Property intelligent customer service system based on AI large model

Abstract

The invention relates to the technical field of intelligent property customer service, in particular to an intelligent property customer service system based on an AI large model, which is used for a back-end server for interaction between front ends, each front end comprises an owner end, a customer service end and a management end, and the back-end server comprises an intelligent customer service module, a big data analysis module, a personalized service module and an information management module. According to the application, through the arrangement of the intelligent customer service module, the demands of the owners can be automatically identified and classified, the rapid response to the demands of the owners is realized, and the application range of the system is expanded by adopting a mode of combining manual customer service and intelligent customer service; through the arrangement of the big data analysis module and the utilization of big data analysis and artificial intelligence technologies, the work condition of customer service staff can be known more comprehensively and accurately, and more valuable decision support is provided for a management layer. By arranging the personalized service module, personalized service schemes can be provided for the owners according to historical appeal information and preferences of the owners, so that the satisfaction degree of the owners is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent property customer service technology, and in particular to a property intelligent customer service system based on an AI large model. Background Art

[0002] Traditional property management customer service relies primarily on manual phone answering and handling of property owners' requests. This model has the following problems: (1) Slow response: The number of customer service staff is limited, and during peak hours, the lines are often busy or the waiting time is too long; (2) Unstable service quality: The professional level and work attitude of customer service staff vary, making it difficult to ensure service quality; (3) Chaotic information management: Property owners' requests and feedback information are scattered in different records, making it difficult to effectively organize and analyze them.

[0003] Currently, some existing property management companies have begun to introduce property management software to improve customer service. These software generally have the following functions: (1) Centralized information management: The owner's personal information, complaint records, etc. are stored in a centralized manner, making it easier for customer service staff to query and process; (2) Automated processes: Through pre-set process templates, tasks are automatically assigned to relevant departments, reducing manual operations; (3) Data analysis: Statistical analysis of customer service data helps management understand service quality and service needs.

[0004] Although existing property management software has improved the efficiency of customer service work to a certain extent, it still has the following shortcomings: (1) Low level of intelligence: Most software can only achieve simple information recording and task allocation, and lack intelligent analysis and prediction functions; (2) Poor owner experience: The software's operating interface is not user-friendly, and owners may encounter operational difficulties during use; (3) Lack of personalized services: It is impossible to provide personalized service solutions based on the specific needs of owners.

[0005] To this end, this application provides a property intelligent customer service system based on AI big model. Summary of the Invention

[0006] Based on this, it is necessary to provide a property intelligent customer service system based on AI big model to address the above technical problems.

[0007] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: a property intelligent customer service system based on an AI big model, a back-end server for front-end interaction, the front-end comprising: an owner end, a customer service end, and a management end, the back-end server comprising: an intelligent customer service module, a big data analysis module, a personalized service module, and an information management module; the owner end is used to input the owner's demand information and send the demand information to the intelligent customer service module, conduct at least one round of interactive dialogue with the customer service end or the intelligent customer service module, and receive and view the demand response results and personalized service plans generated by manual answers or by the AI big model; the customer service end is used to receive demand task requests and owner intention information, conduct at least one round of manual interactive dialogue with the owner end, generate and store demand response results answered manually; the management end is used to view multi-dimensional customer service data, statistical analysis reports and analysis results through the management end interface, and adjust customer service strategies and service plans according to the analysis results; the intelligent customer service module is used to receive demand information, use voice recognition and natural language processing technology to identify the demand information, generate owner intention information, and determine whether the owner intention information belongs to the property service scope type, generate a demand task request of the corresponding type according to the property service scope type to which the owner intention information belongs, and adjust the customer service strategy and service plan according to the analysis results; Based on the complexity of the owner's intent information, the corresponding type of appeal task request and the owner's intent information are sent to the customer service end or the corresponding type of business sub-module. The business sub-module calls the AI big model, conducts at least one round of AI interactive dialogue with the owner end, generates and stores the appeal response results generated by the AI big model, receives a personalized service plan, and sends the appeal response results and personalized service plan generated by the AI big model to the owner end. The big data analysis module is used to collect multi-dimensional customer service data and owner data, generate analysis results based on data mining algorithms, interpret and analyze the analysis results, extract key information related to property management, generate statistical analysis reports, and send the multi-dimensional customer service data, statistical analysis reports, and analysis results to the management end. The owner data includes the owner's historical appeal information, historical appeal response results answered manually, and historical appeal response results generated by the AI big model. The personalized service module is used to combine the owner's historical appeal information and preferences to generate a personalized service plan and send it to the owner end. The information management module is used to encrypt and store owner data, adopt a multi-level access control mechanism to allow authorized personnel to access and quickly retrieve owner data, and publish privacy policy information including the scope of use and protection measures of owner data.

[0008] Optionally, the owner side includes: a data input module for inputting the owner's demand information through the owner side interface and sending the demand information to the intelligent customer service module, the data input module includes one or more of the client APP, mobile terminal, and property self-service terminal, and the type of the demand information includes voice information and / or text information and / or video information; a dialogue interaction module for conducting at least one round of interactive dialogue with the customer service side or the intelligent customer service module, and receiving and viewing the demand response results and personalized service plans generated by manual answers or by the AI big model.

[0009] Optionally, the client terminal includes: a data receiving module for receiving demand task requests and owner intention information for the owner's personalized needs or questions that the system cannot answer; a manual service module for responding to the demand task request, combining the owner's intention information, and conducting at least one round of manual interactive dialogue with the dialogue interaction module to manually answer the customer's personalized needs and specific needs, generate a demand response result answered manually and send it to the dialogue interaction module; a data storage module for storing the demand response result answered manually in the Hadoop distributed file system.

