Customer service method, service device and computer equipment

Abstract

The invention provides a customer service method and device and computer equipment, and relates to the technical field of communication, and the customer service method comprises the steps: obtaining corresponding service requests submitted by a plurality of users in a plurality of different communication channels, and analyzing the corresponding intention information in the service requests based on a natural language processing technology; analyzing preset emotion information of each vocabulary or statement in the intention information; determining a service type required by the user according to the intention information, and matching a corresponding service resource based on the service type; determining a priority order of the service request according to the emotion information and the service resources, and allocating a customer service team or customer service staff corresponding to the service resources according to the priority order; after the service request is distributed, continuous monitoring and data analysis are performed to determine the service quality of the customer service team or the customer service staff for processing the service request, and the evaluation and guidance result is fed back to the corresponding customer service team or the corresponding customer service staff based on the service quality, so that the efficiency and the accuracy of the customer service can be improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a customer service method, service device and computer equipment. Background Art

[0002] With the rapid advancement of Internet technology, users' demand for online services has shown a significant growth trend. In the current digital era, whether it is shopping, entertainment, learning or work, the Internet has been deeply embedded in people's daily lives and has become an indispensable part. Therefore, the quality and efficiency of online services have become one of the important yardsticks for evaluating the market competitiveness of enterprises.

[0003] The traditional customer service model mainly relies on the manual customer service team to respond to user inquiries and problem solving. However, this model has gradually revealed several limitations in practice. The primary problem is that the response speed of manual customer service is difficult to guarantee, especially during peak hours, when users often have to wait for a long time before getting a response, which undoubtedly has a negative impact on user experience and satisfaction. Furthermore, manual customer service is affected by factors such as personal emotional fluctuations and differences in professional ability, and the service quality is unstable. Users may sometimes encounter customer service staff with cold attitudes, vague answers, or inability to effectively solve problems, which further exacerbates user dissatisfaction. Summary of the invention

[0004] The main purpose of the present invention is to provide a customer service method, aiming to improve the efficiency and accuracy of customer service, enhance user experience and satisfaction, and establish a closer and more efficient communication bridge between enterprises and users.

[0005] To achieve the above object, the present invention provides a customer service method, the method comprising: Obtain corresponding service requests submitted by multiple users through multiple different communication channels, and parse the corresponding intent information in the service requests based on natural language processing technology; and parse the preset emotional information of each word or sentence in the intent information; the intent information includes user's suggestions for product consultation, fault repair or complaint; Determine the service type required by the user according to the intention information, and match corresponding service resources based on the service type; Determine the priority ranking of the service request according to the emotion information and the service resource, and assign a customer service team or customer service personnel corresponding to the service resource according to the priority ranking; After the service request is assigned, continuous monitoring and data analysis are performed to determine the service quality of the customer service team or customer service personnel in handling the service request, and based on the service quality, feedback evaluation and guidance results are given to the corresponding customer service team or corresponding customer service personnel.

[0006] Optionally, the parsing the corresponding intent information in the service request based on natural language processing technology; and parsing the emotion information preset in each word or sentence in the intent information includes: Parsing multiple text messages in the service request based on natural language, and combining the multiple text messages according to the text messages to generate corresponding service request paragraphs; According to the context and semantics of the service request segment, identify the user's main appeal points and problem details; Matching the intention information according to the main appeal points and the question details; Extracting keywords from the intention information based on keyword recognition technology to generate multiple words or sentences; Assigning an initial emotion value to each word or sentence based on a preset emotion vocabulary, wherein the preset emotion vocabulary includes common words and emotion labels corresponding to each common word; Use a pre-trained word embedding model to convert each word or sentence into a vector representation to generate vectorized sentences; Identifying the sentence structure of the vectorized sentence based on syntactic analysis technology, wherein the sentence structure includes a subject-predicate-object relationship, modifiers, and conjunctions; Determining contexts of a plurality of the words or the sentences according to the sentence structure; According to the context of the words or the sentences, the initial emotion value and the sentence structure, a pre-trained sentiment analysis model is used to perform an overall emotion assessment on the context of the multiple words or the sentences, and a corresponding emotion category and intensity are generated to determine an emotion recognition result of the user; The user's personal information and historical service request records are obtained, and the emotion recognition result is corrected based on the personal information and the historical service request records to determine the corresponding preset emotion information.

[0007] Optionally, matching the intention information according to the main appeal point and the question details includes: According to the main appeal points and the problem details, the text is parsed and segmented through natural language processing technology to identify the key concepts and themes in the user's expression; Based on a preset intent recognition model, the key concepts and the topics are classified, and intent matching and similarity calculation are performed to determine the intent category that best matches the user's service request; The intent information is determined according to the intent category, the main appeal point and the question details.

[0008] Optionally, the method for presetting emotional information includes: Acquire multiple historical service request records in the historical customer service process, and construct a user portrait model corresponding to the historical service request records; Analyzing user feedback in the historical service request records, wherein the user feedback includes user evaluation of service quality or satisfaction with service content; Training the user portrait model based on a machine learning algorithm to determine whether user feedback in the user portrait model is positive or negative, and generating a comprehensive sentiment score corresponding to each of the historical service request records; Analyze the service attribute data in the historical service request record, wherein the service attribute data includes the service type, processing time or the customer service personnel who processed the request; Associating the comprehensive emotion score with the service attribute data to identify key factors affecting user emotions; Importing the key factors into the user portrait model for training to determine the possible emotional response of the user corresponding to any of the historical service request records; Testing the trained user portrait model based on the test set to verify the accuracy and stability of the trained user portrait model; When the accuracy and stability of the trained user portrait model reach a preset threshold, generating a sentiment prediction model; Multiple emotion prediction models in the emotion prediction model are matched one by one with corresponding intention information in the historical service request record to generate the preset emotion information.

[0009] Optionally, determining the service type required by the user according to the intent information, and matching corresponding service resources based on the service type includes: According to the intent information, a service type corresponding to the intent information is matched through a preset service type mapping table, wherein the preset service type mapping includes a service type corresponding to each of the intent information; After determining the service type, obtaining the user's service request history and service preferences to determine the user's corresponding personalized service requirements; According to the personalized service requirements, query the preset service resource database to determine a list of service resources matching the service type; the preset service resource database includes multiple service resources, and the multiple service resources include a manual customer service team, an intelligent customer service system or a self-service tool; each of the service resources is marked with a corresponding service type, service scope, service efficiency and user evaluation; Analyze the availability, efficiency and user evaluation of each service resource in the service resource list, comprehensively evaluate and select the most suitable service resource.

