can be used for analysis, which helps to improve the efficiency of subsequently acquiring relevant knowledge features from the business knowledge graph.

[0082] See also Figure 2 In some embodiments, step S103 includes but is not limited to steps S201 to S204:

[0083] Step S201, obtain business knowledge features of the business knowledge graph, and perform feature matching between customer service demand features and business knowledge features to obtain candidate business knowledge features.

[0084] Step S202 : obtaining similarity scores of the candidate business knowledge features, and selecting a candidate business knowledge feature with the highest similarity score from the candidate business knowledge features as the first business operation knowledge feature based on the similarity scores.

[0085] Step S203: in response to the feature instruction selected by the preset terminal, a second business operation knowledge feature is acquired from the candidate business knowledge features.

[0086] Step S204: selecting a target business operation knowledge feature from candidate business knowledge features based on the first business operation knowledge feature and the second business operation knowledge feature.

[0087] In step S201 of some embodiments, specifically, before performing business operation retrieval on customer service demand features from a pre-constructed business knowledge graph to obtain target business operation knowledge features, the customer service response method based on the knowledge graph also includes: extracting business operation step entities from customer service data to obtain business operation step entities; extracting operation step entity relationships from the business operation step entities to obtain business operation entity relationships; and constructing a business knowledge graph based on the business operation step entities and business operation entity relationships.

[0088] Furthermore, named entity recognition technology can be used to identify entity reference items of each business operation step in the customer service data, and the entity reference items can be classified through the BERT network to obtain business operation step entities.

[0089] Specifically, in the field of weight loss, customer service data can be obtained from the customized weight loss operation manual stored in the enterprise database. The customer service data includes but is not limited to icebreaking service operation step data for customers' weight loss process, customized weight loss plan operation step data, weight loss remote supervision operation step data and revised customized weight loss plan operation step data, etc.

[0090] For example, in the field of weight loss, the ice-breaking operation step data may include but is not limited to guiding customers to their seats and greeting them, asking customers about the distance of their homes / units, asking customers about their origins, and asking customers about their information (such as age, weight, height). Through entity extraction, we can generate the customer reception operation step entity, the weight loss store source operation step entity, and the operation step entity for asking customers about their basic information, etc.

[0091] Furthermore, in the process of extracting the entity relationship of the operation steps, it is first necessary to identify the semantic connection between the business operation step entities, and based on the semantic connection, identify the subject-predicate-object structure and modification relationship in the sentence through dependency syntactic analysis and semantic role labeling to extract multi-level business operation entity relationships.

[0092] For example, in the field of weight loss, business operation step entities may include ice-breaking service operation step entities, customized service operation step entities, remote supervision operation step entities, and modified customized service operation step entities, and there is a first-to-last relationship between them. There is also a sequential relationship between the two ice-breaking service operation sub-step entities of the ice-breaking service operation step entity, namely, "introducing the customer to a seat and saying hello" and "asking about the customer's source", that is, "asking about the customer's source" is usually performed after "introducing the customer to a seat and saying hello". There is also a corresponding sequential relationship between the specific operation cases corresponding to the operation sub-step entities. The semantic connection between entities can be determined through entity relationship extraction, and the entity relationship between the business operation sub-steps contained in each business operation step and the specific case entity relationship contained in the business operation sub-steps can be extracted through dependency syntactic analysis to obtain multi-level entity relationships between weight loss service operation step entities.

[0093] Specifically, the business knowledge graph refers to the storage of customer service data-related knowledge through a graph network structure. The business operation step entities are used as nodes of the business knowledge graph, and the business operation step entity relationships are used as edges connecting the nodes, so that all business operation step entities and business operation step entity relationships of customer service data are stored through the Neo4j graph structure.

[0094] Furthermore, the node type of the business operation step entity is obtained, and the relationship type of the business operation step entity relationship is obtained, and a business knowledge graph is constructed based on the node type and the relationship type.

[0095] For example, in the field of weight loss, the node type of the seat greeting entity can be the ice-breaking service operation step entity's node type is the ice-breaking service operation type, the node type of the customized service operation step entity is the customized service operation type, and the relationship type between them is the follow-up type, that is, the operation step following the ice-breaking service operation step entity is the customized service operation type, and all entities and entity relationships are stored in a Neo4j graph structure to generate a business knowledge graph.

[0096] In this embodiment, by constructing a corresponding business knowledge graph for customer service data, it is possible to quickly find relevant business operation step knowledge by querying the business knowledge graph when the customer service responds later, provide accurate guidance for the customer service response, and help improve the efficiency and accuracy of customer service responses.

[0097] Specifically, business knowledge features refer to vector representations used to quickly retrieve and locate business operation knowledge features in the business knowledge graph. The business knowledge features include but are not limited to icebreaking service operation step features, customized service operation step features, remote supervision operation step features, and modified customized service operation step features, and each feature contains corresponding operation step sub-features.

[0098] For example, in the field of weight loss, if the customer service demand characteristic is what is the ice-breaking process when first receiving a customer, then the business knowledge characteristics are guiding the customer to sit down and greet them, asking the customer how far their home is from the weight loss store, and how the customer learned about the store.

[0099] Specifically, the target business operation knowledge characteristics refer to specific business operation characteristics that match the customer service demand characteristics.