[0010] Optionally, the intelligent customer service module includes: an information receiving submodule for receiving demand information. If the demand information is of the voice information and / or video information type, the voice information contained in the voice information and / or video information is converted into computer-readable text information through a voice recognition engine, and a personalized service plan is received; a natural language processing submodule for identifying the demand information through natural language processing technology to obtain the owner's intention information; a task allocation submodule for determining whether the owner's intention information belongs to the property service scope type. If the owner's intention information belongs to the property service scope type, a demand task request of the corresponding type is generated, and according to the complexity of the owner's intention information, if the owner's intention information is for the owner's personalized needs or a question that the system cannot answer, the demand task request of the corresponding type and the owner's intention information are sent to the manual service module, otherwise the corresponding type The appeal task request and the owner's intention information are sent to the business sub-module of the corresponding type; the business sub-module is used to receive the appeal task request and the owner's intention information, conduct at least one round of AI interactive dialogue with the dialogue interaction module, call multiple AI large models to generate multiple initial appeal response results of text type, fuse the multiple initial appeal response results of text type to obtain text type appeal response results, and convert the text type appeal response results into voice type appeal response results through text-to-speech technology, and the voice type appeal response results are used as the appeal response results generated by the AI large model and sent to the dialogue interaction module, and send a personalized service plan; wherein, the type of the business sub-module corresponds to the property service scope type; the data storage sub-module is used to store the appeal response results generated by the AI large model in the Hadoop distributed file system.

[0011] Optionally, the types of property service scope include: consultation, problem guidance and confirmation, automatic problem reporting and repair promotion, emergency event appeasement and emergency linkage, one-click rescue, notification of the recipient of the problem reporting and repair, AI dialogue to gently remind the overdue bills or express gratitude for no overdue bills, AI automatic telephone collection of overdue bills and collection of opinions from overdue owners, AI-assisted office, AI exclusive training and AI service drawing.

[0012] Optionally, the management end includes: a visualization module, which is used to view multi-dimensional customer service data, statistical analysis reports and analysis results in a visual form through the management end interface; an application and feedback module, which is used to adjust customer service strategies and service plans based on the analysis results, and apply them in actual property management work, collect feedback information on the application effects, and send it to the big data analysis module.

[0013] Optionally, the big data analysis module includes: a data collection and integration submodule, which is used to call multi-dimensional customer service data through the property charging API, and integrate the called multi-dimensional customer service data to build a data analysis library and store it in the Hadoop distributed file system. The multi-dimensional customer service data includes property management data, owner property fee payment records, parking space usage data, payable fee details, owner satisfaction survey data, historical payment history data, community financial revenue and expenditure data, report and repair record data, address book information, announcement notification information and management details information; a data cleaning and preprocessing submodule, which is used to call owner data through the information management module, and clean and preprocess the owner data to obtain preprocessed owner data, and store it in the Hadoop distributed file system; the first model construction The construction and training sub-module is used to build multiple data mining models through the Hadoop platform, and use the multi-dimensional customer service data and pre-processed owner data in the data analysis library to train and optimize the multiple data mining models to obtain multiple trained data mining models; the result analysis sub-module is used to generate multiple analysis results based on the trained multiple data mining models, interpret and analyze the analysis results, extract key information related to property work, and generate statistical analysis reports. The key information related to property work includes common problems and potential risks, and sends multi-dimensional customer service data, statistical analysis reports and analysis results to the visualization module; the continuous optimization sub-module is used to receive feedback information on the application effect, and call the latest multi-dimensional customer service data and the latest owner data to continuously optimize the data mining model.

[0014] Optionally, the result analysis submodule also includes: an interpretation and analysis submodule, which is used to use NLP technology to extract keywords from the analysis results, obtain common problem types, keywords and their frequencies, and build and train a risk warning model, and predict potential risks based on the trained risk warning model.

[0015] Optionally, the personalized service module includes: a data collection submodule, which is used to call multi-dimensional customer service data related to the owner through the property charging API, and collect the owner's historical demand information and preferences; a second model construction and training submodule, which is used to build and train the owner portrait model and multiple recommendation algorithms through the Hadoop platform, generate owner behavior label results based on the trained owner portrait model, combine the recommendation algorithm, generate personalized service plans, and send them to the information receiving submodule.

[0016] Optionally, the information management module includes: a data encryption storage submodule, which is used to encrypt the owner data using data encryption technology and store the encrypted owner data in the Hadoop distributed file system; an access control submodule, which is used to adopt a multi-level access control mechanism to allow authorized personnel to access and quickly retrieve the encrypted owner data; and a privacy policy publishing submodule, which is used to publish privacy policy information including the scope of use and protection measures of the owner data.

[0017] The advantages and beneficial effects of the present invention are: 1. Improve response speed: It can automatically identify and classify the owner's demands, greatly improving the response speed; 2. Improve service quality: By using big data analysis and artificial intelligence technology, it can more comprehensively and accurately understand the situation of customer service work, provide more valuable decision support for management, and improve the professionalism and accuracy of services; 3. Optimize information management: By establishing a unified information management module, efficient storage and rapid retrieval of information can be achieved; 4. Provide personalized services: According to the owner's historical demand information and preferences, personalized service plans are provided to improve the owner's satisfaction. 5. Improve user experience: The system operation interface is friendly, and the owner is more convenient and quick during use; 6. Reduce the workload of manual customer service of the property: Using a full-process, automated intelligent customer service system to reduce manual customer service intervention and replace the customer service seat system can not only effectively reduce the workload of manual customer service of the property, realize personalized family services, and allow property customer service to go out of the office to serve the owner's family affairs, but also improve the quality of property services, owner satisfaction and service efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a structural diagram of an embodiment of a property intelligent customer service system based on an AI big model according to the present application.

[0019] Figure 2 It is a structural diagram of an embodiment of the owner side according to the present application.

[0020] Figure 3 It is a structural diagram of an embodiment of the client terminal according to the present application.

[0021] Figure 4 It is a structural diagram of an embodiment of the management terminal according to the present application.

[0022] Figure 5 This is a structural diagram of an embodiment of the intelligent customer service module according to the present application.

