Virtual service personnel interaction method, electronic device, storage medium and program product

Abstract

This application provides a virtual service personnel interaction method, electronic device, computer-readable storage medium, and computer program product. The method includes: receiving an access request from a first terminal device; establishing a communication connection between the first terminal device and a target server; using the first terminal device to receive user input information; determining one service item from a plurality of service items as the user's target service item; obtaining a first interactive video corresponding to the target service item from a question-and-answer database corresponding to the virtual service personnel; and using the first terminal device to play the first interactive video, in which the virtual service personnel introduce the procedures and / or guidelines corresponding to the target service item. This application provides interactive videos related to target service items so that users can more clearly understand the required service content and operation steps, improving user satisfaction and efficiency.

Description

Technical Field

[0001] This application relates to the technical fields of virtual humans and artificial intelligence, and in particular to methods for virtual service personnel interaction, electronic devices, computer-readable storage media, and computer program products. Background Technology

[0002] With the rapid development of internet technology and the fast pace of life, more and more services are shifting online. To improve user convenience and efficiency, some businesses and government departments have begun using virtual service personnel to provide online consultations and service guidance. Virtual service personnel can provide real-time interaction based on user needs, effectively answering user questions and improving user satisfaction.

[0003] Existing virtual service provider interaction methods still have shortcomings in processing user input information and providing personalized services. Therefore, this application provides a virtual service provider interaction method, electronic device, computer-readable storage medium, and computer program product to improve upon the prior art. Summary of the Invention

[0004] The purpose of this application is to provide a virtual service personnel interaction method, electronic device, computer-readable storage medium, and computer program product that can quickly determine the target service item based on input information and provide interactive videos related to the target service item so that users can more clearly understand the required service content and operation steps, thereby improving user satisfaction and efficiency.

[0005] The objective of this application is achieved through the following technical solution:

[0006] Firstly, this application provides a method for interacting with virtual service personnel, the method comprising:

[0007] Receive an access request from a first terminal device, establish a communication connection between the first terminal device and a target server, wherein the target server is used to provide virtual service personnel interaction functions;

[0008] The first terminal device is used to receive user input information, each of which is text information, voice information, or image information;

[0009] Based on the input information, one service item is determined from multiple service items as the user's target service item;

[0010] Obtain the first interactive video corresponding to the target service item from the question and answer database corresponding to the virtual service personnel, wherein the question and answer database includes first interactive videos corresponding to multiple service items;

[0011] The first interactive video is played using the first terminal device, in which the virtual service personnel introduce the procedures and / or guidelines corresponding to the target service item.

[0012] The beneficial effects of this technical solution are as follows: By establishing a communication connection between the first terminal device and the target server, the interactive function of virtual service personnel is realized, improving the real-time performance and stability of the system and providing users with a smoother online consultation experience. By receiving user input information, including text, voice, and image information, it can meet the needs of different users, provide a more convenient interaction method, and enhance the user experience. By determining the target service item from multiple service items based on user input information, user needs can be quickly identified, improving the targeting and accuracy of virtual service personnel. Obtaining the first interactive video corresponding to the target service item from the question-and-answer database corresponding to the virtual service personnel provides detailed interactive videos related to the target service item, allowing users to more intuitively understand the relevant information of the target service item, improving user satisfaction and efficiency. In the first interactive video, the virtual service personnel introduce the corresponding service procedures and / or service guides for the target service item, providing users with detailed guidance information, enabling users to more clearly understand the procedures and requirements, improving efficiency and accuracy.

[0013] In summary, the virtual service personnel interaction method provided in this application can receive various types of input information, quickly identify the target service item, and provide detailed procedures and guides, effectively improving user convenience, efficiency, and satisfaction. Simultaneously, this method ensures real-time performance and stability during the interaction between the user and the virtual service personnel, guaranteeing that users can find the required procedures and guides within a short time. Compared to existing technologies, it better adapts to user needs, increases user reliance on and satisfaction with online services, and enhances the user experience. Therefore, this technical solution has high application value and practicality in the field of virtual service personnel interaction, providing enterprises and government departments with an effective way to improve customer service and enhance user experience.

[0014] In some possible implementations, determining one service item from a plurality of service items as the user's target service item based on the input information includes:

[0015] Using the semantic extraction model corresponding to the input information, semantic information is extracted from the input information;

[0016] Obtain the similarity between the semantic information and each of the service items, and take the maximum similarity as the highest similarity.

[0017] When the highest similarity is greater than the preset similarity, the service item corresponding to the highest similarity is taken as the target service item.

[0018] The beneficial effects of this technical solution are as follows: By using a semantic extraction model to extract semantic information from input information, it is possible to more accurately understand user needs and effectively improve the response accuracy of virtual service personnel. By calculating the similarity between semantic information and various service items, the service item that best matches the user's input information can be quickly found, further improving the targeting and accuracy of virtual service personnel. By setting preset similarity, misjudgments and incorrect matches can be effectively avoided, ensuring that only highly matching service items are provided to the user as the target service item. This technical solution enhances the accuracy and reliability of virtual service personnel in handling complex problems and fuzzy inputs, thereby improving user satisfaction and efficiency. In summary, by adopting a semantic extraction model, calculating similarity, and setting preset similarity, this technical solution can more accurately determine the user's target service item and provide users with more precise and personalized services. This implementation method has high practicality and efficiency in the field of virtual service personnel interaction, which helps to further improve user experience and satisfaction, enabling enterprises and government departments to better meet the needs of various groups.

[0019] In some possible implementations, determining one service item from a plurality of service items as the user's target service item based on the input information further includes:

[0020] When the highest similarity is not greater than the preset similarity, the first terminal device is used to play the second interactive video to guide the user to click the form button and fill in the form;

[0021] When a click operation is received on the form button, the form is displayed using the first terminal device. The form is provided with a service item input box and / or a service item drop-down menu.

[0022] Based on the form's completion results, the user's target service item is obtained.

[0023] The beneficial effects of this technical solution are as follows: When the highest similarity score is no greater than a preset similarity score, playing a second interactive video guides the user to click the form button and fill out the form. This helps to collect more information about user needs, more effectively obtain detailed user requirements, and improve the accuracy of target service items, thereby providing users with more precise and personalized services. The form includes service item input boxes and / or service item drop-down menus, giving users greater flexibility when providing their needs and facilitating their selection and input of required services. It provides an effective supplementary means when virtual service personnel cannot accurately identify user needs, helping to improve user satisfaction and efficiency. In summary, when the highest similarity score does not meet the preset condition, this technical solution guides users to fill out a form to obtain more information about their needs, thereby better meeting users' actual needs and improving overall efficiency and user satisfaction.

[0024] In some possible implementations, the process of obtaining the similarity between the semantic information and each of the service items includes:

[0025] For each of the aforementioned service items, perform the following processing:

[0026] Input the semantic information and the preset text corresponding to the service item into a similarity model to obtain the similarity between the semantic information and the preset text, which is used as the similarity between the semantic information and the service item;

[0027] The training process of the similarity model includes:

[0028] Obtain a training set, which includes multiple training data, each of which includes a sample semantic information, a sample text, and annotation data on the similarity between the sample semantic information and the sample text;

[0029] For each of the training data, perform the following processing:

[0030] The sample semantic information and sample text in the training data are input into a preset deep learning model to obtain the predicted similarity data between the sample semantic information and the sample text.

[0031] The model parameters of the deep learning model are updated based on the sample semantic information and the predicted and labeled data of the similarity between the sample text and the sample semantic information.

[0032] Check whether the preset training termination condition is met; if yes, use the trained deep learning model as the similarity model; if no, continue to train the deep learning model using the next training data.

[0033] The beneficial effects of this technical solution are as follows: For each service item, inputting semantic information and preset text into the similarity model enables more accurate calculation of the similarity between semantic information and each service item, improving the matching accuracy of virtual service personnel. The training process of the similarity model employs a deep learning model, utilizing sample semantic information and sample text from the training data. This allows the similarity model to better capture the correlation between semantic information and text, improving the accuracy of similarity calculation. By detecting preset training termination conditions, it ensures that the similarity model completes training while meeting certain accuracy requirements, thereby improving the model's generalization ability and practicality. This technical solution, by using a deep learning model for similarity calculation, can better adapt to service items in different fields, improving the versatility and application value of virtual service personnel in various scenarios. It exhibits high accuracy and reliability in processing user input information, thereby improving user satisfaction and efficiency, enabling enterprises and government departments to better meet customer needs. In summary, this technical solution, through training a similarity model using a deep learning model, achieves more accurate similarity calculation between semantic information and service items, thus improving the accuracy and practicality of virtual service personnel in identifying target service items.

[0034] In some possible implementations, the method further includes:

[0035] The first terminal device is used to receive feedback information from the user, and each piece of feedback information is text information, voice information, or image information;

[0036] Based on the feedback information, identify the user's emotion category;

[0037] Based on the user's emotion category, determine whether the user is satisfied with the service;

[0038] If not satisfied, increment the cumulative count by one;

[0039] When the cumulative number of times exceeds the preset number of times, the cumulative number of times is cleared to zero, and a prompt message is sent to the second terminal device of the preset personnel corresponding to the target service item, so as to prompt the preset personnel to update the similarity model or update the preset text corresponding to the target service item.