[0010] Optionally, querying a preset service resource database according to the personalized service requirement to determine a service resource list matching the service type includes: Filter out service resources that match the current service request from a preset service resource database; According to the personalized service requirements, the screened service resources are prioritized to obtain a ranking result; wherein the priority ranking is based on a comprehensive ranking of the service resources' response time, historical processing success rate, and user satisfaction evaluation; According to the ranking result, determine the first N service resources as a candidate service resource list, where N is a preset threshold; The analyzing the availability, efficiency and user evaluation of each service resource in the service resource list, and comprehensively evaluating and selecting the most suitable service resource include: Evaluate whether each of the service resources in the candidate service resource list can meet the user's current service request, and generate a corresponding evaluation result; the evaluation requirements include the availability of the service resource and whether the service scope covers the content requested by the user; Based on the evaluation result, at least one service resource is determined as a service resource recommended to the user.

[0011] Optionally, the evaluating whether each of the service resources in the candidate service resource list can satisfy the user's current service request and generating a corresponding evaluation result includes: According to the service resources in the candidate service resource list, the current performance indicators of each service resource are obtained one by one to check the current working status, load condition and ability to process user requests of each service resource, wherein the performance indicators include response time, concurrent processing capability and historical processing success rate; Refine and quantify the user's service request to determine the specific content and specific requirements of the service request, to determine the complexity and urgency of the service request and the user's expectations of service quality, and to determine the corresponding evaluation criteria; Compare the performance indicators of service resources with the service request requirements of users to evaluate whether the service resources can meet the needs of users; If the performance indicators of the service resources meet or exceed the user's requirements, it is considered that the service resources can meet the user's current service request and generate a positive evaluation result; If the performance indicators of the service resources fail to meet the user's requirements, a negative evaluation result is generated and the deficiencies are marked.

[0012] Optionally, after the service request is assigned, the continuous monitoring and data analysis to determine the service quality of the customer service team or customer service personnel in handling the service request includes: Obtaining user feedback data during customer service, including user satisfaction scores, interaction records during the service process, and user suggestions for service improvements; Classify and sort the feedback data, and extract key information points of the feedback data according to different classifications; Based on the key information points, and through machine learning algorithms or statistical analysis methods, potential optimization points in the service process are identified to determine the corresponding user behavior patterns or service efficiency bottlenecks in the customer service process, and generate corresponding analysis results; Generate corresponding improvement and adjustment measures based on the analysis results, which include optimizing service processes, improving service efficiency, or improving user interface design; After implementing improvement and adjustment measures, continuously monitor user feedback and service results to iteratively optimize the improvement and adjustment measures.

[0013] In addition, to achieve the above objectives, the present invention also provides a service device, which includes: a memory, a processor, and a customer service program stored in the memory and executable on the processor, wherein the customer service program is configured to implement the customer service method described above.

[0014] In addition, to achieve the above object, the present invention also provides a computer device, including the service device as described above.

[0015] The embodiment of the present invention obtains corresponding service requests submitted by multiple users in multiple different communication channels, and then parses the corresponding intent information in the service request based on natural language processing technology; and parses the emotional information preset by each word or sentence in the intent information, and then determines the service type required by the user based on the intent information, and determines the priority ranking of the service request based on matching the corresponding service resources with the emotional information based on the service type, and allocates the customer service team or customer service personnel corresponding to the service resources according to the priority ranking. After the service request is allocated, continuous monitoring and data analysis are carried out to determine the service quality of the customer service team or customer service personnel in handling the service request, and based on the service quality feedback evaluation and guidance results to the corresponding customer service team or the corresponding customer service personnel, the efficiency and accuracy of customer service can be improved, and the user experience and satisfaction can be enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0018] Figure 1 It is a flowchart of a customer service method according to an embodiment of the present invention; Figure 2 A schematic diagram of a customer service method flow chart of another embodiment of the present invention; Figure 3 A schematic diagram of a customer service method flow chart of another embodiment of the present invention; Figure 4 A schematic diagram of a customer service method flow chart of yet another embodiment of the present invention; Figure 5 A schematic diagram of a customer service method flow chart of another embodiment of the present invention; Figure 6 A schematic diagram of a customer service method flow chart of another embodiment of the present invention; Figure 7 A schematic diagram of a customer service method flow chart of another embodiment of the present invention; Figure 8 The figure is a flowchart of a customer service method according to another embodiment of the present invention.

[0019] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments, and the well-known modules, units and their connections, links, communications or operations are not shown or described in detail. In addition, the described features, architectures or functions can be combined in any way in one or more embodiments. It should be understood by those skilled in the art that the various embodiments described below are only used for illustration and are not used to limit the scope of protection of the present invention. It can also be easily understood that the modules or units or processing methods in each embodiment shown in this article and the drawings can be combined and designed according to various different configurations. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0021] The limitations of various nouns or methods referred to in the following embodiments, except for those that are logically untenable, are generally based on the broad concepts that can be implemented under the premise of the disclosure in the embodiments. Under such understanding, various specific subordinate specific limitations of nouns or methods should be regarded as the inventive content of the present invention, and should not be understood narrowly or interpreted in a biased manner on the grounds that the specification does not disclose the specific limitations. Similarly, under the premise that it can be implemented logically, the order of the steps in the method is flexible and changeable, and the specific subordinate specific limitations in the broad concepts of various nouns or methods all fall within the scope of protection of the present invention.

[0022] The main solution of the embodiment of the present application is: by obtaining corresponding service requests submitted by multiple users in multiple different communication channels, and then parsing the corresponding intent information in the service request based on natural language processing technology; and parsing the emotional information preset by each word or sentence in the intent information, and then determining the service type required by the user based on the intent information, and matching the corresponding service resources based on the service type combined with the emotional information, determine the priority ranking of the service request, and allocate the customer service team or customer service personnel corresponding to the service resources according to the priority ranking, and after the service request is allocated, continuously monitor and analyze the data to determine the service quality of the customer service team or customer service personnel in handling the service request, and feedback evaluation and guidance results based on the service quality.

[0023] In this embodiment, for the convenience of description, the following description is made with the service device as the execution subject.