[0023] Figure 6 It is a structural diagram of an embodiment of a big data analysis module according to the present application.

[0024] Figure 7 It is a structural diagram of an embodiment of a personalized service module according to the present application.

[0025] Figure 8 It is a structural diagram of an embodiment of the information management module according to the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0027] Existing property management software has improved the efficiency of customer service to a certain extent, but it still has the following shortcomings: (1) Low level of intelligence: Most software can only achieve simple information recording and task allocation, and lack intelligent analysis and prediction functions; (2) Poor owner experience: The software's operating interface is not user-friendly, and owners may encounter operational difficulties during use; (3) Lack of personalized service: It is impossible to provide personalized service solutions based on the specific needs of owners.

[0028] Based on the above problems, this application proposes a property intelligent customer service system based on an AI big model. The system can automatically identify and classify the owner's demands by setting up an intelligent customer service module, and achieve a rapid response to the owner's demands; and based on the complexity of the owner's intention information, the corresponding type of demand task request and the owner's intention information are sent to the customer service end or the corresponding type of business sub-module for at least one round of manual interactive dialogue or AI interactive dialogue, and a combination of manual customer service and intelligent customer service is adopted to improve the applicability of this system. At the same time, during the AI interactive dialogue process, multiple AI big models are called to generate multiple initial demand response results, and the multiple initial demand response results are integrated to improve the accuracy and reliability of the intelligent reply, thereby ensuring the effectiveness of the intelligent reply; and by setting up a big data analysis module and utilizing big data analysis and artificial intelligence technology, the situation of customer service work can be understood more comprehensively and accurately, providing management with more valuable decision-making support and improving the professionalism and accuracy of the service; in addition, by setting up a personalized service module, it is possible to provide personalized service plans based on the owner's historical demand information and preferences, so as to improve the owner's satisfaction.

[0029] Reference Attachment Figure 1 , shows a structural diagram of an embodiment of a property intelligent customer service system based on an AI large model according to the present application. The system can be specifically applied to various electronic devices.

[0030] The present embodiment describes a property intelligent customer service system based on an AI big model, which is used for a front-end 100 to interact with a back-end server 200, wherein the front-end 100 includes: an owner end 101, a customer service end 102 and a management end 103, and the back-end server 200 includes: an intelligent customer service module 201, a big data analysis module 202, a personalized service module 203 and an information management module 204.

[0031] According to an embodiment of the present disclosure, the owner terminal 101 is used to input the owner's demand information and send the demand information to the intelligent customer service module 201, conduct at least one round of interactive dialogue with the customer service terminal 102 or the intelligent customer service module 201, and receive and view the demand response results and personalized service plans generated by manual answers or by the AI big model.

[0032] According to the embodiments of the present disclosure, Figure 2 The owner terminal 101 includes: a data input module 1011, which is used to input the owner's demand information through the owner terminal 101 interface and send the demand information to the intelligent customer service module 201. The data input module 1011 includes one or more of the client APP, mobile terminal, and property self-service terminal. The type of the demand information includes voice information and / or text information and / or video information; a dialogue interaction module 1012, which is used to conduct at least one round of interactive dialogue with the customer service terminal 102 or the intelligent customer service module 201, and receive and view the demand response results and personalized service plans generated by manual answers or by the AI big model.

[0033] According to the embodiment of the present disclosure, the data input module 1011 supports multiple access methods such as telephone, online customer service, client APP, etc., which is convenient for owners to make requests anytime and anywhere; at the same time, in addition to the customer service function, it can also be combined with the community management module to realize the management of community facilities, activities, etc.

[0034] According to the embodiments of the present disclosure, the client APP and the back-end server 200 can exchange information through the Volcano Engine real-time transmission network, the mobile terminal (such as a mobile phone / telephone) and the back-end server 200 can exchange information through a network framework constructed by the public switched telephone network (i.e. PSTN), the SIP2RTC gateway and the Volcano Engine real-time transmission network, and the property self-service terminal and the back-end server 200 can exchange information through WIFI.

[0035] According to an embodiment of the present disclosure, by setting up a dialogue interaction module 1012, if encountering personalized needs of the owner or questions that the system cannot answer, at least one round of manual interactive dialogue is conducted with the customer service end 102, and the response results and personalized service plans answered manually are received and viewed; otherwise, at least one round of AI interactive dialogue is conducted with the intelligent customer service module 201, and the response results and personalized service plans generated by the AI big model are received and viewed.

[0036] According to an embodiment of the present disclosure, the client terminal 102 is used to receive the appeal task request and the owner's intention information, conduct at least one round of manual interactive dialogue with the owner terminal 101, and generate and store the appeal response result answered manually.

[0037] According to the embodiments of the present disclosure, Figure 3 The client terminal 102 includes: a data receiving module 1021, which is used to receive demand task requests and owner intention information for the owner's personalized needs or questions that the system cannot answer; a manual service module 1022, which is used to respond to the demand task request, combine the owner's intention information, and conduct at least one round of manual interactive dialogue with the dialogue interaction module 1012 to manually answer the customer's personalized needs and specific needs, generate a demand response result answered by the manual answer, and send it to the dialogue interaction module 1012; a data storage module 1023, which is used to store the demand response result answered by the manual answer in the Hadoop distributed file system.

[0038] According to the embodiment of the present disclosure, by setting up a manual service module 1022 and combining it with the intelligent customer service module 201, and adopting a combination of manual customer service and intelligent customer service, the demand information for the owner's personalized needs or questions that the system cannot answer can be automatically transferred to the manual service module 1022 and processed by manual customer service to improve the scope of application of this system.

[0039] According to the embodiments of the present disclosure, the present application uses the Hadoop distributed file system (ie, HDFS) to store massive customer service data, supporting high concurrent access and efficient data retrieval functions.

[0040] According to an embodiment of the present disclosure, the management terminal 103 is used to view multi-dimensional customer service data, statistical analysis reports and analysis results through the management terminal 103 interface, and adjust customer service strategies and service plans based on the analysis results.