[0040] The beneficial effects of this technical solution are as follows: By receiving user feedback and identifying user emotion categories, it is possible to better understand user satisfaction with the services provided by virtual service personnel, which helps to further optimize service quality. Based on the user's emotion category, detecting whether the user is satisfied with the service allows for the timely identification of problematic service items, enabling targeted improvements. When the cumulative number of occurrences exceeds a preset limit, a prompt message is sent to the second terminal device of the preset personnel, helping to remind relevant personnel to update the similarity model or adjust the preset text, improving the matching accuracy of virtual service personnel and user satisfaction. This method enhances the accuracy and reliability of virtual service personnel in handling user needs, thereby improving user satisfaction and efficiency, enabling enterprises and government departments to better meet customer needs. Through user feedback and emotion recognition, this technical solution helps to collect valuable user opinions and suggestions, providing strong support for further optimizing the work effectiveness of virtual service personnel, ensuring that relevant personnel pay timely attention to user satisfaction issues, and preventing a decline in service quality. In summary, this technical solution, by collecting user feedback, identifying user emotion categories, and sending prompt messages, can continuously optimize the service quality and matching accuracy of virtual service personnel.

[0041] In some possible implementations, the method further includes:

[0042] The first terminal device is used to display a plurality of second interactive videos related to the first interactive video, each of which is used to introduce the precautions, frequently asked questions or related service items of the target service item;

[0043] When a click operation is received for one of the second interactive videos, the clicked second interactive video is played using the first terminal device.

[0044] The beneficial effects of this technical solution are as follows: By displaying multiple second interactive videos related to the first interactive video on the first terminal device, users can be provided with richer and more comprehensive service information, helping them better understand the precautions, frequently asked questions, or related service items for the target service. Users can choose to watch second interactive videos related to the target service according to their needs and interests, increasing user initiative and interactivity in the process of obtaining service information, thereby improving user satisfaction. This method provides users with a more convenient and personalized way to obtain information, helping to meet the needs and preferences of different users.

[0045] In some possible implementations, during the first interactive video, the virtual service representative guides the user to click the document download button to download the certification document;

[0046] The method further includes:

[0047] When a click operation is received on the document download button, the certificate document corresponding to the document download button is downloaded to the first terminal device.

[0048] The beneficial effects of this technical solution are as follows: In the first interactive video, the virtual service personnel guide the user to click the document download button, enabling the user to easily obtain the required supporting documents, thus improving user experience and satisfaction. By responding to the user's click, the supporting documents are downloaded to the first terminal device, realizing the convenience and efficiency of the document download function and saving the user's time and effort. This technical solution simplifies the process of obtaining supporting documents, thereby improving efficiency. Implementing the supporting document download function during the interaction with the virtual service personnel further enriches the service content that the virtual service personnel can provide, allowing users to complete more service items on a single platform.

[0049] In some possible implementations, the step of downloading the proof document corresponding to the document download button to the first terminal device when a click operation is received on the document download button includes:

[0050] When a click operation is received on the document download button, the first terminal device displays the document download interface corresponding to the document download button, and the document download interface is provided with an identity recognition button.

[0051] When a click operation is received on the identity recognition button, the first terminal device is used to obtain the user's identity information to be identified;

[0052] Detect whether the identity information to be identified matches the user's authenticated identity information;

[0053] If a match is found, the proof document is downloaded to the first terminal device.

[0054] The beneficial effects of this technical solution are as follows: Upon receiving a click on the document download button, a document download interface is displayed. This interface includes an identity verification button to trigger user authentication steps, thus providing a secure identity verification mechanism and ensuring the security and privacy protection of downloaded documents. After the user clicks the identity verification button, the first terminal device obtains the user's identity information to be verified, helping to ensure that the obtained document matches the user's identity. By detecting the matching between the identity information to be verified and the authenticated identity information, malicious users can be prevented from downloading others' documents to a certain extent, improving information security. If the identity information to be verified matches the authenticated identity information, the document will be downloaded to the first terminal device, enabling the user to successfully obtain the required document, improving user experience and satisfaction.

[0055] In some possible implementations, the step of downloading the proof document to the first terminal device if a match is found includes:

[0056] If a match is found, the user information database corresponding to the target service item is queried based on the authenticated identity information to generate the proof document based on the query results.

[0057] Download the proof document to the first terminal device.

[0058] The beneficial effects of this technical solution are as follows: It queries the user information database corresponding to the target service item based on the authenticated identity information, generates a supporting document based on the query results, ensuring that the generated supporting document is based on the user's real information, and guaranteeing the timeliness, reliability, and accuracy of the generated supporting document. During the interaction with virtual service personnel, the generation and download functions of the supporting document are implemented, further enriching the service content that virtual service personnel can provide.

[0059] In some possible implementations, the method further includes:

[0060] Based on the user's authenticated identity information and the interaction data between the user and the virtual service personnel, one or more recommendation policies corresponding to the user are obtained and displayed in list form on the first terminal device.

[0061] The beneficial effects of this technical solution are as follows: By combining the user's authenticated identity information and interaction data between the user and virtual service personnel, one or more recommended policies corresponding to the user can be obtained more accurately, thereby improving the targeting and accuracy of the recommendations. Displaying recommended policies in list form on the first terminal device provides users with a clear and intuitive way of presenting information, helping them quickly browse and understand policies relevant to them. This method implements a policy recommendation function, helping users to understand policy information relevant to themselves in a timely manner, increasing users' attention to and awareness of policies, helping them benefit from policies, and preventing them from missing out on policy benefits relevant to them. By recommending policies related to users' identity and needs, it helps to improve the coverage and implementation effect of policies, thereby improving the overall effectiveness of policy implementation.

[0062] Secondly, this application provides an electronic device including a memory and at least one processor, the memory storing a computer program, and the at least one processor being configured to execute the computer program to perform the following steps:

[0063] Receive an access request from a first terminal device, establish a communication connection between the first terminal device and a target server, wherein the target server is used to provide virtual service personnel interaction functions;

[0064] The first terminal device is used to receive user input information, each of which is text information, voice information, or image information;

[0065] Based on the input information, one service item is determined from multiple service items as the user's target service item;

[0066] Obtain the first interactive video corresponding to the target service item from the question and answer database corresponding to the virtual service personnel, wherein the question and answer database includes first interactive videos corresponding to multiple service items;

[0067] The first interactive video is played using the first terminal device, in which the virtual service personnel introduce the procedures and / or guidelines corresponding to the target service item.

[0068] Thirdly, this application provides a computer-readable storage medium storing a computer program that, when executed by at least one processor, implements the steps of any of the above-described methods or the functions of any of the above-described electronic devices.

[0069] Fourthly, this application provides a computer program product comprising a computer program that, when executed by at least one processor, implements the steps of any of the above methods or the functions of any of the above electronic devices. Attached Figure Description

[0070] This application will be further described below with reference to the accompanying drawings and specific embodiments.

[0071] Figure 1 This is a flowchart illustrating a virtual service personnel interaction method provided in an embodiment of this application.

[0072] Figure 2 This is a schematic diagram of a process for obtaining a target service item provided in an embodiment of this application.

[0073] Figure 3 This is a schematic diagram of a process for prompting preset personnel to update, provided in an embodiment of this application.

[0074] Figure 4 This is a schematic diagram of a process for downloading a proof document provided in an embodiment of this application.

[0075] Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of this application.

[0076] Figure 6 This is a schematic diagram of the structure of a computer program product provided in an embodiment of this application. Detailed Implementation

[0077] The technical solutions in this application will be described below with reference to the accompanying drawings and specific embodiments. It should be noted that, without conflict, the various embodiments or technical features described below can be arbitrarily combined to form new embodiments.

[0078] In this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any implementation or design described as "exemplary" or "for example" in this application should not be construed as being better or more advantageous than other implementations or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0079] The descriptions of "first," "second," etc., appearing in the embodiments of this application are for illustrative purposes and to distinguish the objects being described. They have no order and do not indicate any special limitation on the quantity in the embodiments of this application, nor do they constitute any limitation on the embodiments of this application.

[0080] The technical field and related terms of the embodiments of this application are briefly described below.

[0081] Virtual objects include virtual humans, virtual animals, and virtual cartoon characters. Virtual humans, in particular, are anthropomorphic figures constructed using CG technology and operating in code form, possessing various interactive capabilities such as language communication, facial expressions, and action demonstrations. Virtual human technology has rapidly developed in the field of artificial intelligence and has been applied in many technological areas, including film, media, games, finance, culture and tourism, education, and healthcare. It allows for the customization of virtual hosts, virtual anchors, virtual idols, virtual customer service representatives, virtual lawyers, virtual service personnel, virtual doctors, virtual guides, and virtual assistants, and can also generate videos with a single click from text or audio. Among virtual humans, service-oriented virtual humans primarily function to replace real-life services and provide daily companionship; they are virtualizations of real-world service roles. Their industrial value mainly lies in reducing costs in existing service industries, thereby improving efficiency and reducing costs in the existing market.

[0082] Artificial Intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess perception, reasoning, and decision-making capabilities. AI technology is a comprehensive discipline involving a wide range of fields, encompassing both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly include computer vision, speech processing, natural language processing, as well as machine learning / deep learning, autonomous driving, and intelligent transportation.

[0083] Machine Learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. A computer program can learn experience E given a certain type of task T and a performance metric P. If its performance on task T can be measured by P, it improves with experience E. Machine learning specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence.

[0084] Deep learning is a special type of machine learning that learns to represent the world using nested hierarchical concepts, achieving tremendous functionality and flexibility. Each concept is defined as being associated with a simpler one, while more abstract representations are computed in a less abstract manner. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and learning by demonstration.

[0085] Virtual object interaction applications provide interactive functionality using virtual objects. These virtual objects can simulate human communication and behavior and interact with users. Such software (referring to virtual human interaction applications) is typically driven by artificial intelligence and natural language processing technologies and can interact with users through text, voice, images, forms, and other methods.

[0086] With the rapid development of internet technology and the fast pace of life, more and more services are shifting online. To improve user convenience and efficiency, some businesses and government departments have begun using virtual service personnel to provide online consultations and service guidance. Virtual service personnel can provide real-time interaction based on user needs, effectively answering user questions and improving user satisfaction.