[0024] Due to the traditional customer service model of existing technology, it mainly relies on the manual customer service team to respond to user inquiries and problem solving. However, this model has gradually revealed several limitations in practice. The primary problem is that the response speed of manual customer service is difficult to guarantee, especially during peak hours, when users often have to wait for a long time before getting a response, which undoubtedly has a negative impact on user experience and satisfaction. Furthermore, manual customer service is affected by factors such as personal emotional fluctuations and differences in professional ability, and the service quality is unstable. Users may sometimes encounter customer service staff with cold attitudes, vague answers, or inability to effectively solve problems, which further exacerbates user dissatisfaction.

[0025] This application provides a solution that can improve the efficiency and accuracy of customer service and enhance user experience and satisfaction.

[0026] To this end, the present invention proposes a customer service method; it can be understood that a service device for storing and executing the following method is provided in the computer device, and the service device can be implemented by a main controller, such as MCU (Microcontroller Unit), DSP (Digital Signal Process), FPGA (Field Programmable Gate Array), SOC (System On Chip), etc.

[0027] Reference Figure 1 In one embodiment of the present invention, the customer service method includes steps S100-S400, wherein: S100, obtaining corresponding service requests submitted by multiple users in multiple different communication channels, parsing corresponding intent information in the service requests based on natural language processing technology; and parsing the emotion information preset in each word or sentence in the intent information; S200, determining the service type required by the user according to the intent information, and matching corresponding service resources based on the service type; S300, determining the priority ranking of service requests according to the emotional information and service resources, and allocating a customer service team or customer service personnel corresponding to the service resources according to the priority ranking; S400. After the service request is assigned, continuous monitoring and data analysis are performed to determine the service quality of the customer service team or customer service personnel in handling the service request, and feedback evaluation and guidance results based on the service quality to the corresponding customer service team or corresponding customer service personnel.

[0028] Among them, intent information includes users' product consultation, fault repair or complaint suggestions.

[0029] In this embodiment, the communication channels may include telephone hotlines, online chat platforms, emails, social media channels, and mobile application customer service portals. Each communication channel carries different forms of user service requests, and the service device can use a unified interface to receive and integrate these requests to ensure the comprehensiveness and consistency of information.

[0030] For telephone hotlines, we can use advanced speech recognition technology to convert user voice into text, and then analyze intentions and emotions. Online chat platforms and social media channels directly receive user input in text form, and quickly capture users' real needs and potential emotions through natural language processing (NLP) algorithms. Emails use email parsing technology to automatically extract key information and process it. As for the mobile application customer service portal, we have integrated a variety of interactive methods, such as text, voice, and pictures, to meet users' needs in different scenarios.

[0031] When parsing intent information, we need to not only focus on the needs explicitly expressed by users, but also explore the potential needs of users through contextual understanding and semantic expansion. At the same time, by pre-setting a rich emotional vocabulary library and sentence templates, we can accurately identify the positive, negative or neutral emotions expressed by users, providing an important basis for subsequent priority sorting and service allocation.

[0032] When matching service resources, we divide the customer service team or customer service personnel into multiple professional groups according to the different service types. Each group has the expertise and ability to handle specific types of service requests to ensure the efficiency and professionalism of the service. At the same time, we have also established a dynamic adjustment mechanism for service resources to flexibly adjust resource allocation according to the actual situation of service requests and the load of the customer service team to ensure timely response and processing of service requests.

[0033] In terms of priority sorting, multiple factors are considered, including the matching degree of emotional information and service resources, the urgency of service requests, and the user's history. Through comprehensive evaluation through the algorithm model, reasonable priority labels are assigned to service requests to ensure that important and urgent requests are given priority.

[0034] Finally, after the service request is assigned, a service quality monitoring and feedback mechanism is established. Through data analysis tools, the service quality of the customer service team or customer service personnel is continuously monitored and evaluated, and problems are discovered and improved in a timely manner. At the same time, users can also provide feedback on the service process through the evaluation system, and use user opinions and suggestions as an important basis for improving service quality. These feedback results will directly guide the corresponding customer service team or customer service personnel to make improvements and upgrades, thereby continuously improving our customer service quality and user satisfaction.

[0035] In actual application, for example, a service request for product consultation submitted by user A on an online chat platform is obtained, and the content is, for example: "The XX model smart bracelet I bought cannot connect to the mobile phone. How can I solve this problem?" Through natural language processing technology, the service device first parses the intent information of the service request as the user's need for product consultation, and identifies the user's emotion as "confused" or "anxious."

[0036] Next, the service device matches the service type of "product consultation" with a manual customer service team or an intelligent customer service system and gives it to the user. In the priority sorting stage, since this service request does not show a very high degree of urgency (such as no mention of issues that affect the user's health or safety), and the user's emotions are relatively stable and not obviously anxious, the service device classifies it as a medium priority for processing. At the same time, since there are no similar high-frequency problems or negative feedback in the user's history records, there is no additional impact on the priority. Based on this, the corresponding customer service team responsible for the technical support of smart bracelet products is assigned. This team is not only proficient in product operation and troubleshooting, but also has the ability to respond quickly.

[0037] After the service request was assigned to the corresponding customer service staff, the customer service staff immediately contacted user A through the online chat platform and asked in detail about the specific circumstances of the problem, including the model of the bracelet, the model of the mobile phone, the solutions that have been tried, etc. Based on this information, the customer service staff quickly located the cause of the problem and gave specific operation steps to guide the user to try to connect.

[0038] During the communication between the customer service staff and the user, the statements in the communication process are continuously monitored and data analysis is performed. Assuming that the customer service staff actively responds to every question of the user and successfully helps user A solve the problem of connecting the smart bracelet to the mobile phone, the customer service staff can be given a relatively good service quality evaluation based on this; if it is monitored that the customer service staff is obviously unprofessional or perfunctory in the reply process, resulting in the problem not being solved or user complaints, feedback of low-quality service is output to the customer service staff, and their unprofessionalism is pointed out and adjusted.

[0039] This embodiment obtains corresponding service requests submitted by multiple users in multiple different communication channels, and then parses the corresponding intent information in the service request based on natural language processing technology; and parses the emotional information preset in each word or sentence in the intent information, and then determines the service type required by the user based on the intent information, and determines the priority ranking of the service request based on matching the corresponding service resources based on the service type combined with the emotional information, and allocates the customer service team or customer service personnel corresponding to the service resources according to the priority ranking. After the service request is allocated, continuous monitoring and data analysis are carried out to determine the service quality of the customer service team or customer service personnel in handling the service request, and feedback evaluation and guidance results based on the service quality are given to the corresponding customer service team or the corresponding customer service personnel, which can improve the efficiency and accuracy of customer service and enhance user experience and satisfaction.