[0100] Furthermore, the similarity score between customer service demand features and business knowledge features can be calculated using similarity algorithms such as cosine similarity, Euclidean distance, or Manhattan distance, and the top k business knowledge features with the highest similarity scores can be selected as candidate business knowledge features to search for business operation step entities that may meet customer service demand features in the business knowledge graph.

[0101] In this embodiment, by matching customer service demand features with business knowledge features, candidate business knowledge features are obtained. By comparing the similarities between vectors, candidate business knowledge features that may match the customer service demand features can be quickly retrieved, which helps to improve the efficiency of subsequent business operation retrieval.

[0102] In step S202 of some embodiments, specifically, the similarity score can be determined by calculating the similarity of the customer service demand features and the business knowledge features based on a cosine similarity algorithm.

[0103] Specifically, the first business operation knowledge feature is a business operation knowledge feature that is selected based on the similarity score and best matches the customer service demand feature.

[0104] Specifically, the similarity scores are sorted from high to low, and based on the sorted scores, a candidate business knowledge feature with the highest similarity score is selected from the candidate business knowledge features as the first business operation knowledge feature.

[0105] In this embodiment, based on the similarity score, the candidate business knowledge feature with the highest similarity score is selected from the candidate business knowledge features as the first business operation knowledge feature. The relationship between different candidate business knowledge features and customer service demand features can be quantified through the similarity score to determine the degree of matching between the features, thereby screening out the candidate knowledge feature with the highest degree of matching as the first business operation knowledge feature, which helps to improve the accuracy of business operation retrieval.

[0106] In step S203 of some embodiments, specifically, the preset terminal may be a mobile phone, iPad or other terminal.

[0107] Specifically, the selected feature instruction may be an option button instruction clicked by the customer service staff.

[0108] Specifically, after obtaining the candidate business knowledge features, the candidate business knowledge features are converted into candidate business knowledge texts, and the candidate business knowledge texts are displayed to the terminal through a pop-up window. The customer service personnel can select at least one candidate business knowledge text by clicking the option button corresponding to the candidate business knowledge text related to the customer service demand feature, and use the candidate business knowledge feature corresponding to the selected candidate business knowledge text as the second business operation knowledge feature.

[0109] In this embodiment, by responding to the feature instruction selected by the preset terminal, the second business operation knowledge feature is obtained from the candidate business knowledge features. The customer service personnel can select the business operation knowledge feature that is adapted to the customer service demand characteristics based on their work experience, so as to provide targeted high-quality services to different customers, which helps to improve the quality of subsequent customer service.

[0110] In step S204 of some embodiments, specifically, the target business operation knowledge feature refers to a feature that best matches the customer service demand feature.

[0111] Specifically, if the first business operation knowledge feature is consistent with the second business operation knowledge feature, the first business operation knowledge feature or the second business operation knowledge feature is used as the target business operation knowledge feature; if the first business operation knowledge feature is inconsistent with the second business operation knowledge feature, the second business operation knowledge feature is used as the target business operation knowledge feature.

[0112] For example, in the field of weight loss, if the customer service demand characteristic is what the ice-breaking process is, the first business operation knowledge characteristic is guiding customers to sit down and greet them, and the second business operation knowledge characteristic is asking about the customers' origins and how they know about the store, then the target business operation knowledge characteristic is asking about the customers' origins and how they know about the store.

[0113] In this embodiment, the target business operation knowledge feature is selected from the candidate business knowledge features based on the first business operation knowledge feature and the second business operation knowledge feature. The candidate business knowledge feature can be quickly selected through the business knowledge graph, and the final target business operation knowledge feature is selected in combination with the experience of customer service personnel, so that the most matching knowledge feature can be selected for different demand features, rather than just a simple matching of features, which facilitates the subsequent provision of personalized high-quality services to different customers and helps to improve the quality of subsequent customer service.

[0114] See also Figure 3 In some embodiments, step S104 includes but is not limited to steps S301 to S305:

[0115] Step S301, obtaining an ice-breaking reception service feature of the ice-breaking service feature, and obtaining a reception operation knowledge feature from the ice-breaking operation knowledge feature based on the ice-breaking reception service feature.

[0116] Step S302: obtaining an ice-breaking inquiry service feature of the ice-breaking service feature, and obtaining an ice-breaking inquiry knowledge feature from the ice-breaking operation knowledge feature based on the ice-breaking inquiry service feature.

[0117] Step S303, obtaining the ice-breaking consumption behavior service feature of the ice-breaking service feature, and obtaining the ice-breaking consumption behavior knowledge feature from the ice-breaking operation knowledge feature based on the ice-breaking consumption behavior service feature.

[0118] Step S304, integrating the ice-breaking reception service characteristics, reception operation knowledge characteristics, ice-breaking inquiry service characteristics, ice-breaking inquiry knowledge characteristics, ice-breaking consumption behavior service characteristics and ice-breaking consumption behavior knowledge characteristics to obtain integrated ice-breaking service characteristics.

[0119] Step S305: input the integrated ice-breaking service characteristics into a preset indication template to construct indication information, so as to generate ice-breaking service indication information as target business operation indication information.

[0120] In step S301 of some embodiments, specifically, the customer service demand feature may be an ice-breaking service feature.