[0087] Existing virtual service interaction methods still have shortcomings in processing user input and providing personalized services. For example, some existing methods can only process text information and cannot recognize or process voice or image information. Furthermore, some virtual service providers lack specificity in their service delivery, failing to quickly determine the target service based on user input, causing users to spend more time interacting with virtual service providers and reducing efficiency. Additionally, existing methods fail to provide clear and intuitive procedures and guides when playing interactive videos, further reducing user satisfaction.

[0088] Based on this, this application provides a virtual service personnel interaction method, electronic device, computer-readable storage medium, and computer program product to improve the prior art.

[0089] The solutions provided in this application involve technologies such as virtual humans, interaction design, artificial intelligence, 3D modeling, and cloud computing, and are specifically illustrated through the following embodiments. It should be noted that the order of description of the following embodiments is not intended to limit the preferred order of the embodiments.

[0090] (Virtual service personnel interaction methods)

[0091] See Figure 1 , Figure 1 This is a flowchart illustrating a virtual service personnel interaction method provided in an embodiment of this application.

[0092] This application provides a virtual service personnel interaction method, the method comprising:

[0093] Step S101: Receive an access request from the first terminal device and establish a communication connection between the first terminal device and the target server, wherein the target server is used to provide virtual service personnel interaction functions;

[0094] Step S102: Use the first terminal device to receive user input information, where each piece of input information is text information, voice information, or image information;

[0095] Step S103: Based on the input information, determine one of the multiple service items as the user's target service item;

[0096] Step S104: Obtain the first interactive video corresponding to the target service item from the Q&A library corresponding to the virtual service staff, where the Q&A library includes first interactive videos corresponding to multiple service items;

[0097] Step S105: Use the first terminal device to play the first interactive video. In the first interactive video, the virtual service staff introduces the办事流程 and / or办事指南 corresponding to the target service item.

[0098] The virtual service staff interaction method can run on an electronic device. The electronic device and the first terminal device (used by the user) can be independent of each other, or the electronic device can be integrated with the first terminal device. When the electronic device and the first terminal device are independent of each other, the electronic device can be a device with computing capabilities such as a computer, a server (including a cloud server), etc. The embodiments of the present application do not limit the first terminal device. For example, it can be a smart first terminal device with a display screen and a speaker such as a mobile phone, a tablet computer, a laptop computer, a desktop computer, a smart wearable device, etc. Or, the first terminal device can be a workstation or a console with a display screen and a speaker. The display screen can be a touch display screen or a non-touch display screen.

[0099] In the embodiments of the present application, the virtual service staff includes one or more of a virtual person, a virtual animal, and a virtual cartoon image. As an example, the virtual service staff is the virtual person "JING" (Chinese name:镜). As an example, the name of the virtual service staff interaction application is "Virtual Service Staff, Making Your办事不再难".

[0100] The embodiments of the present application do not limit the multiple service items. The multiple service items of an enterprise can include, for example, financial services, repairs, returns and exchanges, complaint handling, safety production management, safety accident emergency handling, recruitment, training, performance appraisal, etc. The multiple service items of a government department can include, for example, enterprise registration, land use application, safety production approval, medical insurance reimbursement, social security card replacement, tax exemption application, talent subsidy application, etc.

[0101] Virtual service personnel can provide users with procedures and guides for various service items. For example, a company's virtual service personnel can provide procedures and guides for product or service applications, after-sales service, complaint handling, workplace safety, and human resource management. Product or service application procedures and guides primarily target the company's products or services, such as insurance products and financial services. The company will develop relevant procedures and guides to help customers understand the application process, conditions, and standards for products or services. After-sales service procedures and guides primarily target the company's after-sales services, such as repairs and returns / exchanges. The company will develop relevant procedures and guides to help customers understand the application process, conditions, and standards for after-sales service. Complaint handling procedures and guides primarily target customer complaint handling, such as complaint acceptance, investigation, and resolution. The company will develop relevant procedures and guides to help customers understand the complaint handling process and related regulations. Workplace safety procedures and guides primarily target the company's workplace safety management and emergency response to safety accidents. Companies will develop relevant procedures and guidelines to help employees and customers understand the company's safety production management system and emergency response measures. Human resource management procedures and guidelines primarily target the company's human resource management processes, such as recruitment, training, and performance appraisal. Companies will develop relevant procedures and guidelines to help employees understand the company's human resource management system and related policies and regulations.

[0102] Virtual service personnel from government departments can provide procedures and guides for administrative approvals, public services, policies and regulations, and government information disclosure. Administrative approval procedures and guides primarily target matters requiring administrative approval, such as business registration, land use applications, and work safety approvals. The government will develop relevant procedures and guides to help applicants understand the application process and reduce problems. Public service procedures and guides mainly target basic living services for the public, such as medical services, education services, and social security services. The government will develop relevant procedures and guides to help citizens understand the application procedures, conditions, and standards for these services. Policy and regulation procedures and guides mainly target government policies and regulations, such as tax incentives and subsidies. The government will develop relevant procedures and guides to help relevant units and individuals understand the application conditions and procedures for these policies and regulations. Government information disclosure procedures and guides are a form of government disclosure of information, such as the organizational structure, responsibilities, and work processes of government departments. The government will develop relevant procedures and guidelines to help citizens understand the internal operations of government departments and the implementation of relevant policies.

[0103] The target server can run one or more applications to provide interactive functionality for virtual service personnel. These applications can be written in one or more programming languages, such as Java, Python, Node.js, etc., and can utilize various frameworks and libraries to implement various functions, such as natural language processing, speech-to-text, and image recognition. To improve the availability and performance of the target server, a load balancer can be used to distribute requests across multiple servers, and a failover mechanism can be used to automatically switch to a backup server in case of server failure. Electronic devices and the target server can be independent of each other, or the electronic devices can be integrated with the target server.

[0104] The Q&A database corresponding to the virtual service personnel needs to be stored on one or more servers, which can use cloud storage or their own storage to store the videos.

[0105] The question and answer database corresponding to the virtual service personnel stores the first interactive videos corresponding to multiple service items. For example, the question and answer database corresponding to the virtual service personnel of government departments can store the first interactive videos corresponding to service items such as enterprise registration, land use application, safety production approval, medical insurance reimbursement, social security card replacement, tax reduction and exemption application, and talent subsidy application.

[0106] Therefore, by establishing a communication connection between the first terminal device and the target server, the interactive function of the virtual service personnel is realized, improving the real-time performance and stability of the system and providing users with a smoother online consultation experience. By receiving user input information, including text, voice, and image information, it can meet the needs of different users, provide a more convenient interaction method, and enhance the user experience. By determining the target service item from multiple service items based on user input information, user needs can be quickly identified, improving the targeting and accuracy of the virtual service personnel. Retrieving the first interactive video corresponding to the target service item from the question-and-answer database corresponding to the virtual service personnel provides detailed interactive videos related to the target service item, allowing users to more intuitively understand the relevant information of the target service item, improving user satisfaction and efficiency. In the first interactive video, the virtual service personnel introduce the corresponding service procedures and / or service guides for the target service item, providing users with detailed guidance information, enabling users to more clearly understand the procedures and requirements, improving efficiency and accuracy.

[0107] In summary, the virtual service personnel interaction method provided in this application can receive various types of input information, quickly identify the target service item, and provide detailed procedures and guides, effectively improving user convenience, efficiency, and satisfaction. Simultaneously, this method ensures real-time performance and stability during the interaction between the user and the virtual service personnel, guaranteeing that users can find the required procedures and guides within a short time. Compared to existing technologies, it better adapts to user needs, increases user reliance on and satisfaction with online services, and enhances the user experience. Therefore, this technical solution has high application value and practicality in the field of virtual service personnel interaction, providing enterprises and government departments with an effective way to improve customer service and enhance user experience.

[0108] Give an example to illustrate the interaction process between a user and the virtual service provider "Xiao A".

[0109] Users access the virtual service personnel "Xiao A" of government departments (i.e., access the target server used to provide interactive functions of "Xiao A") through a browser or dedicated government service application (such as "Government Service App") using smartphones or computers (i.e., the first terminal device).

[0110] Users can interact with the virtual service provider "Xiao A" through text, voice, or images. For example, a user might enter the text message: "I want to know about the business registration process." In this example, the user's target service is business registration.

[0111] Based on the input information provided by the user, search for and play the first interactive video related to enterprise registration in the question and answer database corresponding to the virtual service personnel "Xiao A".

[0112] In the first interactive video, the virtual service representative "Xiao A" details the business registration process, including: a. preparing relevant materials, such as a copy of the business license; b. logging into the government website for pre-approval of the company name; c. submitting the business registration application form and related materials; d. paying the registered capital; and e. receiving the business license.

[0113] In addition to the procedures, the virtual service representative "Xiao A" also provides a guide to business registration in the first interactive video, including: a. the conditions that need to be met for business registration; b. the relevant laws and policies for business registration; c. a list of materials that need to be submitted; d. the time requirements and fees for business registration; and e. frequently asked questions and answers.

[0114] Through the steps described above, the virtual service representative "Xiao A" effectively interacted with users, helping them understand the relevant procedures and guidelines for business registration. Users can then more easily complete the business registration process based on the provided information.

[0115] See Figure 2 , Figure 2 This is a schematic diagram of a process for obtaining a target service item provided in an embodiment of this application.

[0116] In some embodiments, determining one service item from a plurality of service items as the user's target service item based on the input information (i.e., step S103) includes:

[0117] Using the semantic extraction model corresponding to the input information, semantic information is extracted from the input information;

[0118] Obtain the similarity between the semantic information and each of the service items, and take the maximum similarity as the highest similarity.