[0041] According to the embodiments of the present disclosure, Figure 4The management terminal 103 includes: a visualization module 1031, which is used to view multi-dimensional customer service data, statistical analysis reports and analysis results in a visual form through the management terminal 103 interface; an application and feedback module 1032, which is used to adjust customer service strategies and service plans based on the analysis results, and apply them in actual property management work, collect feedback information on the application effect, and send it to the big data analysis module 202.

[0042] According to an embodiment of the present disclosure, a visualization module 1031 is set up so that multi-dimensional customer service data, statistical analysis reports and analysis results can be displayed and viewed in the form of charts, reports, etc. through the management terminal 103 interface, which facilitates property management personnel to make decisions.

[0043] According to an embodiment of the present disclosure, an application and feedback module 1032 is set up to apply the adjusted customer service strategy and service plan to actual property management work, and then combined with the intelligent customer service module 201 for dialogue communication, making it more vivid and using AI customer service as a communication medium to provide owners with customized and personalized services.

[0044] According to an embodiment of the present disclosure, the intelligent customer service module 201 is used to receive appeal information, use voice recognition and natural language processing technology to identify the appeal information, generate owner intention information, and determine whether the owner intention information belongs to the property service scope type. According to the property service scope type to which the owner intention information belongs, a corresponding type of appeal task request is generated, and according to the complexity of the owner intention information, the corresponding type of appeal task request and the owner intention information are sent to the customer service terminal 102 or the corresponding type of business sub-module, and the AI big model is called through the business sub-module to conduct at least one round of AI interactive dialogue with the owner terminal 101, generate and store the appeal response results generated by the AI big model, and receive personalized service plans, and send the appeal response results and personalized service plans generated by the AI big model to the owner terminal 101.

[0045] According to the embodiments of the present disclosure, Figure 5The intelligent customer service module 201 includes: an information receiving submodule 2011, which is used to receive the appeal information. If the appeal information is of the voice information and / or video information type, the voice information contained in the voice information and / or video information is converted into computer-readable text information through the voice recognition engine, and a personalized service plan is received; a natural language processing submodule 2012, which is used to identify the appeal information through natural language processing technology to obtain the owner's intention information; a task allocation submodule 2013, which is used to determine whether the owner's intention information belongs to the property service scope type. If the owner's intention information belongs to the property service scope type, a corresponding type of appeal task request is generated, and according to the complexity of the owner's intention information, if the owner's intention information is for the owner's personalized needs or a question that the system cannot answer, the corresponding type of appeal task request and the owner's intention information are sent to the manual service module 1022, otherwise the corresponding type of appeal task request is sent to the manual service module 1022. The task request and the owner's intention information are sent to the business sub-module of the corresponding type; the business sub-module 2014 is used to receive the demand task request and the owner's intention information, conduct at least one round of AI interactive dialogue with the dialogue interaction module 1012, call multiple AI large models to generate multiple initial demand response results of text type, fuse the multiple initial demand response results of text type to obtain the demand response result of text type, and convert the demand response result of text type into the demand response result of voice type through text-to-speech technology, and the demand response result of voice type is used as the demand response result generated by the AI large model, and sent to the dialogue interaction module 1012, and send a personalized service plan; wherein, the type of the business sub-module 2014 corresponds to the property service scope type; the data storage sub-module 2015 is used to store the demand response result generated by the AI large model in the Hadoop distributed file system.

[0046] According to an embodiment of the present disclosure, an information receiving submodule 2011 is set to receive appeal information. If the type of the appeal information is voice information and / or video information, the voice information contained in the voice information and / or video information is converted into computer-readable text information through a voice recognition engine (such as ASR) so that it can be subsequently processed in natural language.

[0047] According to an embodiment of the present disclosure, the appeal information is identified through the natural language processing sub-module 2012 to generate the owner's intention information, and it is judged and classified according to the preset property service scope type. If the owner's intention information belongs to the property service scope type, a corresponding type of appeal task request is generated. If the owner's intention information does not belong to the property service scope type, the owner can be automatically replied, such as: "This issue does not fall within the scope of property services. Is there any other service I can provide for you?"

[0048] According to an embodiment of the present disclosure, by setting a task assignment submodule 2013, tasks are automatically assigned to the corresponding business submodule 2014 or the data receiving module 1021 of the customer service end 102 according to the preset property service scope type, combined with the demand task request and the owner's intention information, and processed by the corresponding business submodule 2014 or customer service personnel (or customer service department) to improve work efficiency.

[0049] According to the embodiment of the present disclosure, by setting a business submodule 2014, whose type corresponds to the property service scope type, it is used to respond to the demand task request corresponding to the property service scope type, conduct at least one round of interactive dialogue in combination with the owner's intention information, and call multiple AI big models. The AI big models include but are not limited to the DeepSeek big model focusing on the dialogue context, the Hunyuan big model focusing on the dialogue performance, the Wenxin Yiyan big model focusing on the dialogue reasoning, and the Tongyi Qianxiang big model based on the dialogue knowledge base. Based on the aforementioned multiple AI big models, multiple initial demand response results of text types are generated, that is, the initial demand focusing on the dialogue context. Response results, initial demand response results focusing on dialogue performance, initial demand response results focusing on dialogue reasoning, and initial demand response results based on dialogue knowledge base, these initial demand response results are integrated (a multimodal large model can be used) to obtain text-type demand response results, and the text-type demand response results are converted into voice-type demand response results through text-to-speech technology (such as TTS), and the voice-type demand response results are used as the demand response results generated by the AI large model, and the demand response results generated by each round of interactive dialogue are sent to the dialogue interaction module 1012, as well as a personalized service plan.