[0121] Specifically, the target business operation knowledge feature may be an ice-breaking operation knowledge feature.

[0122] Furthermore, the ice-breaking service feature may be an ice-breaking reception service feature, and the ice-breaking reception service feature is used to represent reception guidance features for meeting a customer for the first time.

[0123] For example, in the field of weight loss, the characteristics of ice-breaking reception service can be how to guide customers to their seats and greet them when meeting them for the first time, how to ask customers how far their homes / units are from the weight loss store, how to ask customers where they learned about the weight loss store, etc.

[0124] Specifically, the reception operation knowledge features corresponding to the ice-breaking operation knowledge features are extracted from the business knowledge graph. The reception operation knowledge features are used to guide customer service staff in their reception actions and words when they first meet customers.

[0125] For example, in the field of weight loss, how to guide customers to take a seat and greet them, the corresponding reception operation knowledge features can be smiling, nodding, and greeting to guide customers to take a seat. How to ask customers how far their homes / units are from the weight loss store, the corresponding reception operation knowledge features can be "Is our store close?" How to ask customers about their knowledge of weight loss stores, the corresponding reception operation knowledge features can be "through what channels did you learn about brand B and store B?"

[0126] Furthermore, during the reception process, if the customer appears nervous, customer service can adopt a gentler and more patient attitude to appease the customer based on the guidance in the reception operation knowledge characteristics. This personalized reception method helps to ease the customer's nervousness and make them more willing to share personal information and specific weight loss needs, thus laying the foundation for the subsequent provision of personalized services.

[0127] In step S302 of some embodiments, specifically, the ice-breaking service feature may be an ice-breaking inquiry service feature.

[0128] For example, in the area of ​​weight loss, the ice-breaking inquiry service feature could be how to understand the customer's basic information and medical history.

[0129] Specifically, the ice-breaking inquiry knowledge feature corresponding to the ice-breaking operation knowledge feature is extracted from the business knowledge graph, and the ice-breaking inquiry knowledge feature is used to express the words for understanding the basic body condition of the customer.

[0130] For example, in the area of ​​weight loss, is it easy to understand your height, weight, body fat percentage, medical history? Or is it easy to understand your weight loss goals and health status?

[0131] In this embodiment, by obtaining the ice-breaking inquiry service characteristics and obtaining the ice-breaking consultation knowledge characteristics from the ice-breaking operation knowledge characteristics based on these characteristics, it can ensure that customer service can more accurately understand the customer's weight loss goals and weight loss plan requirements in the initial communication with the customer, which helps to quickly identify the customer's specific needs and expectations, thereby laying the foundation for the subsequent provision of personalized services.

[0132] In step S303 of some embodiments, specifically, the ice-breaking service feature may be an ice-breaking consumption behavior service feature.

[0133] For example, in the field of weight loss, the ice-breaking consumer behavior service feature can be how to understand the customer's budget for weight loss and the desire to purchase customized weight loss products.

[0134] Specifically, the ice-breaking consumption behavior knowledge features corresponding to the ice-breaking operation knowledge features are extracted from the business knowledge graph. The ice-breaking consumption behavior knowledge features are used to express the words for understanding the customer's consumption situation and the customer's customized weight loss products.

[0135] For example, in the area of ​​weight loss, is it convenient to understand your budget for weight loss based on an assessment of your body data?

[0136] In this embodiment, by obtaining the ice-breaking consumption behavior service characteristics and obtaining the ice-breaking consumption behavior knowledge characteristics from the ice-breaking operation knowledge characteristics, it is possible to better understand the customer's purchasing habits and preferences, so as to help customer service make customized weight loss product recommendations based on the customer's consumption history and budget.

[0137] In step S304 of some embodiments, specifically, the ice-breaking reception service features, reception operation knowledge features, ice-breaking inquiry service features, ice-breaking inquiry knowledge features, ice-breaking consumption behavior service features and ice-breaking consumption behavior knowledge features are spliced, and the interactions and common meanings between the ice-breaking features are learned, and finally the fused ice-breaking service features are output.

[0138] In this embodiment, by integrating the ice-breaking reception service characteristics, reception operation knowledge characteristics, ice-breaking inquiry service characteristics, ice-breaking inquiry knowledge characteristics, ice-breaking consumption behavior service characteristics and ice-breaking consumption behavior knowledge characteristics, it is possible to integrate information from multiple dimensions such as the customer's weight loss needs, weight loss plans, and weight loss consumption behaviors, which helps customer service personnel to fully understand customer needs, thereby providing more accurate and personalized services, and facilitating the subsequent improvement of customer service quality.

[0139] In step S305 of some embodiments, specifically, the instruction template is used to integrate icebreaking service features and icebreaking operation knowledge features to generate icebreaking service demand data that can be understood by the subsequent business service operation model.

[0140] For example, in the field of weight loss, the instruction template can be: [Customer service demand stage], asking questions about [ice-breaking weight loss service]. Based on the business knowledge graph, ice-breaking service guidance is given: [Weight loss service guidance].