[0040] Optionally, refer to Figure 2 Another embodiment of the present invention provides a customer service method based on the above Figure 1In the embodiment shown, parsing the corresponding intent information in the service request based on natural language processing technology; and parsing the preset emotion information of each word or sentence in the intent information includes steps S1010-S1100, wherein: S1010, parsing multiple text information in the service request based on natural language, and combining the multiple text information according to the text information to generate a corresponding service request paragraph; S1020, identifying the user's main appeal and question details according to the context and semantics of the service request segment; S1030, matching intention information according to the main appeal points and problem details; S1040, extracting keywords from the intention information based on keyword recognition technology to generate multiple words or sentences; S1050, assigning an initial emotion value to each word or sentence based on a preset emotion vocabulary; S1060, using a pre-trained word embedding model to convert each word or sentence into a vector representation to generate a vectorized sentence; S1070, identifying the sentence structure of the vectorized sentence based on syntactic analysis technology; S1080, determining the context of multiple words or sentences according to the sentence structure; S1090, based on the context, initial emotion value and sentence structure of the words or sentences, using a pre-trained sentiment analysis model to perform an overall emotion assessment on the context of multiple words or sentences, and generate corresponding emotion categories and intensities to determine the emotion recognition result of the user; S1100: Obtain the user's personal information and historical service request records, and modify the emotion recognition result based on the personal information and historical service request records to determine the corresponding preset emotion information.

[0041] Among them, the preset emotional vocabulary includes common words and the emotional labels corresponding to each common word, and the sentence structure includes subject-predicate-object relationship, modifiers and conjunctions.

[0042] In this embodiment, the text information can be any form of text description used by the user when submitting a service request, including but not limited to a description of the problem, a description of the requirements, or a question about the use of the product. The text information is first captured by the service device and used as input for the natural language processing technology. Through the natural language parsing technology, the service device can understand and analyze the meaning of the text information and identify the key information points therein, such as the user's specific demands, problems encountered, or questions about the product.

[0043] Next, in order to more accurately grasp the user's intention, the service device will combine the parsed text information to generate a complete service request paragraph. This step helps capture the context and semantic logic in the paragraph, so as to more accurately understand the user's true intention.

[0044] After determining the context and semantics of the service request, the service device will use a preset algorithm or model, such as an intent recognition model, to identify the user's main appeal and problem details. These appeals and problem details are important bases for subsequent service resource matching and priority sorting.

[0045] Subsequently, based on the identified main appeal points and problem details, the service device will further match the corresponding intent information. Intent information is a highly generalized and abstract form of user service requests, which can help the service device quickly understand the user's service needs and match them with the most appropriate service resources.

[0046] After determining the intent information, the service device will use keyword recognition technology to extract keywords from the intent information. These keywords represent the core information points in the user service request, and they will serve as an important basis for subsequent sentiment analysis and resource matching.

[0047] In order to identify the user's emotions, the service device will assign an initial emotion value to each keyword or sentence based on a preset emotion vocabulary. This emotion vocabulary contains common words and their corresponding emotion labels, such as "good" for positive emotions and "bad" for negative emotions. By matching keywords with the emotion vocabulary, the service device can preliminarily determine the general direction of the user's emotions.

[0048] However, relying solely on the sentiment labels of keywords is not enough to accurately judge user emotions. Therefore, the service device also uses a pre-trained word embedding model to convert each keyword or sentence into a vector representation to generate a vectorized sentence. This step helps capture the semantic relationship and contextual information between words, and assign emotions to some neutral words, or avoid different emotions in different semantics for the same word, so as to provide richer data support for subsequent sentiment analysis.

[0049] After obtaining the vectorized sentence, the service device will use syntactic analysis technology to identify the sentence structure. The sentence structure includes grammatical elements such as subject-verb-object relationship, modifiers and conjunctions. These grammatical elements are important for understanding Finally, the service device will use the pre-trained sentiment analysis model to evaluate the overall sentiment of the sentence based on the context, initial sentiment value, and sentence structure of the vocabulary or sentence. This evaluation process will take into account multiple factors, such as the sentiment label of the vocabulary, the semantic relationship between the vocabulary, and the sentence structure of the sentence, to generate the corresponding sentiment category and intensity. Through this process, the service device can accurately identify the user's emotional state and provide strong support for subsequent service resource matching and priority sorting. The overall meaning and emotional tendency of the sentence play an important role.

[0050] Optionally, refer to Figure 3 Another embodiment of the present invention provides a customer service method based on the above Figure 2 In the embodiment shown, matching the intent information according to the main appeal point and the problem details includes steps S1031-S1033, wherein: S1031. Based on the main appeal points and question details, the text is parsed and segmented using natural language processing technology to identify key concepts and themes in the user's expression; S1032: Based on a preset intent recognition model, key concepts and topics are classified, and intent matching and similarity calculation are performed to determine the intent category that best matches the user's service request; S1033. Determine the intent information based on the intent category, main appeal points and question details.

[0051] In this embodiment, the main appeal point can be the core demand or problem clearly expressed by the user in the service request, which is directly related to the service or solution the user expects to obtain. The problem details are an elaboration or supplement to the main appeal point, which may include a specific scenario description, difficulties encountered, and attempted solution steps. This information is crucial for the service device because it forms the basis for understanding the user's true intentions and service needs.

[0052] In order to more accurately match the intent information, the service device will first use natural language processing technology to parse and segment the user's text, where segmentation is the words that split the text. This step can identify key concepts and topics in the text, which usually represent the core content of the user's expression. For example, in a service request about product use, key concepts may include product names, function names, or error codes; topics may be questions, suggestions, or feedback about product use.

[0053] Subsequently, the service device will classify the parsed key concepts and topics based on the preset intent recognition model, and perform intent matching and similarity calculation. This model is usually learned through a large amount of training data and can identify different types of service requests and their corresponding intent categories. By calculating the similarity between the user service request and the preset intent category in the model, the service device can determine the intent category that best matches the user service request.

[0054] Finally, based on the matched intent categories, main appeal points and problem details, the service device will comprehensively determine the intent information. This intent information not only summarizes the user's service needs, but also contains enough details for subsequent service resource matching and priority sorting. For example, in the customer service of an e-commerce platform, the intent information may include "return and exchange application", "product consultation" or "account problem", etc. Each intent category will also include a specific problem description and the user's expected solution.