[0050] According to the embodiments of the present disclosure, the types of property service scope include: consultation, problem guidance and confirmation, automatic problem reporting and repair promotion, emergency event appeasement and emergency linkage, one-click rescue, notification of the collector of problem reporting and repair, AI dialogue to gently remind customers of overdue bills or express gratitude for not having overdue bills, AI automatic telephone collection of overdue bills and collection of opinions from owners with overdue bills, AI-assisted office, AI exclusive training and AI service drawing.

[0051] According to the embodiments of the present disclosure, for consultation types within the scope of property services, for example, when a property company applies the property intelligent customer service system based on the AI big model provided by this application on a project basis, the owner dials the phone button through the client APP to enter the scene of communication and dialogue with the AI. After entering the scene of the dialogue, the owner input module can first initiate a concise self-introduction "Hello! Property customer service of a certain community, happy to serve you?", and the owner then speaks the content. According to the consultation questions input by the business, the owner's intention information is generated, and based on the consultation type within the scope of property services provided by the owner's intention information, a consultation type appeal task request is generated. In response to the consultation type appeal task request and the owner's intention information, the AI big model is called to generate a appeal response result for answering.

[0052] According to the embodiments of the present disclosure, for the problem guidance confirmation type within the scope of property services, the problem guidance confirmation type is mainly used for the automatic guidance confirmation type, which is used to automatically guide the owner to confirm the problem of the event within the scope of property services at home. For example, when the property company applies the property intelligent customer service system based on the AI large model provided by this application on a project basis, the owner dials the phone button through the client APP to enter the scene of the communication dialogue with A1. After entering the scene of the dialogue, the owner input module can first initiate a concise self-introduction "Hello! Property customer service of a certain community, I am happy to serve you?", and then the owner says the content. According to the content described by the owner, the owner is guided to check and confirm one by one step by step, and finally the problem is found. Through human-computer interaction recognition and analysis, it is determined what type of work of the property company of this project needs manual intervention, and the single-person collection reporting and repair process is automatically promoted.

[0053] According to the embodiments of the present disclosure, for problems within the scope of property services, the automatic reporting and repair promotion type of the problem is mainly used to automatically record reports and repairs and complete automatic dispatching. For example, when the property company applies the property intelligent customer service system based on the AI large model provided by this application on a project basis, the owner dials the phone button through the client APP to enter the scene of the communication dialogue with A1. After entering the dialogue scene, the owner input module can first initiate a concise self-introduction "Hello! Property customer service of a certain community, I am happy to serve you?", and the owner then speaks the content, and through human-computer interaction recognition and analysis, it is determined what type of work of the property company of this project requires manual intervention, and the reporting and repair process of a single person collecting the order is automatically promoted.

[0054] According to the embodiments of the present disclosure, for the emergency event appeasement and emergency linkage type within the property service scope, the emergency event appeasement and emergency linkage type is mainly used to automatically identify emergency events within the property service scope, automatically complete the reporting and repair process of multiple people in this property project collecting orders at the same time, realize the synchronous linkage of multiple people in multiple work categories of this project, and continuously guide the owners to avoid risks and appease their emotions according to the category and nature of emergency events within the property service scope; for example, when the property company applies the property intelligent customer service system based on the AI big model provided by this application on a project basis, the owner enters the scene of the communication dialogue with A1 by dialing the phone button on the client APP. After entering the scene of the dialogue, the owner can first initiate a concise self-introduction by inputting the module "Hello! Property customer service of a certain community, I am happy to serve you?", and then the owner speaks the content, and the AI big model content recognition and analysis shows that it belongs to an emergency event, and then the reporting and repair process of multiple people in this property project collecting orders at the same time is forcibly initiated automatically, realizing the synchronous linkage of multiple people in multiple work categories of this project, and automatically and continuously guiding the owners to avoid risks and appease their emotions.

[0055] According to the embodiments of the present disclosure, for the one-click help type within the scope of property services, a one-click help is used to initiate a reporting and repair process in which multiple people in this property project collect orders at the same time, thereby realizing the synchronous linkage of multiple people in multiple work categories of this property project, and automatically switching to the call to forcefully enter the call state to the owner. For example, when the property company applies the property intelligent customer service system based on the AI large model provided by this application on a project basis, the owner directly triggers and initiates the reporting and repair process in which multiple people in this property project collect orders at the same time through the orange SOS help button that can be moved up and down on the side of the client APP homepage, thereby realizing the synchronous linkage of multiple people in multiple work categories of this property project, and forcibly switching to the human-computer interaction interface, and automatically entering the call scene to the owner. The content of the call can be solidified as "Hello! Property customer service of a certain community, we are happy to serve you? Mr. / Ms., are you there? Please answer me!" Repeated until the owner speaks.

[0056] According to the embodiments of the present disclosure, for the notification type of the recipient of incident reports and repairs within the scope of property services, the notification type of the recipient of incident reports and repairs is mainly used to automatically complete the incident report and repair dispatch process, and automatically notify the recipient by phone that "you have a (normal event / emergency event) dispatch, please handle it in time or contact the owner"; for example, in the case where the property company applies the property intelligent customer service system based on the AI large model provided by this application on a project basis, the property company configures whether the dispatch requires telephone notification according to (normal event / emergency event) in the property system on a project basis. When there is traffic and the configuration requires telephone notification, the AI automatically notifies the recipient by phone that "you have a (normal event / emergency event) dispatch, please handle it in time or contact the owner".

[0057] According to the embodiments of the present disclosure, for AI conversations within the scope of property services, which subtly remind owners of overdue fees or convey gratitude for those who do not owe fees, at the end of at least one round of interactive conversations, when it is identified that the owner of this conversation has overdue fees, the owner will be subtly reminded that "the settlement of your property fees will be the top incentive for our service, please, please!"; when it is identified that the owner of this conversation has no overdue fees, gratitude will be conveyed with "Thank you for your support, we will continue to work hard to provide you with better services! Thank you!".