[0119] When the highest similarity is greater than the preset similarity, the service item corresponding to the highest similarity is taken as the target service item.

[0120] In some embodiments, when the input information is text information, the semantic extraction model corresponding to the input information is a pre-trained language model based on deep learning; when the input information is speech information, the semantic extraction model corresponding to the input information includes a speech-to-text model based on deep learning and a pre-trained language model based on deep learning; when the input information is image information, the semantic extraction model corresponding to the input information is a semantic segmentation model based on deep learning.

[0121] Therefore, deep learning-based models are used to extract semantic features from the input information. Different deep learning models are employed for semantic extraction depending on the type of input information, including pre-trained language models, speech-to-text models, and semantic segmentation models.

[0122] For textual information, pre-trained language models, such as BERT and GPT, are used to extract semantic features from the input text. These models, through pre-training on a large amount of text, can learn the structure and semantic information of the language, thus effectively extracting the semantic information of the input text.

[0123] For speech information, deep learning-based speech-to-text models, such as CTC and Transformer, are used to convert speech information into text information, and pre-trained language models are used to extract semantic features from the text information. In this way, semantic information related to the input information can be extracted from the speech information.

[0124] For image information, deep learning-based semantic segmentation models, such as UNet and DeepLab, are used to segment the image and extract the semantic information corresponding to each pixel. This allows for the accurate extraction of semantic information related to the input information from the image.

[0125] The advantages of this approach are that using deep learning-based semantic extraction models can more accurately extract the semantic features of input information, thereby improving the accuracy and efficiency of semantic understanding; using pre-trained language models can improve the capabilities of natural language processing, including text classification, sentiment analysis, and machine translation; and employing different deep learning models can enable the processing of multimodal information, including text, speech, and image information, thus achieving more comprehensive information processing.

[0126] Preset similarity refers to a preset similarity threshold to determine the degree of matching between input information and target service items. This application does not limit the preset similarity; it can be, for example, 0.5, 0.65, 0.75, 0.8, 0.95, etc.

[0127] Therefore, by using a semantic extraction model to extract semantic information from input information, we can more accurately understand user needs and effectively improve the response accuracy of virtual service personnel. By calculating the similarity between semantic information and various service items, we can quickly find the service item that best matches the user's input information, further improving the targeting and accuracy of virtual service personnel. By setting a preset similarity, we can effectively avoid misjudgments and incorrect matching, ensuring that only highly matching service items are provided to users as target service items. This technical solution enhances the accuracy and reliability of virtual service personnel in handling complex problems and fuzzy inputs, thereby improving user satisfaction and efficiency. In summary, by adopting a semantic extraction model, calculating similarity, and setting a preset similarity, this technical solution can more accurately determine the user's target service item, providing users with more precise and personalized services. This implementation method has high practicality and efficiency in the field of virtual service personnel interaction, helping to further improve user experience and satisfaction, and enabling enterprises and government departments to better meet the needs of various groups.

[0128] See also Figure 2 In some embodiments, determining one service item from multiple service items as the user's target service item based on the input information (i.e., step S103) further includes:

[0129] When the highest similarity is not greater than the preset similarity, the first terminal device is used to play the second interactive video to guide the user to click the form button and fill in the form;

[0130] When a click operation is received on the form button, the form is displayed using the first terminal device. The form is provided with a service item input box and / or a service item drop-down menu.

[0131] Based on the form's completion results, the user's target service item is obtained.

[0132] Therefore, when the highest similarity score is not greater than the preset similarity score, playing a second interactive video to guide the user to click the form button and fill out the form helps to collect more information about the user's needs, more effectively obtain the user's detailed requirements, and improve the accuracy of the target service items, thereby providing users with more accurate and personalized services. The form is equipped with service item input boxes and / or service item drop-down menus, giving users greater flexibility when providing their needs, and also making it convenient for users to select and input the required service items. It provides an effective supplementary means when virtual service personnel cannot accurately identify user needs, helping to improve user satisfaction and efficiency. In summary, this technical solution, when the highest similarity score does not meet the preset condition, guides users to fill out a form to obtain more information about their needs, thereby better meeting the user's actual needs and improving overall efficiency and user satisfaction.

[0133] Give an example to illustrate the interaction process between a user and the virtual service provider "Xiao A".

[0134] Users access the virtual service personnel "Xiao A" of government departments through a browser or a dedicated government service application (such as "Government Affairs Connect") using a smartphone or computer (i.e., the first terminal device).

[0135] Users can interact with the virtual service provider "Xiao A" via voice. For example, a user can input a voice message saying, "I want to know about the process of setting up a company."

[0136] Use a semantic extraction model corresponding to the speech information to extract semantic information from the input information, such as "company establishment" or "information".

[0137] The virtual service provider "Xiao A" can offer a variety of services, such as business registration, business registration changes, and tax registration.

[0138] The similarity between the semantic information extracted from the input information and each service item is calculated. For example, "Business Registration: 0.90, Business Registration Change: 0.65, Tax Registration: 0.70". In this example, the highest similarity is with the business registration service item, which is 0.90. The preset similarity is, for example, 0.75.

[0139] When the highest similarity is greater than the preset similarity (0.90 > 0.75 in this example), the service item corresponding to the highest similarity will be taken as the target service item, i.e., enterprise registration.

[0140] If the highest similarity score is no greater than a preset similarity score, the first terminal device will play a second interactive video of the virtual service provider "Xiao A," guiding the user to click the form button and fill out the form. After the user clicks the form button, the first terminal device will display a form with service item input boxes and / or service item drop-down menus to allow the user to provide more detailed information about their needs. The user fills in the corresponding information on the form, such as "Business Registration." Based on the results, it is determined that the user's target service item is business registration.

[0141] Through the steps described above, the virtual service provider "Xiao A" can determine the user's desired service by calculating similarity and guiding the user to fill out a form when the user provides vague information. This helps to provide more accurate services and meet user needs.

[0142] In some embodiments, the process of obtaining the similarity between the semantic information and each of the service items includes:

[0143] For each of the aforementioned service items, perform the following processing:

[0144] Input the semantic information and the preset text corresponding to the service item into a similarity model to obtain the similarity between the semantic information and the preset text, which is used as the similarity between the semantic information and the service item;

[0145] The training process of the similarity model includes:

[0146] Obtain a training set, which includes multiple training data, each of which includes a sample semantic information, a sample text, and annotation data on the similarity between the sample semantic information and the sample text;

[0147] For each of the training data, perform the following processing:

[0148] The sample semantic information and sample text in the training data are input into a preset deep learning model to obtain the predicted similarity data between the sample semantic information and the sample text.

[0149] The model parameters of the deep learning model are updated based on the sample semantic information and the predicted and labeled data of the similarity between the sample text and the sample semantic information.

[0150] Check whether the preset training termination condition is met; if yes, use the trained deep learning model as the similarity model; if no, continue to train the deep learning model using the next training data.

[0151] Therefore, by inputting semantic information and pre-defined text into the similarity model for each service item, the similarity between semantic information and each service item can be calculated more accurately, improving the matching accuracy of virtual service personnel. The training process of the similarity model employs a deep learning model, utilizing sample semantic information and sample text from the training data. This allows the similarity model to better capture the correlation between semantic information and text, improving the accuracy of similarity calculation. By detecting pre-defined training termination conditions, it can be ensured that the similarity model completes training while meeting certain accuracy requirements, thereby improving the model's generalization ability and practicality. This technical solution, by using a deep learning model for similarity calculation, can better adapt to service items in different fields, improving the versatility and application value of virtual service personnel in various scenarios. It exhibits high accuracy and reliability in processing user input information, thereby improving user satisfaction and efficiency, enabling enterprises and government departments to better meet customer needs. In summary, this technical solution, by training a similarity model using a deep learning model, achieves more accurate similarity calculation between semantic information and service items, thus improving the accuracy and practicality of virtual service personnel in identifying target service items.

[0152] See Figure 3 , Figure 3 This is a schematic diagram of a process for prompting preset personnel to update, provided in an embodiment of this application.

[0153] In some embodiments, the method further includes:

[0154] The first terminal device is used to receive feedback information from the user, and each piece of feedback information is text information, voice information, or image information;

[0155] Based on the feedback information, identify the user's emotion category;

[0156] Based on the user's emotion category, determine whether the user is satisfied with the service;

[0157] If not satisfied, increment the cumulative count by one;

[0158] When the cumulative number of times exceeds the preset number of times, the cumulative number of times is cleared to zero, and a prompt message is sent to the second terminal device of the preset personnel corresponding to the target service item, so as to prompt the preset personnel to update the similarity model or update the preset text corresponding to the target service item.

[0159] In some embodiments, the method further includes: if satisfied, then doing nothing.

[0160] In some embodiments, the method further includes: when the cumulative number of times is not greater than the preset number of times, no operation is performed.

[0161] The embodiments of this application do not limit the preset number of times, which may be, for example, 3 times, 5 times, 10 times, 100 times, etc.

[0162] In this application embodiment, the preset personnel may be, for example, designated staff responsible for updating the similarity model or preset text.

[0163] The embodiments of this application do not limit the second terminal device, which may be, for example, a mobile phone, tablet computer, laptop computer, desktop computer, smart wearable device or other smart terminal device, or the second terminal device may be a workstation or console.

[0164] In some embodiments, identifying the user's emotion category based on the feedback information includes: inputting the feedback information into an emotion recognition model to identify the user's emotion category. The emotion recognition model may, for example, be a deep learning-based emotion recognition model.

[0165] The embodiments of this application do not limit the categories of emotions, which may include, for example, happiness, sadness, anger, fear, surprise, anxiety, disgust, neutrality, etc.