[0141] For example, the customer service asked how to communicate with customers about weight loss during the ice-breaking stage. Based on the business knowledge graph, they received ice-breaking service guidance: first smile, nod, and greet the customer to guide them to sit down. Ask them if they are close to our store and ask them about their weight loss goals and health status (such as is it convenient to know your height, weight, body fat percentage, and medical history). Further, based on the customer's physical data assessment, ask them about their weight loss budget (such as is it convenient to know your budget for weight loss based on the assessment of your physical data?). Based on the customer's answer, guide the customer to purchase a customized weight loss plan.

[0142] In this embodiment, by integrating the icebreaking service characteristics into a preset instruction template to construct instruction information, icebreaking service instruction information is generated as target business operation instruction information, so that customer service personnel can take quick action based on the instruction information without having to reconsider how to interact with customers during each communication. This helps to quickly build customer trust, lays the foundation for subsequent personalized weight loss plans for customers, and significantly improves customer service quality.

[0143] See also Figure 4 In some embodiments, step S104 further includes but is not limited to steps S401 to S404:

[0144] Step S401 : obtaining customized customer guidance features of customized service features, and obtaining customized guidance knowledge features from customized operation knowledge features based on the customized customer guidance features.

[0145] Step S402: Obtain customized service solution features of the customized service features, and obtain customized solution knowledge features from the customized operation knowledge features based on the customized service solution features.

[0146] Step S403, the customized customer guidance features, customized guidance knowledge features, customized service solution features and customized solution knowledge features are integrated to obtain integrated customized service features.

[0147] Step S404: input the integrated customized service characteristics into a preset indication template to construct indication information, so as to generate customized service indication information as target business operation indication information.

[0148] In step S401 of some embodiments, specifically, the customer service demand feature may be a customized service feature.

[0149] Specifically, the target business operation knowledge feature may be a customized operation knowledge feature.

[0150] Furthermore, the customized service features may be customized customer guidance features.

[0151] For example, in the area of ​​weight loss, a customized customer coaching feature can provide personalized diet and exercise recommendations based on the customer's health status.

[0152] Specifically, the customized guidance knowledge features corresponding to the customized operation knowledge features are extracted from the business knowledge graph.

[0153] For example, in the field of weight loss, the customized guidance knowledge features a specific guidance plan for a one-week weight loss diet plan and a weight loss exercise arrangement.

[0154] In this embodiment, by obtaining customized customer guidance features and obtaining customized guidance knowledge features from customized operation knowledge features based on the customized customer guidance features, it is possible to ensure that customer service personnel can provide targeted weight loss guidance in initial communication with customers, which helps to build customer trust and improve the professionalism of the service.

[0155] In step S402 of some embodiments, specifically, the customized service solution feature may be an ice-breaking consumption behavior service feature.

[0156] For example, in the field of weight loss, the customized service plan feature may be the specific plan content of how to provide customized weight loss services to customers.

[0157] Specifically, the customized solution knowledge features corresponding to the customized operation knowledge features are extracted from the business knowledge graph.

[0158] For example, in the field of weight loss, a one-month customized plan with knowledge features including weight loss products, weight loss diet plan, weight loss exercise plan and body care plan is customized for customer A.

[0159] In this embodiment, by obtaining customized service plan characteristics and obtaining customized plan knowledge characteristics from customized operation knowledge characteristics based on these characteristics, customer service personnel can provide different customers with personalized weight loss service plans that meet their needs and budgets, thereby meeting the personalized weight loss needs of different customers and significantly improving the quality of customer service.

[0160] In step S403 of some embodiments, specifically, the customized customer guidance features, customized guidance knowledge features, customized service solution features, and customized solution knowledge features are vector-joined to achieve feature fusion.

[0161] In this embodiment, by integrating customized customer guidance features, customized guidance knowledge features, customized service plan features and customized plan knowledge features, the personalized weight loss needs of different customers and personalized weight loss plan customization are integrated, thereby improving the customer service quality during the customized service process.

[0162] In step S404 of some embodiments, specifically, a clear customized instruction guide is provided to the customer service personnel through an instruction template.

[0163] For example, the instruction template can be: [Customer service demand stage], asking questions about [customized weight loss service]. Based on the business knowledge graph, customized service guidance is given: [Customized weight loss plan guidance].

[0164] For example, in the field of weight loss, the instruction template can guide customer service staff to first understand the customer's basic information, weight loss goals and weight loss plan, then provide personalized diet and exercise suggestions based on this information, and finally guide customers to choose a customized weight loss plan that suits them.

[0165] In this embodiment, by inputting the integrated customized service characteristics into a preset instruction template to construct instruction information, customized service instruction information is generated as target business operation instruction information, which can instruct customer service personnel to accurately and quickly guide customers to choose appropriate customized weight loss plans, thereby further improving customer service quality.

[0166] See also Figure 5 In some embodiments, step S104 further includes but is not limited to steps S501 to S504:

[0167] Step S501 : acquiring a remote guidance service feature of a remote supervision service feature, and acquiring a remote guidance knowledge feature from a remote supervision knowledge feature based on the remote guidance service feature.

[0168] Step S502: acquiring a remote supervision customer service feature of the remote supervision service feature, and acquiring a remote supervision customer knowledge feature from the remote supervision knowledge feature based on the remote supervision customer service feature.