[0058] According to the embodiments of the present disclosure, for the AI automatic telephone collection of overdue fees and the collection of opinions from owners of overdue fees within the scope of property services, the AI automatic telephone collection of overdue fees and the collection of opinions from owners of overdue fees are mainly used to automatically complete the telephone collection of overdue fees and the collection of opinions from owners. For example, when the property company applies the AI large model-based property intelligent customer service system provided by this application on a project basis, the property company configures the monthly collection start date and daily collection time period in the system on a project basis. During the configured date and time period, the company automatically collects overdue fees and collects opinions from owners of overdue fees by phone, and the telephone charges and collected information need to be recorded in the system.

[0059] According to the embodiments of the present disclosure, for the AI-assisted office type within the scope of property services, the AI-assisted office type can assist in the drafting of office documents and the query of legal applications in the property industry according to the needs of property company employees. For example, when a property company applies the property intelligent customer service system based on the AI large model provided by this application on a project basis, a text entry window is provided to the property company employees, and the employees input the requirements for drafting office documents in the property industry or descriptions of corresponding events in the industry. The document content is then automatically generated or the relevant applicable laws and regulations are listed, and the generated text can be copied by employees.

[0060] According to the embodiments of the present disclosure, for the AI exclusive training type within the scope of property services, the AI exclusive training type can provide uploading of exclusive knowledge of the property company and each project, so as to achieve more accurate replies or services to the owners of the corresponding projects under the property company in the capacity of a property company customer service, and achieve more specialized, more precise and more efficient professionalism. For example, when the property company applies the property intelligent customer service system based on the AI large model provided by this application on a project basis, a window for uploading knowledge is provided to the property company, and the uploaded knowledge content is used as the first search of the property company's project knowledge, and also as an increment to the platform company's knowledge base.

[0061] According to the embodiments of the present disclosure, for the AI service drawing types within the scope of property services, the AI service drawing class can automatically generate property service decision analysis diagrams on a project basis, so that the decision-makers of the property company can grasp the analysis data charts of employees, owners, economy, events, etc. of each project, so as to provide decision-making ideas and suggestions for management adjustments. For example, when the property company applies the property intelligent customer service system based on the AI big model provided by this application on a project basis, the decision analysis diagram on a project basis is generated according to set rules based on factors such as employee behavior, owner interaction records, project economic values, events, etc.

[0062] According to an embodiment of the present disclosure, the big data analysis module 202 is used to collect multi-dimensional customer service data and owner data, generate analysis results based on data mining algorithms, interpret and analyze the analysis results, extract key information related to property work, generate statistical analysis reports, and send the multi-dimensional customer service data, statistical analysis reports and analysis results to the management terminal 103; wherein, the owner data includes the owner's historical demand information, historical demand response results answered manually, and historical demand response results generated by the AI big model.

[0063] According to the embodiments of the present disclosure, Figure 6, the big data analysis module 202 includes: a data collection and integration submodule 2021, which is used to call multi-dimensional customer service data through the property charging API, and integrate the called multi-dimensional customer service data to build a data analysis library and store it in the Hadoop distributed file system. The multi-dimensional customer service data includes property management data, owner property fee payment records, parking space usage data, payable fee details, owner satisfaction survey data, historical payment history data, community financial revenue and expenditure data, report and repair record data, address book information, announcement notification information and management details information; a data cleaning and preprocessing submodule 2022, which is used to call owner data through the information management module 204, and clean and preprocess the owner data to obtain preprocessed owner data, and store it in the Hadoop distributed file system; first model construction and training The training submodule 2023 is used to build a variety of data mining models through the Hadoop platform, and use the multi-dimensional customer service data and pre-processed owner data in the data analysis library to train and optimize the various data mining models to obtain the trained multiple data mining models; the result analysis submodule 2024 is used to generate a variety of analysis results based on the trained multiple data mining models, and interpret and analyze the analysis results, extract key information related to property work, and generate a statistical analysis report. The key information related to property work includes common problems and potential risks, and sends the multi-dimensional customer service data, statistical analysis report and analysis results to the visualization module 1031; the continuous optimization submodule 2025 is used to receive feedback information on the application effect, and call the latest multi-dimensional customer service data and the latest owner data to continuously optimize the data mining model.

[0064] According to an embodiment of the present disclosure, the result analysis submodule 2024 also includes: an interpretation and analysis submodule, which is used to use NLP technology to extract keywords from the analysis results, obtain common problem types, keywords and their frequencies, and build and train a risk warning model, and predict potential risks based on the trained risk warning model.

[0065] According to an embodiment of the present disclosure, by setting up a big data analysis module 202, multi-dimensional customer service data and owner data can be collected, and the owner data can be pre-processed, and a variety of data mining models can be constructed and trained. Based on the trained multiple data mining models, a variety of analysis results can be generated, and the analysis results can be interpreted and analyzed, and key information related to property work can be extracted and displayed in a visual form.

[0066] According to the embodiments of the present disclosure, this application uses data mining technology to analyze multi-dimensional data sources, including property management data, owner property fee payment records, parking space usage data, payable fee details, owner satisfaction survey data, historical payment history, community financial revenue and expenditure data, incident and repair report records, contact information, announcements, and management details, to extract information that is extremely valuable for property management. This information can play a key role in owner property fee payment, significantly improving property management efficiency and owner satisfaction.

[0067] Specifically, by analyzing owners' payment history, payment habits, payment capacity (inferred from parking space usage data, detailed payable fees, and other factors), as well as satisfaction survey data, it is possible to accurately identify the payment characteristics and potential needs of different owner groups. For example, it is possible to identify owners who are active in paying, those who need reminders, and those who may have difficulty paying. Based on these analysis results, the property management company can develop personalized payment reminder strategies. For example, it can reward owners who are active in paying with points, provide customized reminders (such as via text message, phone call, or home visit) to owners who need reminders, and offer flexible payment plans or installment options to owners who may have difficulty paying. Furthermore, by analyzing community revenue and expenditure data and incident reporting and repair data, the property management company can optimize resource allocation, improve service quality, and enhance owners' perception of the value of property management services, thereby promoting timely payment of property management fees. For example, allocating more resources to incident reporting and repair projects that owners frequently report can significantly improve owner satisfaction and, in turn, increase their willingness to pay. Ultimately, through the application of data mining technology, the property management company can achieve refined management of property fee payment, increase payment rates, reduce collection costs, and build a more harmonious and win-win relationship between the property management company and the owners.