[0055] Through such a process, the service device can more accurately understand the user's service request and match it with the most appropriate service resources and solutions. This can not only improve user satisfaction and loyalty, but also improve service efficiency and reduce operating costs.

[0056] Optionally, refer to Figure 4 Another embodiment of the present invention provides a customer service method based on the above Figure 3 In the embodiment shown, the method for presetting emotional information includes steps S1101-S1109, wherein: S1101, obtaining multiple historical service request records in the historical customer service process, and constructing a user portrait model corresponding to the historical service request records; S1102, analyzing user feedback in historical service request records; S1103, training a user portrait model based on a machine learning algorithm to determine whether user feedback in the user portrait model is positive or negative, and generating a comprehensive emotion score corresponding to each historical service request record; S1104, analyzing service attribute data in historical service request records; S1105, associating the comprehensive emotion score with the service attribute data to identify key factors affecting user emotions; S1106, importing key factors into the user portrait model for training to determine the possible emotional response of the corresponding user in any historical service request record; S1107. Testing the trained user portrait model based on the test set to verify the accuracy and stability of the trained user portrait model; S1108. When the accuracy and stability of the trained user portrait model reach a preset threshold, generate a sentiment prediction model; S1109: Match multiple emotion prediction models in the emotion prediction model with corresponding intention information in the historical service request record one by one to generate preset emotion information.

[0057] Among them, user feedback includes users' evaluation of service quality or satisfaction with service content, and service attribute data includes service type, processing time or the customer service staff who handles the service.

[0058] In this embodiment, the historical service request records may include all service requests submitted by users in the past period of time and related information, such as request time, request content, processing results, user feedback, etc. These records are important data sources for building user portrait models and training emotion prediction models.

[0059] First, by collecting and analyzing historical service request records, the service device is able to build a unique profile of each user. This profile not only contains basic information about the user, such as name, contact information, etc., but more importantly, it reflects the user's service preferences, behavioral habits, and emotional characteristics. For example, some users may prefer to seek help through online chat rather than by phone, or they may show higher dissatisfaction with certain types of problems.

[0060] Next, the service device will conduct an in-depth analysis of user feedback to understand the user's true evaluation of the service quality and satisfaction with the service content. User feedback is an important indicator for measuring service effectiveness. It can not only help the service device identify deficiencies in the service process, but also provide valuable reference for subsequent improvements. By training user feedback through machine learning algorithms, the service device can automatically classify user feedback as positive or negative, and generate a comprehensive emotional score corresponding to each historical service request record. This score can intuitively reflect the user's emotional state during the service process and provide a basis for subsequent emotional prediction.

[0061] At the same time, the service device will also pay attention to the service attribute data in the historical service request records. These service attribute data may include service type, processing time, customer service personnel, etc. These service attribute data are important for understanding the changing trends and influencing factors of user emotions. By associating the comprehensive emotion score and service attribute data, the service device can identify the key factors that affect user emotions, such as certain types of service requests are more likely to cause user dissatisfaction, or the handling methods of certain customer service personnel are more likely to gain user approval.

[0062] After determining the key factors that affect user emotions, the service device will import these factors into the user portrait model for training. Through continuous iteration and optimization, the service device can gradually improve the user portrait model so that it can more accurately predict the user's possible emotional response in any historical service request record. This prediction process not only takes into account the user's basic information and historical behavior, but also incorporates the specific content and context information of the current service request, thereby achieving a comprehensive and detailed analysis of user emotions.

[0063] Finally, when the accuracy and stability of the trained user portrait model reaches the preset threshold, the service device will generate an emotion prediction model. This emotion prediction model can automatically predict the emotion of newly received service requests and generate preset emotion information. This emotion information will serve as an important basis for subsequent service resource matching and priority sorting, helping the service device to better meet user needs and expectations.

[0064] Through this method, the service device can accurately predict and effectively manage user emotions, which can improve the user's service experience and satisfaction.

[0065] Optionally, refer to Figure 5 Another embodiment of the present invention provides a customer service method based on the above Figure 1 In the illustrated embodiment, determining the service type required by the user according to the intent information and matching the corresponding service resources based on the service type includes steps S210-S240, wherein: S210, according to the intent information, matching the service type corresponding to the intent information through a preset service type mapping table, where the preset service type mapping includes a service type corresponding to each intent information; S220: After determining the service type, obtain the user's service request history and service preferences to determine the user's corresponding personalized service requirements; S230, querying a preset service resource database according to the personalized service requirements to determine a service resource list that matches the service type; S240, analyzing the availability, efficiency and user evaluation of each service resource in the service resource list, comprehensively evaluating and selecting the most suitable service resource.

[0066] Among them, the preset service resource database includes a variety of service resources, including manual customer service teams, intelligent customer service systems or self-service tools; each service resource is marked with the corresponding service type, service scope, service efficiency and user evaluation.

[0067] In this embodiment, the preset service type mapping table can be a carefully designed table, which records various intention information that the user may express and the standard service types corresponding to these intention information. The function of this preset service type mapping table is to quickly convert the natural language or keywords input by the user into a service type code that can be recognized by the system, thereby providing a basis for subsequent service resource matching. For example, when the user enters "the mobile phone cannot be turned on", the preset service type mapping table can quickly identify that this is an intention about "equipment maintenance" and map it to the corresponding service type.

[0068] After determining the service type, the service device will further consider the user's personalized service requirements. These personalized requirements may come from the user's historical service request records, service preference settings, or even specific needs expressed by the user during the interaction with customer service. Through comprehensive analysis of this information, the service device can more accurately grasp the user's needs, thereby providing a more intimate and personalized service experience.

[0069] Next, the service device will query the preset service resource database to find a list of service resources that match the service type. This database is a huge resource library that contains various types of service resources, such as manual customer service teams, intelligent customer service systems, self-service tools, etc. Each service resource is marked in detail with key information such as its corresponding service type, service scope, service efficiency, and user evaluation. This information helps the service device make more scientific and reasonable decisions when screening service resources.

[0070] In the process of screening service resources, the service device will comprehensively consider factors such as the availability, efficiency and user evaluation of various service resources. Availability refers to whether the service resources can be used by users at the current time point. For example, some manual customer service may be processing other users' requests and cannot respond immediately. Efficiency refers to the speed and quality of service resources in solving problems. For example, some intelligent customer service systems may have the ability to respond quickly and solve problems efficiently. User evaluation is the direct feedback from users on service resources, which can intuitively reflect the pros and cons of service resources. Through a comprehensive evaluation of these factors, the service device can screen out the most suitable service resources and recommend them to users, which can achieve accurate identification of user needs and effective matching of service resources. It can not only improve users' service experience and satisfaction, but also improve service efficiency and quality.