[0169] Step S503, the remote guidance service feature, the remote guidance knowledge feature, the remote supervision customer service feature and the remote supervision customer knowledge feature are integrated to obtain an integrated remote supervision service feature.

[0170] Step S504: input the integrated remote supervision service characteristics into a preset indication template to construct indication information, so as to generate remote supervision service indication information as target business operation indication information.

[0171] In step S501 of some embodiments, specifically, the customer service demand feature may be a remote supervision service feature.

[0172] Specifically, the target business operation knowledge feature may be a remote supervision knowledge feature.

[0173] Furthermore, the remote supervision service feature may be a remote guidance service feature.

[0174] For example, in the area of ​​weight loss, remote coaching services could be characterized as how to provide real-time coaching via video calls, or how to send customized exercise and diet plans via an app.

[0175] Specifically, remote supervision knowledge features corresponding to remote guidance service features are extracted from the business knowledge graph to represent specific remote guidance strategies and methods.

[0176] For example, in the field of weight loss, remote supervision knowledge features can be used to ask customers about their hunger and provide relative diet analysis and guidance, remind customers of their sleep time, and guide customers to exercise through videos.

[0177] In this embodiment, by identifying remote supervision service characteristics and acquiring corresponding remote guidance knowledge characteristics, customer service personnel can provide customers with specific guidelines and action steps, achieve continuous tracking of customer service, and further improve customer service quality.

[0178] In step S502 of some embodiments, specifically, the remote supervision service feature may be a remote supervision customer service feature.

[0179] For example, in the field of weight loss, the remote supervision customer service feature can be how to generate online message reminders about the customer's weight loss.

[0180] Specifically, remote supervision customer knowledge features corresponding to remote supervision knowledge features are extracted from the business knowledge graph.

[0181] For example, in the field of weight loss, the knowledge characteristics of remote supervision customers with a weekly cycle can be: Dear customer A, whether you have gained weight or maintained your original weight, I hope you will do offline exercise in time. Due to the cold weather, fat is most active and stores energy to prepare for winter, which is the rhythm of gaining weight. I hope you will continue to lose weight and build a good figure. Come on! And tomorrow the instructor of store D will come to the store to summarize the weight loss of the stage. I hope you can participate in it when you are free tomorrow.

[0182] In this embodiment, by obtaining remote supervision customer service characteristics and obtaining remote supervision customer knowledge characteristics from remote supervision knowledge characteristics based on the remote supervision customer service characteristics, it is helpful to monitor the customer's weight loss situation and give timely encouragement to the customer, thereby improving the customer's satisfaction with the weight loss service.

[0183] In step S503 of some embodiments, specifically, feature fusion can be achieved by vector concatenating the remote guidance service feature, the remote guidance knowledge feature, the remote supervision customer service feature, and the remote supervision customer knowledge feature.

[0184] In this embodiment, by integrating remote guidance service characteristics, remote guidance knowledge characteristics, remote supervision customer service characteristics and remote supervision customer knowledge characteristics, the customer's real-time execution and satisfaction in the customized weight loss plan are integrated, thereby improving the customer service quality during the remote supervision process.

[0185] In step S504 of some embodiments, specifically, a clear remote supervision instruction guide is provided to the customer service personnel through an instruction template.

[0186] For example, the instruction template can be: [Customer service demand stage], asked about [remote supervision weight loss service]. According to the business knowledge graph, remote supervision weight loss guidance is given: [Remote supervision weight loss guidance].

[0187] For example, in the area of ​​weight loss, an instruction template can guide customer service staff to first understand the customer’s current health status and weight loss goals through a video call, and then provide personalized diet and exercise recommendations based on this information, and regularly track the customer’s progress and feedback through the application.

[0188] In this embodiment, by integrating the remote supervision service characteristics into a preset instruction template to construct instruction information, remote supervision service instruction information is generated as target business operation instruction information, which can instruct customer service personnel to provide a clear remote supervision action guide, including how to use remote tools to communicate with customers, how to adjust service plans based on customer feedback, and how to ensure customer satisfaction and participation when using remote services, thereby further improving the quality of customer weight loss services.

[0189] See also Figure 6 In some embodiments, step S104 further includes but is not limited to steps S601 to S604:

[0190] Step S601, obtaining customer return service characteristics of the plan revision service characteristics, and obtaining return service knowledge characteristics from the plan revision knowledge characteristics based on the customer return service characteristics.

[0191] Step S602: Acquire a correction service solution feature of the planned correction service feature, and acquire a correction solution knowledge feature from the planned correction knowledge feature based on the correction service solution feature.

[0192] Step S603, integrating the customer return service characteristics, the return service knowledge characteristics, the correction service plan characteristics and the correction plan knowledge characteristics to obtain the integrated plan correction service characteristics.

[0193] Step S604: input the integrated plan modification service characteristics into a preset indication template to construct indication information, so as to generate plan modification service indication information as target business operation indication information.

[0194] In step S601 of some embodiments, specifically, the customer service demand feature may be a plan modification service feature.

[0195] Specifically, the target business operation knowledge feature may be a plan modification knowledge feature.

[0196] Furthermore, the plan revision service feature may be a customer return visit service feature.

[0197] For example, in the field of weight loss, the customer follow-up service feature can be how to follow up with customers about their weight loss status when they come to the store.