[0166] In some embodiments, detecting whether a user is satisfied with the service based on the user's emotion category includes: obtaining a correspondence between emotion categories and satisfaction assessment results, wherein the satisfaction assessment results indicate satisfaction or dissatisfaction; obtaining a satisfaction assessment result corresponding to the user's emotion category based on the user's emotion category and the correspondence; and confirming whether the user is satisfied or dissatisfied with the service based on the satisfaction assessment result corresponding to the user's emotion category.

[0167] In other words, a pre-established correspondence between emotion categories and satisfaction assessment results can be created. For example, when the emotion category is "happy," the corresponding satisfaction assessment result is "satisfied"; when the emotion category is "anxious," the corresponding satisfaction assessment result is "dissatisfied." Then, based on the user's satisfaction assessment result corresponding to their emotion category, it is determined whether the user is satisfied with the service. If the user's satisfaction assessment result is "satisfied," then the user is confirmed to be satisfied with the service; if the user's satisfaction assessment result is "dissatisfied," then the user is confirmed to be dissatisfied with the service.

[0168] In this embodiment, each service item corresponds to a preset text and a preset action. The first interactive video corresponding to the service item in the question-and-answer database is synthesized based on the preset text and preset action of the service item. Specifically, the synthesis process may include: obtaining the preset text and preset action corresponding to the service item; performing speech synthesis based on the preset text to obtain speech data; driving the lip movements and facial expressions of the virtual service personnel based on the speech data to obtain facial driving data; driving the actions of the virtual service personnel based on the preset action to obtain action driving data; and generating the first interactive video corresponding to the service item based on the speech data, the facial driving data, and the action driving data. In this first interactive video, the lip movements, facial expressions, and actions of the virtual service personnel will change during the interaction process to closely resemble a real service personnel.

[0169] Therefore, by receiving user feedback and identifying user emotion categories, it is possible to better understand user satisfaction with the services provided by virtual service personnel, which helps to further optimize service quality. Based on the user's emotion category, detecting whether the user is satisfied with the service allows for the timely identification of problematic service items, enabling targeted improvements. When the cumulative number of occurrences exceeds a preset limit, a prompt message is sent to the second terminal device of the preset personnel, helping to remind relevant personnel to update the similarity model or adjust the preset text, improving the matching accuracy of virtual service personnel and user satisfaction. This method enhances the accuracy and reliability of virtual service personnel in handling user needs, thereby improving user satisfaction and efficiency, enabling enterprises and government departments to better meet customer needs. Through user feedback and emotion recognition, this technical solution helps to collect valuable user opinions and suggestions, providing strong support for further optimizing the work effectiveness of virtual service personnel, ensuring that relevant personnel pay timely attention to user satisfaction issues, and preventing a decline in service quality. In summary, this technical solution, by collecting user feedback information, identifying user emotion categories, and sending prompt messages, can continuously optimize the service quality and matching accuracy of virtual service personnel.

[0170] Give an example to illustrate the interaction process between a user and the virtual service provider "Xiao A".

[0171] Users access the virtual service personnel "Xiao A" of government departments through a browser or a dedicated government service application (such as "Government Affairs Connect") using a smartphone or computer (i.e., the first terminal device).

[0172] Users can provide feedback to the virtual service provider "Xiao A" via text, voice, or images. For example, a user could input a voice message: "The business registration process you just provided is not very clear." In this example, the target service is business registration.

[0173] By analyzing user feedback, we can identify the user's emotional category, such as "dissatisfaction".

[0174] Based on the user's emotion category, determine whether the user is satisfied with the service. In this example, the user is dissatisfied with the service.

[0175] Each time a user is dissatisfied with the service, the cumulative count is incremented by one. For example, the current cumulative count is 3. The preset count is, for example, 5 times, which serves as the threshold for user dissatisfaction with the service.

[0176] When the cumulative number of requests exceeds the preset number (e.g., the cumulative number reaches 6 times), a notification message is sent to the preset personnel's mobile phone (i.e., the second terminal device). The message may read: "We have received multiple unsatisfactory feedbacks regarding the business registration service. Please check and update the similarity model or the preset text." The preset text is the standard response used by the virtual service personnel "Xiao A" when answering user questions, such as an explanation of the business registration process.

[0177] In some embodiments, the method further includes:

[0178] The first terminal device is used to display a plurality of second interactive videos related to the first interactive video, each of which is used to introduce the precautions, frequently asked questions or related service items of the target service item;

[0179] When a click operation is received for one of the second interactive videos, the clicked second interactive video is played using the first terminal device.

[0180] Therefore, by displaying multiple second interactive videos related to the first interactive video on the first terminal device, users can be provided with richer and more comprehensive service information, helping them better understand the precautions, frequently asked questions, or related service items for the target service. Users can choose to watch the second interactive videos related to the target service based on their needs and interests, increasing user initiative and interactivity in the process of obtaining service information, thereby improving user satisfaction. This method provides users with a more convenient and personalized way to obtain information, helping to meet the needs and preferences of different users.

[0181] Give an example to illustrate the interaction process between a user and the virtual service provider "Xiao A".

[0182] Users access the virtual service provider "Xiao A" of government departments using a smartphone or computer (i.e., the first terminal device) through a browser or a dedicated government service application (such as "Government Affairs App"). In this example, the target service is business registration.

[0183] The virtual service provider "Xiao A" plays an interactive video introducing the business registration process and service guide based on the user's questions about business registration.

[0184] After playing the first interactive video, the first terminal device displays several second interactive videos related to business registration, such as: "Materials Required for Business Registration", "Tax Procedures for Business Registration", and "Frequently Asked Questions about Business Registration".

[0185] Important considerations for business registration, such as the valid documents required during the registration process and age restrictions for applicants. Common issues related to business registration, such as choosing the right business type, registered capital requirements, registered address, company name regulations, and restrictions on the legal representative's identity. Related services for business registration, such as tax registration and social insurance registration.

[0186] When a user clicks on one of the second interactive videos, such as "Materials Required for Business Registration," the first terminal device will play the video, which details the various materials and submission requirements needed for business registration.

[0187] See Figure 4 , Figure 4 This is a schematic diagram of a process for downloading a proof document provided in an embodiment of this application.

[0188] In some embodiments, during the first interactive video, the virtual service personnel guide the user to click the document download button to download the certification document;

[0189] The method further includes:

[0190] When a click operation is received on the document download button, the certificate document corresponding to the document download button is downloaded to the first terminal device.

[0191] This application does not limit the type of supporting documents, which may include, for example, birth certificates, household registration books, identity certificates, marriage certificates, divorce certificates, academic certificates, employment certificates, social security certificates, property ownership certificates, etc.

[0192] Therefore, in the first interactive video, the virtual service personnel guide the user to click the document download button, enabling the user to easily obtain the required supporting documents, thus improving user experience and satisfaction. By responding to the user's click, the supporting documents are downloaded to the first terminal device, realizing the convenience and efficiency of the document download function and saving the user's time and effort. This technical solution simplifies the process of obtaining supporting documents, thereby improving efficiency. Implementing the supporting document download function during the interaction with the virtual service personnel further enriches the services that the virtual service personnel can provide, allowing users to complete more service items on a single platform.

[0193] For example, suppose a user wants to know how to download academic credentials. They can access a specific website or mobile app using their smartphone or computer. On the website or app, a first interactive video featuring a virtual service provider, "A," plays. In this video, A explains the uses of academic credentials, such as job applications, promotions, and studying abroad, and how to obtain them.

[0194] In the first interactive video, "Xiao A" guides the user to click the "Document Download" button on the page, explaining that clicking this button will automatically download the academic certificate document. The user then clicks the "Document Download" button as instructed by Xiao A. Based on the user's click, the academic certificate document is downloaded to the user's primary device. This interactive method allows users to conveniently obtain their academic certificate documents, improving efficiency. Simultaneously, the virtual service representative "Xiao A" provides detailed guidance throughout the process, making the process of obtaining academic certificates clearer and more efficient for users.

[0195] See also Figure 4 In some embodiments, the step of downloading the certificate document corresponding to the document download button to the first terminal device when a click operation is received on the document download button includes:

[0196] When a click operation is received on the document download button, the first terminal device displays the document download interface corresponding to the document download button, and the document download interface is provided with an identity recognition button.

[0197] When a click operation is received on the identity recognition button, the first terminal device is used to obtain the user's identity information to be identified;

[0198] Detect whether the identity information to be identified matches the user's authenticated identity information;

[0199] If a match is found, the proof document is downloaded to the first terminal device.

[0200] In this embodiment, the identity information to be identified may be, for example, an account and password to be identified, fingerprint information, facial information, voiceprint information, iris information, etc. Correspondingly, the authenticated identity information is the authenticated account and password, fingerprint information, facial information, voiceprint information, iris information, etc.

[0201] Therefore, upon receiving a click on the document download button, a document download interface is displayed. This interface includes an identity verification button to trigger user authentication, providing a secure identity verification mechanism that ensures the security and privacy of downloaded documents. After the user clicks the identity verification button, the first terminal device obtains the user's identity information, helping to ensure that the obtained document matches their identity. By detecting the match between the identity information to be identified and the authenticated identity information, malicious users can be prevented from downloading others' documents to a certain extent, improving information security. If the identity information to be identified matches the authenticated identity information, the document will be downloaded to the first terminal device, allowing the user to successfully obtain the required document, improving user experience and satisfaction.

[0202] See also Figure 4 In some embodiments, the step of downloading the proof document to the first terminal device if a match is found includes:

[0203] If a match is found, the user information database corresponding to the target service item is queried based on the authenticated identity information to generate the proof document based on the query results.

[0204] Download the proof document to the first terminal device.