[0071] Optionally, refer to Figure 6 Another embodiment of the present invention provides a customer service method based on the above Figure 5In the illustrated embodiment, according to the personalized service requirement, querying the preset service resource database to determine a service resource list matching the service type includes steps S231-S233, wherein: S231, filtering out service resources in a preset service resource database that match the current service request; S232. Prioritize the screened service resources according to the personalized service requirements to obtain a ranking result; S233. According to the sorting result, determine the first N service resources as a candidate service resource list, where N is a preset threshold.

[0072] Optionally, analyzing the availability, efficiency and user evaluation of each service resource in the service resource list, comprehensively evaluating and selecting the most suitable service resource includes steps S241-S242, wherein: S241, evaluating whether each service resource in the candidate service resource list can satisfy the user's current service request, and generating a corresponding evaluation result; S242: Based on the evaluation result, determine at least one service resource as a service resource recommended to the user.

[0073] Among them, priority sorting is based on a comprehensive score ranking of service resource response time, historical processing success rate and user satisfaction evaluation; the evaluation requirements include the availability of service resources and whether the service scope covers the content of user requests.

[0074] In this embodiment, the personalized service requirement may be a specific requirement put forward by the user based on historical service experience, specific needs or preferences. When processing a service request, the service device will first filter out all available service resources that match the current service request from the preset service resource database. These service resources may include different manual customer service teams, intelligent customer service systems or self-service tools, each of which has different characteristics and service capabilities. Among them, the preset service resource database can be a dynamically updated storage system that continuously collects, organizes and classifies service resource information from different channels. These channels include but are not limited to multiple departments within the company, third-party service providers, and user feedback. The database is designed to quickly respond to and meet diverse service needs, ensuring that when a service request is received, the service resources that best meet the user's personalized needs can be quickly located.

[0075] The service device will then prioritize the resources based on the user's personalized service requirements. This step is crucial because it directly determines which service resources will be recommended to the user first. In order to make a scientific and reasonable ranking, the service device will take into account multiple factors, such as the response time of the service resource (i.e., the time difference between the user's request and the response), the historical processing success rate (i.e., the proportion of similar problems successfully solved by the resource in the past), and the user satisfaction evaluation (i.e., the user's direct feedback on the service quality provided by the resource). Through comprehensive scoring, the service device can come up with a clear ranking result, thereby ensuring that the service resources that best meet the user's needs are placed in front.

[0076] After obtaining the ranking results, the service device will filter out the first N service resources as a candidate service resource list. Here, N is a preset threshold, which is adjusted according to the actual situation of the system and user needs. By limiting the number of candidate resources, the service device can process service requests more efficiently and avoid users being troubled by too many choices.

[0077] Finally, the service device will conduct a comprehensive evaluation of each resource in the candidate service resource list. The evaluation includes the availability of the service resource (i.e., whether the resource can be used by the user at the current time point), efficiency (i.e., the speed and quality of the resource in solving problems), and user evaluation (i.e., the user's overall evaluation of the resource). Through these evaluation results, the service device can determine at least one most suitable service resource and recommend it as the final service resource to the user. This process not only realizes the accurate identification of user needs and the effective matching of service resources, but also improves the user's service experience and satisfaction.

[0078] Optionally, refer to Figure 7 Another embodiment of the present invention provides a customer service method based on the above Figure 6 In the embodiment shown, evaluating whether each service resource in the candidate service resource list can satisfy the user's current service request and generating a corresponding evaluation result includes steps S2411-S2415, wherein: S2411. According to the service resources in the candidate service resource list, the current performance indicators of each service resource are obtained one by one to check the current working status, load condition and ability of each service resource to process user requests; S2412, refine and quantify the user's service request to determine the specific content and specific requirements of the service request, to determine the complexity and urgency of the service request and the user's expectations of service quality, and to determine the corresponding evaluation criteria; S2413, comparing the performance indicators of the service resources with the service request requirements of the user to evaluate whether the service resources can meet the user's needs; S2414. If the performance indicator of the service resource meets or exceeds the user's requirement, it is considered that the service resource can meet the user's current service request and a positive evaluation result is generated; S2415. If the performance indicators of the service resources fail to meet the user's requirements, a negative evaluation result is generated and the deficiencies are marked.

[0079] Among them, performance indicators include response time, concurrent processing capability and historical processing success rate.

[0080] In this embodiment, the candidate service resource list can be based on a detailed evaluation of the user's specific service request, and the service device can more finely judge the applicability of each service resource. Among them, the service device will first obtain the real-time performance indicators of each service resource in the candidate service resource list, which include but are not limited to response time, concurrent processing capability, and historical processing success rate. These performance indicators can intuitively reflect the current working status, load status, and ability to process user requests of the service resource.

[0081] Subsequently, the service device will conduct an in-depth analysis and quantitative processing of the user's service request. This step aims to transform the user's service request into a measurable standard, which includes the specific content, complexity, urgency and user's expectations of service quality. By refining the service request, the service device can more accurately grasp the user's needs and thus formulate a more scientific and reasonable evaluation standard.

[0082] After obtaining the performance indicators of the service resources and the detailed requirements of the user's service request, the service device will conduct a comparative analysis between the two. During this process, the service device will compare the performance indicators of the service resources with the specific requirements of the user's service request one by one to evaluate whether the service resources can meet the actual needs of the user. If the performance indicators of the service resources meet or exceed the user's requirements, the service device will consider that the service resources are able to meet the user's current service request and generate a positive evaluation result. Conversely, if the performance indicators of the service resources fail to meet the user's requirements, the service device will generate a negative evaluation result and clearly point out the deficiencies for subsequent improvements or adjustments.

[0083] Through the above evaluation steps, the service device can ensure that the service resources recommended to the user not only have sufficient capabilities to process the user's request, but also can meet the user's personalized needs to the greatest extent. This evaluation method based on the user's specific service request not only improves the accuracy of service resource matching, but also enhances the user's service experience and satisfaction. At the same time, this evaluation method also provides strong data support for the optimization and upgrading of service resources, which helps to continuously improve service quality and efficiency.