[0198] Specifically, the return visit service knowledge features corresponding to the customer return visit service features are extracted from the business knowledge graph.

[0199] For example, in the field of weight loss, the knowledge features of the follow-up service can be asking the customer about their weight gain or loss, asking about the customer's eating and exercise habits, and the difficulties the customer encounters in implementing the weight loss plan.

[0200] In this embodiment, by identifying the remote supervision service characteristics and obtaining the corresponding remote guidance knowledge characteristics, it is possible to continuously track the weight loss service provided to customers, understand the customer's satisfaction with the weight loss effect, and improve the quality of customer service.

[0201] In step S602 of some embodiments, specifically, the planned modification service feature may be a modification service scheme feature.

[0202] For example, in the field of weight loss, the characteristics of the modified service plan can provide modified service plan characteristics for how to combine the customer's current weight loss effect, weight loss mood, and weight loss goals.

[0203] Specifically, the correction plan knowledge features corresponding to the correction service plan features are extracted from the business knowledge graph.

[0204] For example, in the field of weight loss, the revised plan knowledge feature can be used to provide a revised weight loss plan for customers who encounter a plateau while implementing a weight loss plan, combining the customer's current weight loss effect, weight loss mood, and weight loss goals, with new diet suggestions, exercise plans, and weight loss products.

[0205] In this embodiment, by obtaining the correction service plan characteristics in the plan correction service characteristics, and obtaining the correction plan knowledge characteristics from the plan correction knowledge characteristics based on these characteristics, it is possible to provide customers with personalized weight loss update plans, enable customers to match appropriate weight loss plans at different weight loss stages, so as to help customers achieve their weight loss goals and improve their satisfaction with the weight loss service.

[0206] In step S603 of some embodiments, specifically, the customer return service features, the return service knowledge features, the revised service solution features and the revised solution knowledge features are vector-joined to achieve feature fusion.

[0207] In this embodiment, by integrating customer follow-up service characteristics, follow-up service knowledge characteristics, correction service plan characteristics and correction plan knowledge characteristics, customer service personnel are helped to fully understand the real-time situation of customer weight loss, so as to formulate more accurate correction weight loss plan measures, thereby improving the quality of customer weight loss service.

[0208] In step S604 of some embodiments, specifically, a clear solution modification guidance guide is provided to the customer service personnel through an instruction template.

[0209] For example, the instruction template can be: [Customer service demand stage], asked about [correction weight loss service]. According to the business knowledge map, the correction weight loss plan guidance is given: [correction weight loss plan guidance].

[0210] For example, in the field of weight loss, the instruction template can guide customer service staff to first ask the customer about the weight gain or loss and the difficulties the customer encounters during the weight loss period, then pay attention to the customer's emotional changes during the weight loss period, and finally generate a revised weight loss plan based on the customer's latest weight loss goal, weight loss effect, and weight loss tolerance.

[0211] In this embodiment, the fusion plan correction service characteristics are input into a preset instruction template to construct instruction information to generate plan correction service instruction information as target business operation instruction information, which facilitates the subsequent instruction business service operation model to generate an accurate correction weight loss plan to improve the accuracy of customer service responses.

[0212] See also Figure 7 In some embodiments, step S105 includes but is not limited to steps S701 to S703:

[0213] Step S701: Use the business service operation model to identify the business operation intention of the target business operation indication information to obtain a business operation indication diagram.

[0214] Step S702: performing a business operation instruction search on target business operation instruction information based on the business operation instruction diagram to obtain business operation instruction search information.

[0215] Step S703: perform inference linking on the business operation instruction search information to obtain target response data that matches the customer service demand data.

[0216] In step S701 of some embodiments, specifically, the business service operation model may be LLMs (Large Language Models), which is mainly composed of a Transformer network.

[0217] Specifically, Transformer can be used to extract knowledge keywords of the target business operation instruction information, and perform intent analysis on the instruction keyword content to determine the specific needs of customer service.

[0218] For example, in the field of weight loss, the model identifies that the customer service demand is that the customer wants to adjust a specific diet plan or exercise program, and then identifies that the business action indication is the intention to modify the weight loss program.

[0219] In this embodiment, by utilizing the business service operation model to identify the business operation intent of the target business operation instruction information, the instruction information can be deeply understood through a large language model to identify the core intent of the knowledge information and provide accurate responses, which helps to improve the accuracy of subsequent customer service responses.

[0220] In step S702 of some embodiments, specifically, the business operation indication search information refers to a series of business operation service steps that may be related to the business operation indication diagram.

[0221] Specifically, the business operation instruction search is performed on the target business operation instruction information based on the business operation instruction diagram in order to find a specific business operation instruction that matches the instruction diagram in the business knowledge graph.

[0222] For example, in the field of weight loss, the target business operation instruction information is how to implement the question about remote supervision of weight loss service. Regarding the question about remote supervision of weight loss service, the customer service staff can first understand the customer's current health status and weight loss goals through video calls, and then provide personalized diet and exercise suggestions based on this information, and regularly track the customer's progress and feedback through the application.

[0223] In step S703 of some embodiments, specifically, since the target response data is response data generated based on the customer service demand data, the target response data also includes at least one response content from the four stages of ice-breaking, customization, remote supervision and correction.