[0068] According to the embodiments of the present disclosure, the present application deploys Hadoop, a storage and computing engine for massive data, as an open source distributed computing framework for building the back-end server 200. With its high reliability, high scalability and high efficiency, it has become an ideal choice for processing massive data. Among them, in terms of data storage, the Hadoop distributed file system (i.e., HDFS) can be used to store massive customer service data, including but not limited to property management data, property owner property fee payment records, parking space usage data, payable fee details, property owner satisfaction survey data, historical payment records, community financial revenue and expenditure data, report and repair records, address book information, announcements and notifications, and management details; HDFS, with its high fault tolerance and high throughput, can ensure the secure storage and fast access of data; in terms of data processing, Hadoop's MapReduce computing model can be used to perform large-scale parallel processing on the data stored in HDFS; MapReduce can decompose complex computing tasks into multiple simple subtasks and execute them in parallel on multiple nodes, greatly improving data processing efficiency; in terms of task scheduling, the Hadoop YARN resource management system can be responsible for the unified management and scheduling of cluster resources to ensure that each computing task can use cluster resources efficiently and fairly.

[0069] According to an embodiment of the present disclosure, a data cleaning and preprocessing submodule 2022 is set to call the owner data through the information management module 204, and clean and preprocess the owner data to ensure the accuracy, completeness and consistency of the data. The cleaning and preprocessing include but are not limited to removing duplicate data, filling missing values, correcting erroneous data, etc. to generate preprocessed owner data.

[0070] According to an embodiment of the present disclosure, by setting a first model construction and training submodule 2023, using the Hadoop platform, multiple data mining models are constructed, and the multiple data mining models are trained and optimized using the multi-dimensional customer service data and pre-processed owner data in the data analysis library to obtain multiple trained data mining models to ensure the accuracy and generalization ability of the model, wherein the data mining models include but are not limited to owner payment prediction models, owner satisfaction prediction models, owner behavior analysis models, Apriori algorithms, K-means algorithms, etc.

[0071] According to the embodiments of the present disclosure, the Apriori algorithm is mainly used for association rule mining to mine owner behavior patterns. Its principle is: Apriori algorithm, as a classic association rule mining algorithm, is used to discover frequent item sets and association rules in data sets; by analyzing the owner's payment records, incident reporting and repair records, satisfaction survey data, etc., the Apriori algorithm can be used to discover potential correlations between owner behaviors; for example, analysis shows that among the group of owners who frequently report incidents and repairs, the proportion of property fees paid on time is low. This prompt indicates that the response speed and quality of incident reporting and repairs need to be further improved; in specific applications, the correlation between the owner's payment behavior and other behaviors (such as incident reporting, repair reporting, complaints and suggestions) is identified to help the property discover potential problems, optimize service processes, and improve owner satisfaction; for example, through association analysis, it is found that the owner group who frequently participates in community activities has a higher payment enthusiasm. This prompt indicates that the owner's sense of belonging and willingness to pay can be enhanced by organizing more community activities.

[0072] According to the embodiments of the present disclosure, the K-means algorithm is mainly used for cluster analysis and segmentation of owner groups. The principle is as follows: the K-means algorithm is a distance-based clustering algorithm that divides data points into K clusters, so that the similarity of data points within the same cluster is high, while the similarity of data points between different clusters is low; by setting a suitable K value, owners can be divided into different groups, such as those who are actively paying, those who are occasionally late paying, and those who are long-term arrears, etc.; in specific applications, owners are grouped in a refined manner, and personalized payment reminder strategies and communication plans are formulated for different groups. For example, for owners who are actively paying, more points rewards and honorary titles can be given to encourage them to continue to maintain their payment; for owners who are occasionally late paying, friendly reminders can be made through text messages, phone calls, etc.; for owners who are long-term arrears, in-depth communication is required to understand the reasons for their arrears and provide corresponding solutions.

[0073] According to an embodiment of the present disclosure, by setting a result analysis submodule 2024, the analysis results of the model are interpreted and analyzed to extract valuable information, wherein NLP technology can be used to extract keywords from the analysis results to obtain common problem types, keywords and their frequencies, for example, common problem types such as call duration, connection rate, customer satisfaction, and keywords such as customer complaints, consultations, and repairs and their frequencies, and the trained risk warning model can be used to predict and identify potential risks, for example, long service response time leads to decreased customer satisfaction; frequent occurrence of a certain type of problem may lead to collective complaints; high equipment failure rate affects the quality of property services, etc.

[0074] According to the embodiments of the present disclosure, service quality indicators such as response time and resolution rate can be displayed through statistical analysis reports, and by analyzing customer needs such as complaints, consultations, repairs and other hot issues, a decision-making basis can be provided to the management.

[0075] According to an embodiment of the present disclosure, by setting a continuous optimization submodule 2025, based on the collected feedback information on the application effect, the latest multi-dimensional customer service data and the latest owner data are continuously called to continuously optimize the data mining model.

[0076] According to an embodiment of the present disclosure, the personalized service module 203 is used to combine the owner's historical demand information and preferences to generate a personalized service plan and send it to the owner terminal 101.

[0077] According to the embodiments of the present disclosure, Figure 7 The personalized service module 203 includes: a data collection submodule 2031, which is used to call multi-dimensional customer service data related to the owner through the property charging API, and collect the owner's historical demand information and preferences; a second model construction and training submodule 2032, which is used to build and train the owner portrait model and multiple recommendation algorithms through the Hadoop platform, generate the owner behavior label result based on the trained owner portrait model, combine the recommendation algorithm, generate a personalized service plan, and send it to the information receiving submodule 2011.