[0084] Optionally, refer to Figure 8 Another embodiment of the present invention provides a customer service method based on the above Figure 1 In the illustrated embodiment, after the service request is assigned, continuous monitoring and data analysis to determine the service quality of the customer service team or customer service personnel in handling the service request includes steps S410-S450, wherein: S410, obtaining user feedback data during customer service; S420, classify and sort the feedback data, and extract key information points of the feedback data according to different classifications; S430, based on key information points and through machine learning algorithms or statistical analysis methods, identify potential optimization points in the service process to determine the corresponding behavior patterns of users or service efficiency bottlenecks in the customer service process, and generate corresponding analysis results; S440, generating corresponding improvement and adjustment measures according to the analysis results, the improvement and adjustment measures including optimizing service processes, improving service efficiency or improving user interface design; S450, and after implementing improvement and adjustment measures, continue to monitor user feedback and service effects to iteratively optimize the improvement and adjustment measures.

[0085] Among them, feedback data includes user satisfaction scores, interaction records during the service process, and user suggestions for service improvements.

[0086] In this embodiment, the key information points may include the user's evaluation of the service attitude, the speed and effect of problem solving, the communication fluency during the service process, etc. Through in-depth analysis of these key information points, it is possible to more comprehensively understand the user's true feelings and needs for customer service. Using machine learning algorithms or statistical analysis methods, the service device can mine potential optimization points from a large amount of feedback data. These optimization points may involve multiple aspects such as the design of service processes, the training of customer service personnel, and the improvement of technical tools.

[0087] Among them, when the service request is assigned to the customer service team or customer service personnel, the service device will begin to continuously monitor the entire service process. During this process, the service device will collect user feedback data, including but not limited to user satisfaction scores, interaction records during the service process, and users' specific suggestions for service improvements. These data are important bases for evaluating service quality and discovering potential problems.

[0088] Next, the service device will classify and sort the collected feedback data in order to more clearly identify different types of problems and user concerns. In the process of classification and sorting, the service device will extract key information points in the feedback data, which are directly related to various aspects of service quality, such as service attitude, problem solving efficiency, communication effectiveness, etc.

[0089] With these key information points, the service device can use machine learning algorithms or statistical analysis methods to conduct in-depth analysis. Through these analysis methods, the service device can identify potential optimization points in the service process, which may be the root cause of user dissatisfaction or low service efficiency. At the same time, the service device can also predict the user's future service needs based on the user's behavior patterns and feedback data, thereby providing more accurate service suggestions to the customer service team.

[0090] After identifying potential optimization points and predicting user needs, the service device will generate corresponding improvement and adjustment measures. These measures may include optimizing service processes, improving the professional capabilities of customer service personnel, improving technical tools to improve service efficiency, etc. By implementing these improvement and adjustment measures, we expect to significantly improve the quality and efficiency of customer service and enhance user satisfaction and loyalty.

[0091] Finally, after implementing improvement and adjustment measures, the system will continue to monitor user feedback and service effects. By comparing the data changes before and after the improvement, the service device can evaluate the effectiveness of the improvement measures and further optimize and adjust the deficiencies. This iterative optimization process will ensure that our customer service can continue to adapt to market changes and user needs.

[0092] In actual applications, for example, when customer service staff X is providing service to user A, the interaction records of customer service staff X during the service process and the satisfaction scores given by users during the communication process will be monitored in real time. The service device will classify and sort the interaction records and satisfaction scores in the feedback data, and extract key information points, such as the service attitude, response speed, and problem-solving ability of customer service staff X. Then, using machine learning algorithms or statistical analysis methods, the service device will deeply analyze these key information points and identify service efficiency bottlenecks or user behavior patterns that may exist in the service process of user A. For example, if it is found that customer service staff X takes a long time to handle a certain type of problem and the user satisfaction is low, the service device may suggest optimizing the processing process of such problems, or provide targeted training to improve the professional ability of customer service staff X.

[0093] In addition, the service device will also pay attention to users' suggestions for service improvements, which often directly reflect users' real needs and expectations. By integrating user suggestions with data analysis results, the service device can generate improvement and adjustment measures that are closer to user needs. For example, if many users mention that they hope to add a certain self-service function to reduce waiting time, the service device may promote the R&D team to develop corresponding self-service tools and optimize the user interface design to improve user experience.

[0094] After implementing improvement and adjustment measures, the service device will continue to monitor changes in user feedback and service effects. By collecting and analyzing new feedback data, the service device can evaluate the actual effect of the improvement measures and make further optimizations and adjustments accordingly. This data-driven iterative optimization process ensures that customer service is always maintained at a high level to meet the ever-changing needs and expectations of users.

[0095] The present invention also proposes a service device, which includes: the service device includes: a memory, a processor, and a customer service program stored in the memory and executable on the processor, the customer service program is configured to implement the above customer service method.

[0096] It is worth noting that since the service device of the present invention is based on the above-mentioned customer service method, the embodiments of the service device of the present invention include all technical solutions of all embodiments of the above-mentioned customer service method, and the technical effects achieved are also exactly the same, which will not be repeated here.

[0097] The present invention also provides a computer device, which includes the service device as described in the above embodiment.

[0098] It is worth noting that since the computer device of the present invention is based on the above-mentioned service device, the embodiments of the computer device of the present invention include all technical solutions of all embodiments of the above-mentioned service device, and the technical effects achieved are also exactly the same, which will not be repeated here.

[0099] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0100] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0101] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of various embodiments of the present invention.

[0102] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A customer service method, characterized in that: The method comprises: Obtain corresponding service requests submitted by multiple users through multiple different communication channels, and parse the corresponding intent information in the service requests based on natural language processing technology; and parse the preset emotional information of each word or sentence in the intent information; the intent information includes user's suggestions for product consultation, fault repair or complaint; Determine the service type required by the user according to the intention information, and match corresponding service resources based on the service type; Determine the priority ranking of the service request according to the emotion information and the service resource, and assign a customer service team or customer service personnel corresponding to the service resource according to the priority ranking; After the service request is assigned, continuous monitoring and data analysis are performed to determine the service quality of the customer service team or customer service personnel in handling the service request, and based on the service quality, feedback evaluation and guidance results are given to the corresponding customer service team or corresponding customer service personnel.