[0205] Therefore, based on the authenticated identity information, the system queries the user information database corresponding to the target service item, generates a supporting document based on the query results, and ensures that the generated supporting document is based on the user's real information, thus guaranteeing its timeliness, reliability, and accuracy. During the interaction with virtual service personnel, the system enables the generation and download of supporting documents, further enriching the service content that virtual service personnel can provide.

[0206] For example, suppose a user wants to download their social security certificate. They can use a smartphone or computer to access the government department's website or mobile application. On the website or mobile application, the user enters the text: "I need a social security certificate to buy a house." The user's target service is then downloading the social security certificate.

[0207] The first terminal device plays an interactive video related to the social security certificate document, in which Xiao A explains in detail how to download the document. After the video finishes playing, the user clicks the "Document Download Button" on the page, which redirects to the document download interface, which includes an "Identity Recognition Button." When the user clicks the "Identity Recognition Button," the first terminal device obtains the user's identity information to be identified, such as facial information. It checks whether the user's identity information to be identified matches the user's already authenticated identity information (such as facial information submitted during registration or authentication on various government websites or mobile applications). If the identity information to be identified matches the already authenticated identity information, the device queries the user information database corresponding to the social security certificate document based on the authenticated identity information, generates the social security certificate document based on the query results, and downloads the generated document to the user's first terminal device.

[0208] In some embodiments, the method further includes:

[0209] Based on the user's authenticated identity information and the interaction data between the user and the virtual service personnel, one or more recommendation policies corresponding to the user are obtained and displayed in list form on the first terminal device.

[0210] In this embodiment of the application, the user's authenticated identity information may also include name, ID number, age, occupation, region, work unit, work address, position, job title, home address, education, etc.

[0211] In this embodiment, interactive data may include, for example, user input information (such as text, voice, and images), operation records (clicks, swipes, etc.), browsing records (page visits, dwell time, etc.), feedback information (satisfaction, suggestions, etc.), and personal information (authenticated identity information). This data helps improve the service quality of virtual service personnel, personalized recommendations, and user experience.

[0212] Therefore, by combining a user's authenticated identity information with interaction data between the user and virtual service personnel, one or more recommended policies corresponding to the user can be obtained more accurately, thereby improving the targeting and accuracy of recommendations. Displaying recommended policies in list form on the first terminal device provides users with a clear and intuitive way of presenting information, helping them quickly browse and understand policies relevant to them. This method implements a policy recommendation function, helping users to understand policy information relevant to themselves in a timely manner, increasing users' attention to and awareness of policies, helping them benefit from policies, and preventing them from missing out on policy benefits relevant to them. By recommending policies related to users' identity and needs, it helps to improve the coverage and implementation effect of policies, thereby improving the overall effectiveness of policy implementation.

[0213] Let's take a user who is a startup entrepreneur as an example to illustrate the interaction process between the user and "Xiao A," a virtual service provider for the government department.

[0214] Users access government websites or mobile applications using smartphones or computers (i.e., the first terminal device) to interact with virtual service personnel "Xiao A" and inquire about relevant information such as entrepreneurship support policies and tax incentives.

[0215] Based on users' verified identity information (such as name, ID number, age, occupation, and region) and interaction data (such as questions and keywords asked by users), we analyze policies that users may be interested in, such as entrepreneurship support policies, tax incentives, and talent training policies in the user's region.

[0216] The system compiles and lists the entrepreneurship support policies, tax incentives, and talent training policies in the user's region and displays them on the user's primary device.

[0217] For example, the list of recommended policies may include:

[0218] Startup support policies: Government policies such as financial support and preferential venues for start-ups.

[0219] Tax incentives: Tax breaks for startups to reduce their operating costs.

[0220] Talent training policy: The government provides professional training and skills enhancement courses for startups.

[0221] This application does not limit the acquisition method of semantic extraction model and similarity model. In some embodiments, the above-mentioned models can be trained, and in other embodiments, the above-mentioned models can be pre-trained.

[0222] When deep learning is used to train various models, by designing and establishing an appropriate number of neural computing nodes and a multi-layered computational hierarchy, and by selecting suitable input and output layers, a pre-defined deep learning model (i.e., the initial model corresponding to each model) can be obtained. Through the learning and optimization of deep learning models, a functional relationship from input to output can be established. Although it is not possible to find a 100% accurate functional relationship between input and output, it can approximate the real-world correlation as closely as possible. The models trained in this way can obtain corresponding output data based on input data, have a wide range of applications, and provide high accuracy and reliability in calculation results.

[0223] Training deep learning models using the training sets corresponding to each model allows for rapid modeling by learning from only a small number of samples. The training error of the deep learning model gradually decreases during continuous training. The optimal weights are saved and retrieved. The accuracy of the training and validation sets is recorded for parameter tuning (adjusting model parameters). Updating the model parameters of the deep learning model can enable the model to better fit the data, have effective generalization ability, and improve robustness and fitting accuracy.

[0224] In some alternative implementations, data mining can be performed on historical data to obtain sample data in the training set. That is, this sample data can be collected during real interactions between users and human service personnel (i.e., real service personnel). Alternatively, the sample data can be automatically generated using the generative network of a GAN model.

[0225] GAN, or Generative Adversarial Network, consists of a generator network and a discriminator network. The generator network randomly samples data from the latent space as input, and its output should closely mimic real samples in the training set. The discriminator network takes either real samples or the generator network's output as input, aiming to distinguish the generator network's output from real samples as much as possible. The generator network, in turn, tries to deceive the discriminator network. The two networks compete against each other, constantly adjusting their parameters, with the ultimate goal of making it impossible for the discriminator network to determine whether the generator network's output is genuine. Using GAN models allows for the generation of large amounts of sample data for training the aforementioned models, effectively reducing the amount of raw data collected and significantly lowering the costs of data acquisition and annotation.

[0226] The embodiments of this application do not limit the training process of each model. For example, it may adopt a supervised learning training method, a semi-supervised learning training method, or an unsupervised learning training method.

[0227] When using supervised or semi-supervised learning training methods, this application embodiment does not limit the method of obtaining labeled data. For example, manual labeling, automatic labeling, or semi-automatic labeling can be used. When sample data is collected during real-world interactions, real data can be obtained from historical data as labeled data through keyword extraction.

[0228] The embodiments of this application do not limit the training termination conditions during the training process of each model. For example, it may be that the number of training times reaches a preset number (the preset number of times is, for example, 1 time, 3 times, 10 times, 100 times, 1000 times, 10000 times, etc.), or it may be that the training data in the training set has completed one or more training sessions, or it may be that the total loss value obtained in this training is not greater than the preset loss value.

[0229] In a specific application scenario, this application embodiment also provides a virtual service personnel interaction method, the method comprising:

[0230] Receive an access request from a first terminal device, establish a communication connection between the first terminal device and a target server, wherein the target server is used to provide virtual service personnel interaction functions;

[0231] The first terminal device is used to receive user input information, each of which is text information, voice information, or image information;

[0232] Using the semantic extraction model corresponding to the input information, semantic information is extracted from the input information;

[0233] Obtain the similarity between the semantic information and each of the service items, and take the maximum similarity as the highest similarity.

[0234] When the highest similarity is greater than the preset similarity, the service item corresponding to the highest similarity is taken as the target service item;

[0235] When the highest similarity is not greater than a preset similarity, the first terminal device is used to play a second interactive video to guide the user to click the form button and fill in the form; when a click operation is received on the form button, the first terminal device is used to display the form, which is provided with a service item input box and / or a service item drop-down menu; based on the form filling result, the user's target service item is obtained;

[0236] Obtain the first interactive video corresponding to the target service item from the question and answer database corresponding to the virtual service personnel, wherein the question and answer database includes first interactive videos corresponding to multiple service items;

[0237] The first interactive video is played using the first terminal device. In the first interactive video, the virtual service personnel introduce the procedures and / or guidelines for handling the target service item. In addition, in the first interactive video, the virtual service personnel guide the user to click the document download button to download the certificate document.

[0238] The first terminal device is used to receive the user's feedback information;

[0239] Based on the feedback information, identify the user's emotion category;

[0240] Based on the user's emotion category, determine whether the user is satisfied with the service;

[0241] If satisfied, no action is required;

[0242] If not satisfied, increment the cumulative count by one;

[0243] When the cumulative number of times exceeds the preset number of times, the cumulative number of times is cleared to zero, and a prompt message is sent to the second terminal device of the preset personnel corresponding to the target service item, so as to prompt the preset personnel to update the similarity model or update the preset text corresponding to the target service item;

[0244] No operation is performed when the cumulative number of times is not greater than the preset number of times;

[0245] The first terminal device is used to display a plurality of second interactive videos related to the first interactive video, each of which is used to introduce the precautions, frequently asked questions or related service items of the target service item;

[0246] When a click operation is received for one of the second interactive videos, the clicked second interactive video is played using the first terminal device;

[0247] When a click operation is received on the document download button, the first terminal device displays the document download interface corresponding to the document download button, and the document download interface is provided with an identity recognition button.

[0248] When a click operation is received on the identity recognition button, the first terminal device is used to obtain the user's identity information to be identified;

[0249] Detect whether the identity information to be identified matches the user's authenticated identity information;

[0250] If there is no match, no action is taken;

[0251] If a match is found, the user information database corresponding to the target service item is queried based on the authenticated identity information to generate the proof document based on the query results.

[0252] Download the proof document to the first terminal device;

[0253] Based on the user's authenticated identity information and the interaction data between the user and the virtual service personnel, one or more recommendation policies corresponding to the user are obtained and displayed in list form on the first terminal device.