[0224] Specifically, the business operation instruction search information is reasoned and linked. This step involves matching the searched instruction information with the customer's specific needs and generating the final target answer data. In this process, the model may consider the customer's personal preferences, health status, historical feedback, and other relevant factors to ensure that the answer provided not only meets the standards of the business operation, but also meets the customer's personalized needs.

[0225] For example, if a client has specific dietary restrictions or health issues, the model will ensure that the recommended fat loss plan takes these factors into account, resulting in a personalized fat loss plan that is both safe and effective.

[0226] In this embodiment, by performing inference links on the business operation instruction search information, target response data that matches the customer service demand data is obtained, so as to respond to the instruction information generated based on knowledge characteristics and demand characteristics, generate precise business operation step response content, and improve the accuracy of customer service responses.

[0227] In an optional embodiment of the present application, after inferencing and linking the business operation instruction search information to obtain target reply data that matches the customer service demand data, the target reply data is displayed to the terminal device (such as a mobile phone, iPad) through a pop-up window. The customer service personnel can use control keys (such as an input box) to input targeted customer service replies based on the customer's weight loss needs, customer's weight loss goals, and customer's weight loss plans in combination with the target reply data, thereby generating more accurate customer service reply data based on customer needs, further improving the quality of customer service.

[0228] See also Figure 8 The embodiment of the present application also provides a customer service reply device based on a knowledge graph, which can implement the above-mentioned customer service reply method based on a knowledge graph, and the device includes:

[0229] Customer service demand data acquisition module, used to obtain customer service demand data;

[0230] A feature extraction module is used to extract features from customer service demand data to obtain customer service demand features;

[0231] The business operation retrieval module is used to perform business operation retrieval on customer service demand features from a pre-built business knowledge graph to obtain target business operation knowledge features; wherein the business knowledge graph is built based on customer service data;

[0232] An instruction information construction module, used to construct target business operation instruction information based on customer service demand characteristics and target business operation knowledge characteristics;

[0233] The customer service demand response generation module is used to input the target business operation instruction information into the pre-trained business service operation model to generate target response data matching the customer service demand data through the business service operation model.

[0234] The specific implementation of the customer service response device based on the knowledge graph is basically the same as the specific implementation example of the customer service response method based on the knowledge graph mentioned above, and will not be repeated here.

[0235] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned customer service reply method based on knowledge graph when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0236] See also Fig. 9 , Fig. 9 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0237] The processor 901 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0238] The memory 902 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 902 can store processing systems and other applications. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 902, and the processor 901 calls and executes the customer service reply method based on the knowledge graph in the embodiment of this application;

[0239] Input / output interface 903, used to implement information input and output;

[0240] Communication interface 904, used to realize communication interaction between the device and other devices, which can be realized by wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0241] A bus 905 that transmits information between various components of the device (e.g., the processor 901, the memory 902, the input / output interface 903, and the communication interface 904);

[0242] The processor 901 , the memory 902 , the input / output interface 903 and the communication interface 904 are connected to each other in communication within the device via a bus 905 .

[0243] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned customer service response method based on the knowledge graph.

[0244] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0245] The customer service reply method based on knowledge graph, the customer service reply device based on knowledge graph, the electronic device and the storage medium provided in the embodiment of the present application obtain customer service demand data; extract features of the customer service demand data to obtain customer service demand features; retrieve the customer service demand features from the pre-constructed business knowledge graph to obtain target business operation knowledge features; wherein the business knowledge graph is constructed based on the customer service data; based on the customer service demand features and the target business operation knowledge features, the target business operation instruction information is constructed; the target business operation instruction information is input into the pre-trained business service operation model to generate target reply data matching the customer service demand data through the business service operation model. In this way, the quality of customer service can be improved.

[0246] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0247] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0248] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0249] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0250] The terms "first", "second", "third", "fourth", etc. (if any) in the specification 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.

[0251] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0252] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. The mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0253] The units described above 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 network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0254] 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.

[0255] 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, 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 multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0256] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A customer service response method based on knowledge graph, characterized in that: The method comprises: Obtain customer service demand data; Extracting features from the customer service demand data to obtain customer service demand features; Performing a business operation search on the customer service demand feature from a pre-constructed business knowledge graph to obtain a target business operation knowledge feature; wherein the business knowledge graph is constructed based on customer service data; Constructing target business operation instruction information based on the customer service demand characteristics and the target business operation knowledge characteristics; The target business operation instruction information is input into a pre-trained business service operation model to generate target response data matching the customer service demand data through the business service operation model.

2. The method according to claim 1, characterized in that The customer service demand feature is an ice-breaking service feature, the target business operation knowledge feature is an ice-breaking operation knowledge feature, and constructing the target business operation instruction information based on the customer service demand feature and the target business operation knowledge feature includes: Acquire an ice-breaking reception service feature of the ice-breaking service feature, and acquire a reception operation knowledge feature from the ice-breaking operation knowledge feature based on the ice-breaking reception service feature; Acquire an ice-breaking inquiry service feature of the ice-breaking service feature, and acquire an ice-breaking inquiry knowledge feature from the ice-breaking operation knowledge feature based on the ice-breaking inquiry service feature; Acquire an ice-breaking consumption behavior service feature of the ice-breaking service feature, and acquire an ice-breaking consumption behavior knowledge feature from the ice-breaking operation knowledge feature based on the ice-breaking consumption behavior service feature; The ice-breaking reception service feature, the reception operation knowledge feature, the ice-breaking inquiry service feature, the ice-breaking inquiry knowledge feature, the ice-breaking consumption behavior service feature and the ice-breaking consumption behavior knowledge feature are integrated to obtain an integrated ice-breaking service feature; The fused ice-breaking service feature is input into a preset indication template to construct indication information, so as to generate ice-breaking service indication information as the target business operation indication information.