2. The customer service method according to claim 1, characterized in that: The natural language processing technology is used to analyze the corresponding intent information in the service request; And analyzing the preset emotional information of each word or sentence in the intention information includes: Parsing multiple text messages in the service request based on natural language, and combining the multiple text messages according to the text messages to generate corresponding service request paragraphs; According to the context and semantics of the service request segment, identify the user's main appeal points and problem details; Matching the intention information according to the main appeal points and the question details; Extracting keywords from the intention information based on keyword recognition technology to generate multiple words or sentences; Assigning an initial emotion value to each word or sentence based on a preset emotion vocabulary, wherein the preset emotion vocabulary includes common words and emotion labels corresponding to each common word; Use a pre-trained word embedding model to convert each word or sentence into a vector representation to generate vectorized sentences; Identifying the sentence structure of the vectorized sentence based on syntactic analysis technology, wherein the sentence structure includes a subject-predicate-object relationship, modifiers, and conjunctions; Determining contexts of a plurality of the words or the sentences according to the sentence structure; According to the context of the words or the sentences, the initial emotion value and the sentence structure, a pre-trained sentiment analysis model is used to perform an overall emotion assessment on the context of the multiple words or the sentences, and a corresponding emotion category and intensity are generated to determine an emotion recognition result of the user; The user's personal information and historical service request records are obtained, and the emotion recognition result is corrected based on the personal information and the historical service request records to determine the corresponding preset emotion information.

3. The customer service method according to claim 2, characterized in that: The matching of the intention information according to the main appeal point and the question details includes: According to the main appeal points and the problem details, the text is parsed and segmented through natural language processing technology to identify the key concepts and themes in the user's expression; Based on a preset intent recognition model, the key concepts and the topics are classified, and intent matching and similarity calculation are performed to determine the intent category that best matches the user's service request; The intent information is determined according to the intent category, the main appeal point and the question details.

4. The customer service method according to claim 3, characterized in that: The method for preset emotional information includes: Acquire multiple historical service request records in the historical customer service process, and construct a user portrait model corresponding to the historical service request records; Analyzing user feedback in the historical service request records, wherein the user feedback includes user evaluation of service quality or satisfaction with service content; Training the user portrait model based on a machine learning algorithm to determine whether user feedback in the user portrait model is positive or negative, and generating a comprehensive sentiment score corresponding to each of the historical service request records; Analyze the service attribute data in the historical service request record, wherein the service attribute data includes the service type, processing time or the customer service personnel who processed the request; Associating the comprehensive emotion score with the service attribute data to identify key factors affecting user emotions; Importing the key factors into the user portrait model for training to determine the possible emotional response of the user corresponding to any of the historical service request records; Testing the trained user portrait model based on the test set to verify the accuracy and stability of the trained user portrait model; When the accuracy and stability of the trained user portrait model reach a preset threshold, generating a sentiment prediction model; Multiple emotion prediction models in the emotion prediction model are matched one by one with corresponding intention information in the historical service request record to generate the preset emotion information.

5. The customer service method according to claim 1, characterized in that: The determining, according to the intent information, the service type required by the user, and matching corresponding service resources based on the service type includes: According to the intent information, a service type corresponding to the intent information is matched through a preset service type mapping table, wherein the preset service type mapping includes a service type corresponding to each of the intent information; After determining the service type, obtaining the user's service request history and service preferences to determine the user's corresponding personalized service requirements; According to the personalized service requirements, query the preset service resource database to determine a list of service resources matching the service type; the preset service resource database includes multiple service resources, and the multiple service resources include a manual customer service team, an intelligent customer service system or a self-service tool; each of the service resources is marked with a corresponding service type, service scope, service efficiency and user evaluation; Analyze the availability, efficiency and user evaluation of each service resource in the service resource list, comprehensively evaluate and select the most suitable service resource.

6. The customer service method according to claim 5, characterized in that: The querying of a preset service resource database according to the personalized service requirement to determine a service resource list matching the service type includes: Filter out service resources that match the current service request from a preset service resource database; According to the personalized service requirements, the screened service resources are prioritized to obtain a ranking result; wherein the priority ranking is based on a comprehensive ranking of the service resources' response time, historical processing success rate, and user satisfaction evaluation; According to the ranking result, determine the first N service resources as a candidate service resource list, where N is a preset threshold; The analyzing the availability, efficiency and user evaluation of each service resource in the service resource list, and comprehensively evaluating and selecting the most suitable service resource include: Evaluate whether each of the service resources in the candidate service resource list can meet the user's current service request, and generate a corresponding evaluation result; the evaluation requirements include the availability of the service resource and whether the service scope covers the content requested by the user; Based on the evaluation result, at least one service resource is determined as a service resource recommended to the user.

7. The customer service method according to claim 6, characterized in that: The evaluating whether each of the service resources in the candidate service resource list can satisfy the user's current service request and generating a corresponding evaluation result includes: According to the service resources in the candidate service resource list, the current performance indicators of each service resource are obtained one by one to check the current working status, load condition and ability to process user requests of each service resource, wherein the performance indicators include response time, concurrent processing capability and historical processing success rate; Refine and quantify the user's service request to determine the specific content and specific requirements of the service request, to determine the complexity and urgency of the service request and the user's expectations of service quality, and to determine the corresponding evaluation criteria; Compare the performance indicators of service resources with the service request requirements of users to evaluate whether the service resources can meet the needs of users; If the performance indicators of the service resources meet or exceed the user's requirements, it is considered that the service resources can meet the user's current service request and generate a positive evaluation result; If the performance indicators of the service resources fail to meet the user's requirements, a negative evaluation result is generated and the deficiencies are marked.

8. The customer service method according to claim 1, characterized in that: After the service request is assigned, the continuous monitoring and data analysis to determine the service quality of the customer service team or customer service personnel in handling the service request includes: Obtaining user feedback data during customer service, including user satisfaction scores, interaction records during the service process, and user suggestions for service improvements; Classify and sort the feedback data, and extract key information points of the feedback data according to different classifications; Based on the key information points, and through machine learning algorithms or statistical analysis methods, potential optimization points in the service process are identified to determine the corresponding behavior patterns of users or service efficiency bottlenecks in the customer service process, and generate corresponding analysis results; Generate corresponding improvement and adjustment measures based on the analysis results, wherein the improvement and adjustment measures include optimizing service processes, improving service efficiency, or improving user interface design; After implementing improvement and adjustment measures, continuously monitor user feedback and service results to iteratively optimize the improvement and adjustment measures.

9. A service device, characterized in that: The service device includes: a memory, a processor, and a customer service program stored in the memory and executable on the processor, wherein the customer service program is configured to implement the customer service method according to any one of claims 1 to 8.

10. A computer device, characterized in that: Comprising the service device as claimed in claim 9.

Images