[0254] The process of obtaining the similarity between the semantic information and each of the service items includes:

[0255] For each of the aforementioned service items, perform the following processing:

[0256] Input the semantic information and the preset text corresponding to the service item into a similarity model to obtain the similarity between the semantic information and the preset text, which is used as the similarity between the semantic information and the service item;

[0257] The training process of the similarity model includes:

[0258] Obtain a training set, which includes multiple training data, each of which includes a sample semantic information, a sample text, and annotation data on the similarity between the sample semantic information and the sample text;

[0259] For each of the training data, perform the following processing:

[0260] The sample semantic information and sample text in the training data are input into a preset deep learning model to obtain the predicted similarity data between the sample semantic information and the sample text.

[0261] The model parameters of the deep learning model are updated based on the sample semantic information and the predicted and labeled data of the similarity between the sample text and the sample semantic information.

[0262] Check whether the preset training termination condition is met; if yes, use the trained deep learning model as the similarity model; if no, continue to train the deep learning model using the next training data.

[0263] Suppose user MING wants to know about the process of renewing his driver's license. Here is an example of his interaction with "Xiao A," a virtual service representative from the government department.

[0264] MING uses a mobile phone to access government websites or mobile applications by sending an access request. The first terminal device (mobile phone) establishes a communication connection with the target server, which provides interactive functions with virtual service personnel.

[0265] MING enters the text message: "How do I renew my driver's license?" The phone then sends the entered information to the server.

[0266] The server uses a semantic extraction model to extract semantic information from the input information.

[0267] The server calculates the similarity between the input semantic information and each service item, and finds the service item with the highest similarity, namely "driver's license renewal".

[0268] The server retrieves the first interactive video corresponding to "driver's license renewal" from the question and answer database of the virtual service provider "Xiao A".

[0269] MING's phone played the first interactive video. In the video, a virtual service representative, "Xiao A," introduced the process and guidelines for renewing a driver's license, guiding MING to click the document download button to download the supporting documents.

[0270] In the first interactive video, MING clicked the document download button. The phone then displayed the document download interface, which included an identity verification button.

[0271] When MING clicks the identity verification button, the phone obtains MING's identity information to be verified.

[0272] The server checks whether the identity information to be identified matches the already authenticated identity information.

[0273] Once a match is detected, the server queries the user information database corresponding to the driver's license renewal service based on the authenticated identity information and generates a certificate document.

[0274] The proof document was downloaded to MING's mobile phone.

[0275] Based on MING's verified identity information and interaction data with virtual service personnel, the server obtains one or more recommended policies corresponding to MING, including preferential policies for purchasing new energy vehicles, driver safety education and training policies, traffic congestion relief policies, insurance preferential policies, and holiday toll preferential policies, and displays them in a list on the mobile phone.

[0276] As can be seen, user MING easily obtained the information and supporting documents required for renewing his driver's license by interacting with the virtual service provider "Xiao A," and also received relevant recommendation policies.

[0277] Furthermore, personalized recommendation functionality can be added to the virtual service interaction method to provide users with more relevant service suggestions. When determining target service items, the user's historical interaction records, interests, and needs are taken into account to provide more personalized service recommendations that meet user requirements. When determining service items based on user input, a multimodal matching method can be introduced, comprehensively considering the user's text, voice, and image information to improve the matching accuracy of service items. When obtaining semantic information and service item similarity, transfer learning methods can be used, utilizing pre-trained deep learning models for rapid training and optimization to improve model training efficiency and matching accuracy. In the user feedback stage, an image acquisition component can be used to capture user facial images in real time. Combining feedback information and user facial images, the user's emotion category can be identified and user satisfaction can be monitored. When user dissatisfaction is detected, proactive solutions or suggestions can be provided to adjust service strategies in a timely manner and improve user experience. When playing the first and second interactive videos, adaptive streaming media technology can be used to automatically adjust video clarity based on the user's network environment and device performance, ensuring a consistent viewing experience for users in different network environments. In the process of guiding users to download supporting documents, digital signatures and encryption functions can be added to ensure the security and integrity of the downloaded documents. In the identity verification process, multiple verification methods can be supported, such as facial recognition and fingerprint recognition, improving the convenience and security of user verification. In the document generation process, various formats of supporting documents, such as PDF and Word, can be provided according to user needs and preferences, allowing users to easily download the format they require.

[0278] In this embodiment of the application, when a virtual service representative guides a user through operations, different versions of the first interactive video can be provided for different users based on their operating habits, skill levels, and needs, achieving more intelligent and personalized guidance and improving the user experience. For example, for elderly users and visually impaired users, the font size of the first interactive video is larger and the font is bolder; for children, the background music of the first interactive video adopts a cheerful style to soothe children's emotions and shorten the distance between the user and the children.

[0279] (Electronic devices)

[0280] This application also provides an electronic device, the specific embodiments of which are consistent with the embodiments and technical effects achieved in the above method embodiments, and some contents will not be repeated.

[0281] Secondly, this application provides an electronic device including a memory and at least one processor, the memory storing a computer program, and the at least one processor being configured to execute the computer program to perform the following steps:

[0282] Receive an access request from a first terminal device, establish a communication connection between the first terminal device and a target server, wherein the target server is used to provide virtual service personnel interaction functions;

[0283] The first terminal device is used to receive user input information, each of which is text information, voice information, or image information;

[0284] Based on the input information, one service item is determined from multiple service items as the user's target service item;

[0285] Obtain the first interactive video corresponding to the target service item from the question and answer database corresponding to the virtual service personnel, wherein the question and answer database includes first interactive videos corresponding to multiple service items;

[0286] The first interactive video is played using the first terminal device, in which the virtual service personnel introduce the procedures and / or guidelines corresponding to the target service item.

[0287] In some embodiments, the at least one processor is configured to, when executing the computer program, determine one of a plurality of service items as the user's target service item based on the input information in the following manner:

[0288] Using the semantic extraction model corresponding to the input information, semantic information is extracted from the input information;

[0289] Obtain the similarity between the semantic information and each of the service items, and take the maximum similarity as the highest similarity.

[0290] When the highest similarity is greater than the preset similarity, the service item corresponding to the highest similarity is taken as the target service item.

[0291] In some embodiments, when the at least one processor is configured to execute the computer program, it further determines one of a plurality of service items as the user's target service item based on the input information in the following manner:

[0292] When the highest similarity is not greater than the preset similarity, the first terminal device is used to play the second interactive video to guide the user to click the form button and fill in the form;

[0293] When a click operation is received on the form button, the form is displayed using the first terminal device. The form is provided with a service item input box and / or a service item drop-down menu.

[0294] Based on the form's completion results, the user's target service item is obtained.

[0295] In some embodiments, the at least one processor is configured to obtain the similarity between the semantic information and each of the service items when executing the computer program in the following manner:

[0296] For each of the aforementioned service items, perform the following processing:

[0297] Input the semantic information and the preset text corresponding to the service item into a similarity model to obtain the similarity between the semantic information and the preset text, which is used as the similarity between the semantic information and the service item;

[0298] The training process of the similarity model includes:

[0299] Obtain a training set, which includes multiple training data, each of which includes a sample semantic information, a sample text, and annotation data on the similarity between the sample semantic information and the sample text;

[0300] For each of the training data, perform the following processing:

[0301] The sample semantic information and sample text in the training data are input into a preset deep learning model to obtain the predicted similarity data between the sample semantic information and the sample text.

[0302] The model parameters of the deep learning model are updated based on the sample semantic information and the predicted and labeled data of the similarity between the sample text and the sample semantic information.

[0303] Check whether the preset training termination condition is met; if yes, use the trained deep learning model as the similarity model; if no, continue to train the deep learning model using the next training data.

[0304] In some embodiments, when the at least one processor is configured to execute the computer program, it further performs the following steps:

[0305] The first terminal device is used to receive the user's feedback information;

[0306] Based on the feedback information, identify the user's emotion category;

[0307] Based on the user's emotion category, determine whether the user is satisfied with the service;

[0308] If not satisfied, increment the cumulative count by one;

[0309] When the cumulative number of times exceeds the preset number of times, the cumulative number of times is cleared to zero, and a prompt message is sent to the second terminal device of the preset personnel corresponding to the target service item, so as to prompt the preset personnel to update the similarity model or update the preset text corresponding to the target service item.

[0310] In some embodiments, when the at least one processor is configured to execute the computer program, it further performs the following steps:

[0311] The first terminal device is used to display a plurality of second interactive videos related to the first interactive video, each of which is used to introduce the precautions, frequently asked questions or related service items of the target service item;

[0312] When a click operation is received for one of the second interactive videos, the clicked second interactive video is played using the first terminal device.

[0313] In some embodiments, during the first interactive video, the virtual service personnel guide the user to click the document download button to download the certification document;

[0314] When the at least one processor is configured to execute the computer program, it further performs the following steps:

[0315] When a click operation is received on the document download button, the certificate document corresponding to the document download button is downloaded to the first terminal device.

[0316] In some embodiments, the at least one processor is configured to download the verification document corresponding to the document download button to the first terminal device when executing the computer program in the following manner:

[0317] When a click operation is received on the document download button, the first terminal device displays the document download interface corresponding to the document download button, and the document download interface is provided with an identity recognition button.

[0318] When a click operation is received on the identity recognition button, the first terminal device is used to obtain the user's identity information to be identified;

[0319] Detect whether the identity information to be identified matches the user's authenticated identity information;

[0320] If a match is found, the proof document is downloaded to the first terminal device.

[0321] In some embodiments, the at least one processor is configured to download the certification document to the first terminal device in the following manner when executing the computer program:

[0322] If a match is found, the user information database corresponding to the target service item is queried based on the authenticated identity information to generate the proof document based on the query results.

[0323] Download the proof document to the first terminal device.