3. The method according to claim 1, characterized in that The customer service demand feature is a customized service feature, the target business operation knowledge feature is a customized operation knowledge feature, and the target business operation instruction information is constructed based on the customer service demand feature and the target business operation knowledge feature, including: Acquire a customized customer guidance feature of the customized service feature, and acquire a customized guidance knowledge feature from the customized operation knowledge feature based on the customized customer guidance feature; Acquire a customized service solution feature of the customized service feature, and acquire a customized solution knowledge feature from the customized operation knowledge feature based on the customized service solution feature; Fusion of the customized customer guidance feature, the customized guidance knowledge feature, the customized service solution feature, and the customized solution knowledge feature to obtain a fused customized service feature; The integrated customized service characteristics are input into a preset indication template to construct indication information, so as to generate customized service indication information as the target business operation indication information.

4. The method according to claim 1, characterized in that: The customer service demand feature is a remote supervision service feature, the target business operation knowledge feature is a remote supervision knowledge feature, and constructing the target business operation instruction information based on the customer service demand feature and the target business operation knowledge feature includes: Acquire a remote guidance service feature of the remote supervision service feature, and acquire a remote guidance knowledge feature from the remote supervision knowledge feature based on the remote guidance service feature; Acquire a remote supervision customer service feature of the remote supervision service feature, and acquire a remote supervision customer knowledge feature from the remote supervision knowledge feature based on the remote supervision customer service feature; The remote guidance service feature, the remote guidance knowledge feature, the remote supervision customer service feature and the remote supervision customer knowledge feature are integrated to obtain an integrated remote supervision service feature; The fused remote supervision service feature is input into a preset indication template to construct indication information, so as to generate remote supervision service indication information as the target business operation indication information.

5. The method according to claim 1, characterized in that The customer service demand feature is a plan modification service feature, the target business operation knowledge feature is a plan modification knowledge feature, and constructing the target business operation instruction information based on the customer service demand feature and the target business operation knowledge feature includes: Acquire the customer return service feature of the plan revision service feature, and acquire the return service knowledge feature from the plan revision knowledge feature based on the customer return service feature; Acquire a correction service solution feature of the plan correction service feature, and acquire a correction solution knowledge feature from the plan correction knowledge feature based on the correction service solution feature; The customer return service feature, the return service knowledge feature, the correction service plan feature and the correction plan knowledge feature are integrated to obtain a fusion plan correction service feature; The fusion plan modification service characteristics are input into a preset indication template to construct indication information, so as to generate plan modification service indication information as the target business operation indication information.

6. The method according to any one of claims 1 to 5, characterized in that: The step of inputting the target business operation instruction information into a pre-trained business service operation model to generate target response data matching the customer service demand data through the business service operation model includes: Using the business service operation model to identify the business operation intention of the target business operation indication information, to obtain a business operation indication diagram; Performing a business operation instruction search on the target business operation instruction information based on the business operation instruction diagram to obtain business operation instruction search information; The business operation instruction search information is inferred and linked to obtain target response data that matches the customer service demand data.

7. The method according to any one of claims 1 to 5, characterized in that: The business operation retrieval of the customer service demand features from the pre-built business knowledge graph to obtain target business operation knowledge features includes: Acquire the business knowledge features of the business knowledge graph, and perform feature matching between the customer service demand features and the business knowledge features to obtain candidate business knowledge features; Obtaining similarity scores of the candidate business knowledge features, and selecting a candidate business knowledge feature with the highest similarity score from the candidate business knowledge features as a first business operation knowledge feature based on the similarity scores; In response to a feature instruction selected by a preset terminal, acquiring a second business operation knowledge feature from the candidate business knowledge features; The target business operation knowledge feature is selected from the candidate business knowledge features based on the first business operation knowledge feature and the second business operation knowledge feature.

8. A customer service response device based on knowledge graph, characterized in that: The device comprises: Customer service demand data acquisition module, used to obtain customer service demand data; A feature extraction module, used to extract features from the customer service demand data to obtain customer service demand features; A business operation retrieval module is used to perform business operation retrieval on the customer service demand characteristics from a pre-built business knowledge graph to obtain target business operation knowledge characteristics; wherein the business knowledge graph is constructed based on customer service data; An instruction information construction module, used to construct target business operation instruction information based on the customer service demand characteristics and the target business operation knowledge characteristics; The customer service demand response generation module is used to input the target business operation instruction information into a pre-trained business service operation model to generate target response data matching the customer service demand data through the business service operation model.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the customer service response method based on knowledge graph as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the customer service response method based on knowledge graph described in any one of claims 1 to 7 is implemented.

Images