[0324] In some embodiments, when the at least one processor is configured to execute the computer program, it further performs the following steps:

[0325] Based on the user's authenticated identity information and the interaction data between the user and the virtual service personnel, one or more recommendation policies corresponding to the user are obtained and displayed in list form on the first terminal device.

[0326] See Figure 5 , Figure 5 This is a structural block diagram of an electronic device 10 provided in an embodiment of this application.

[0327] Electronic device 10 may include, for example, at least one memory 11, at least one processor 12, and a bus 13 connecting different platform systems.

[0328] The memory 11 may include a (computer) readable medium in the form of volatile memory, such as random access memory (RAM) 111 and / or cache memory 112, and may further include read-only memory (ROM) 113.

[0329] The memory 11 also stores a computer program, which can be executed by the processor 12 to enable the processor 12 to implement the steps of any of the above methods.

[0330] The memory 11 may also include a utility 114 having at least one program module 115, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0331] Accordingly, processor 12 can execute the aforementioned computer program, and can also execute utility 114.

[0332] The processor 12 may employ one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0333] Bus 13 can represent one or more types of bus structures, including a memory bus or memory controller, peripheral bus, graphics acceleration port, processor, or a local bus using any bus structure with multiple bus structures.

[0334] Electronic device 10 can also communicate with one or more external devices, such as keyboards, pointing devices, Bluetooth devices, etc., and with one or more devices capable of interacting with it, and / or with any device that enables it to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication can be performed through input / output interface 14. Furthermore, electronic device 10 can communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 15. Network adapter 15 can communicate with other modules of electronic device 10 via bus 13. It should be understood that, although not shown in the figures, in practical applications, other hardware and / or software modules can be used in conjunction with electronic device 10, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms.

[0335] (Computer-readable storage medium)

[0336] This application also provides a computer-readable storage medium, the specific embodiments of which are consistent with the embodiments and technical effects achieved in the above method embodiments, and some contents will not be repeated.

[0337] The computer-readable storage medium stores a computer program that, when executed by at least one processor, implements the steps of any of the above methods or the functions of any of the above electronic devices.

[0338] A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. In embodiments of this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. Computer-readable storage media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0339] Computer-readable storage media may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable storage medium may also be any computer-readable medium capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, or any suitable combination thereof. Program code for performing operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on a user computing device, partially on a user device, as a standalone software package, partially on a user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing devices can be connected to user computing devices via any type of network, including local area networks (LANs) or wide area networks (WANs), or they can be connected to external computing devices (e.g., via the Internet using an Internet service provider).

[0340] (Computer program products)

[0341] This application also provides a computer program product, the specific embodiments of which are consistent with the embodiments and technical effects achieved in the above method embodiments, and some contents will not be repeated.

[0342] This application provides a computer program product comprising a computer program that, when executed by at least one processor, implements the steps of any of the above-described methods or the functions of any of the above-described electronic devices.

[0343] See Figure 6 , Figure 6 This is a schematic diagram of the structure of a computer program product provided in an embodiment of this application.

[0344] The computer program product is used to implement the steps of any of the above methods or to implement the functions of any of the above electronic devices. The computer program product may employ a portable compact disc read-only memory (CD-ROM) and include program code, and may run on a terminal device, such as a personal computer. However, the computer program product of the present invention is not limited thereto, and the computer program product may employ any combination of one or more computer-readable media.

[0345] This application describes the invention from the perspectives of purpose, performance, progress, and novelty, and it meets the functional enhancement and use requirements emphasized by the Patent Law. The above description and drawings are merely preferred embodiments of this application and are not intended to limit this application. Therefore, all structures, devices, features, etc., that are similar to or identical to those of this application, i.e., all equivalent substitutions or modifications made in accordance with the scope of this patent application, shall fall within the scope of protection of this patent application.

Claims

1. A virtual service person interaction method, characterized by, The method includes: Receive an access request from a first terminal device, establish a communication connection between the first terminal device and a target server, wherein the target server is used to provide virtual service personnel interaction functions; The first terminal device is used to receive user input information, each of which is text information, voice information, or image information; Based on the input information, one service item is determined from multiple service items as the user's target service item; Obtain the first interactive video corresponding to the target service item from the question and answer database corresponding to the virtual service personnel, wherein the question and answer database includes first interactive videos corresponding to multiple service items; The first interactive video is played using the first terminal device. In the first interactive video, the virtual service personnel introduce the procedures and / or guidelines for handling the target service item. The step of determining one service item from multiple service items as the user's target service item based on the input information includes: Semantic information is extracted from the input information; the similarity between the semantic information and each service item is obtained, and the maximum similarity is taken as the highest similarity. When the highest similarity is not greater than the preset similarity, the first terminal device is used to play the second interactive video to guide the user to click the form button and fill in the form; When a click operation is received on the form button, the form is displayed using the first terminal device. The form is provided with a service item input box and / or a service item drop-down menu. Based on the form's completion results, the user's target service item is obtained.

2. The virtual service personnel interaction method according to claim 1, characterized in that, The step of determining one service item from multiple service items as the user's target service item based on the input information includes: Using the semantic extraction model corresponding to the input information, semantic information is extracted from the input information; When the highest similarity is greater than the preset similarity, the service item corresponding to the highest similarity is taken as the target service item.

3. The virtual service personnel interaction method according to claim 1, characterized in that, The process of obtaining the similarity between the semantic information and each of the service items includes: For each of the aforementioned service items, perform the following processing: Input the semantic information and the preset text corresponding to the service item into a similarity model to obtain the similarity between the semantic information and the preset text, which is used as the similarity between the semantic information and the service item; The training process of the similarity model includes: Obtain a training set, which includes multiple training data, each of which includes a sample semantic information, a sample text, and annotation data on the similarity between the sample semantic information and the sample text; For each of the training data, perform the following processing: The sample semantic information and sample text in the training data are input into a preset deep learning model to obtain the predicted similarity data between the sample semantic information and the sample text. The model parameters of the deep learning model are updated based on the sample semantic information and the predicted and labeled data of the similarity between the sample text and the sample semantic information. Check whether the preset training termination condition is met; if yes, use the trained deep learning model as the similarity model; if no, continue to train the deep learning model using the next training data.

4. The virtual service personnel interaction method according to claim 3, characterized in that, The method further includes: The first terminal device is used to receive the user's feedback information; Based on the feedback information, identify the user's emotion category; Based on the user's emotion category, determine whether the user is satisfied with the service; If not satisfied, increment the cumulative count by one; When the cumulative number of times exceeds the preset number of times, the cumulative number of times is cleared to zero, and a prompt message is sent to the second terminal device of the preset personnel corresponding to the target service item, so as to prompt the preset personnel to update the similarity model or update the preset text corresponding to the target service item.

5. The virtual service personnel interaction method according to claim 1, characterized in that, The method further includes: The first terminal device is used to display a plurality of second interactive videos related to the first interactive video, each of which is used to introduce the precautions, frequently asked questions or related service items of the target service item; When a click operation is received for one of the second interactive videos, the clicked second interactive video is played using the first terminal device.

6. The virtual service personnel interaction method according to claim 1, characterized in that, In the first interactive video, the virtual service personnel guide the user to click the document download button to download the certification document; The method further includes: When a click operation is received on the document download button, the certificate document corresponding to the document download button is downloaded to the first terminal device.

7. The virtual service personnel interaction method according to claim 6, characterized in that, When a click operation is received on the document download button, downloading the proof document corresponding to the document download button to the first terminal device includes: When a click operation is received on the document download button, the first terminal device displays the document download interface corresponding to the document download button, and the document download interface is provided with an identity recognition button. When a click operation is received on the identity recognition button, the first terminal device is used to obtain the user's identity information to be identified; Detect whether the identity information to be identified matches the user's authenticated identity information; If a match is found, the proof document is downloaded to the first terminal device.

8. The virtual service personnel interaction method according to claim 7, characterized in that, If a match is found, the proof document is downloaded to the first terminal device, including: If a match is found, the user information database corresponding to the target service item is queried based on the authenticated identity information to generate the proof document based on the query results. Download the proof document to the first terminal device.

9. The virtual service personnel interaction method according to claim 8, characterized in that, The method further includes: Based on the user's authenticated identity information and the interaction data between the user and the virtual service personnel, one or more recommendation policies corresponding to the user are obtained and displayed in list form on the first terminal device.

10. An electronic device, characterized in that, The electronic device includes a memory and at least one processor, the memory storing a computer program, and the at least one processor being configured to execute the computer program to perform the following steps: Receive an access request from a first terminal device, establish a communication connection between the first terminal device and a target server, wherein the target server is used to provide virtual service personnel interaction functions; The first terminal device is used to receive user input information, each of which is text information, voice information, or image information; Based on the input information, one service item is determined from multiple service items as the user's target service item; Obtain the first interactive video corresponding to the target service item from the question and answer database corresponding to the virtual service personnel, wherein the question and answer database includes first interactive videos corresponding to multiple service items; The first interactive video is played using the first terminal device. In the first interactive video, the virtual service personnel introduce the procedures and / or guidelines for handling the target service item. The step of determining one service item from multiple service items as the user's target service item based on the input information includes: Semantic information is extracted from the input information; the similarity between the semantic information and each service item is obtained, and the maximum similarity is taken as the highest similarity. When the highest similarity is not greater than the preset similarity, the first terminal device is used to play the second interactive video to guide the user to click the form button and fill in the form; When a click operation is received on the form button, the form is displayed using the first terminal device. The form is provided with a service item input box and / or a service item drop-down menu. Based on the form's completion results, the user's target service item is obtained.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by at least one processor, implements the steps of the method according to any one of claims 1-9 or the function of the electronic device according to claim 10.

12. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by at least one processor, implements the steps of the method of any one of claims 1-9 or the function of the electronic device of claim 10.

Images