[0078] According to the embodiments of the present disclosure, the specific implementation process of generating and providing personalized service plans includes: collecting historical data of owners, including but not limited to repair records, complaints and suggestions, payment status, etc., analyzing owner preferences, such as service type, service time, communication method, etc., labeling the owners based on the trained owner portrait model, generating owner behavior label results, using recommendation algorithms such as collaborative filtering and matrix decomposition to generate personalized service plans, recommending suitable services to owners, and collecting feedback in real time to continuously optimize the plans.

[0079] According to an embodiment of the present disclosure, the information management module 204 is used to encrypt and store owner data, adopt a multi-level access control mechanism, allow authorized personnel to access and quickly retrieve owner data, and publish privacy policy information including the scope of use and protection measures of owner data.

[0080] According to the embodiments of the present disclosure, Figure 8The information management module 204 includes: a data encryption storage submodule 2041, which is used to encrypt the owner data using data encryption technology and store the encrypted owner data in the Hadoop distributed file system; an access control submodule 2042, which is used to adopt a multi-level access control mechanism to allow authorized personnel to access and quickly retrieve the encrypted owner data; and a privacy policy publishing submodule 2043, which is used to publish privacy policy information including the scope of use and protection measures of the owner data.

[0081] According to an embodiment of the present disclosure, by providing the data encryption storage submodule 2041 , owner data can be encrypted and stored to ensure data security.

[0082] According to an embodiment of the present disclosure, by setting the access control submodule 2042 and adopting a multi-level access control mechanism, it is ensured that only authorized personnel can access sensitive data.

[0083] According to an embodiment of the present disclosure, a privacy policy publishing submodule 2043 is provided to publish privacy policy information, clearly inform the owner of the scope of use and protection measures of the data, and ensure the owner's privacy rights and interests.

[0084] According to an embodiment of the present disclosure, any multiple modules of the owner-side 101, the customer service side 102, the management side 103, the intelligent customer service module 201, the big data analysis module 202, the personalized service module 203, and the information management module 204 can be combined into a single module, or any one of these modules can be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules can be combined with at least part of the functionality of other modules and implemented in a single module. According to an embodiment of the present disclosure, at least one of the owner-side 101, the customer service side 102, the management side 103, the intelligent customer service module 201, the big data analysis module 202, the personalized service module 203, and the information management module 204 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented in hardware or firmware by any other reasonable means of integrating or packaging circuits, or can be implemented in any one of the three implementation methods of software, hardware, and firmware, or in any appropriate combination of any of them. Alternatively, at least one of the owner end 101, the customer service end 102, the management end 103, the intelligent customer service module 201, the big data analysis module 202, the personalized service module 203 and the information management module 204 can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0085] The electronic device according to an embodiment of the present application includes a processor, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage portion into a random access memory (RAM). The processor may, for example, include a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (such as an application-specific integrated circuit (ASIC)), etc. The processor may also include an onboard memory for caching purposes. The processor may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present application.

[0086] Various programs and data required for the operation of the electronic device are stored in the RAM. The processor, ROM, and RAM are connected to each other via a bus. The processor performs various operations of the method flow according to the embodiment of the present application by executing the programs in the ROM and / or RAM. It should be noted that the program can also be stored in one or more memories other than the ROM and RAM. The processor can also perform various operations of the method flow according to the embodiment of the present application by executing the programs stored in the one or more memories.

[0087] According to an embodiment of the present application, the electronic device may further include an input / output (I / O) interface, which is also connected to the bus. The electronic device may also include one or more of the following components connected to the I / O interface: an input portion including a keyboard, a mouse, etc.; an output portion including a cathode ray tube (CRT), a liquid crystal display (LCD), a speaker, etc.; a storage portion including a hard disk, etc.; and a communication portion including a network interface card such as a LAN card, a modem, etc. The communication portion performs communication processing via a network such as the Internet. The drive is also connected to the I / O interface as needed. Removable media, such as magnetic disks, optical disks, magneto-optical disks, semiconductor memories, etc., are installed on the drive as needed so that the computer program read therefrom is installed into the storage portion as needed.

[0088] This application also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments, or may exist independently and not be incorporated into the device / apparatus / system. The computer-readable storage medium carries one or more programs, and when the one or more programs are executed, the method according to the embodiments of this application is implemented.

[0089] According to an embodiment of the present application, a computer-readable storage medium may be a non-volatile computer-readable storage medium, such as but not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present application, a computer-readable storage medium may include the ROM and / or RAM described above and / or one or more memories other than ROM and RAM.

[0090] The present application also includes a computer program product comprising a computer program containing program code for executing the method shown in the flowchart. When the computer program product is executed in a computer system, the program code is used to cause the computer system to implement the item recommendation method provided in the present application.

[0091] When the computer program is executed by the processor, the above functions defined in the system / device of the embodiment of the present application are performed. According to the embodiment of the present application, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0092] In one embodiment, the computer program may be stored on a tangible storage medium, such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may be transmitted and distributed in the form of a signal over a network medium, downloaded and installed via a communication component, and / or installed from a removable medium. The program code contained in the computer program may be transmitted using any suitable network medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0093] In such an embodiment, the computer program can be downloaded and installed from a network via the communication portion, and / or installed from a removable medium. When the computer program is executed by the processor, the above-mentioned functions defined in the system of the embodiment of the present application are performed. According to the embodiment of the present application, the systems, devices, means, modules, units, etc. described above can be implemented by computer program modules.

[0094] According to an embodiment of the present application, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages, specifically, these computing programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages include, but are not limited to, languages such as Java, C++, python, "C" or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect via the Internet).

[0095] Obviously, the embodiments described above are only some of the embodiments of the present application, rather than all of the embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions described in the aforementioned specific embodiments, or to make equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the present application specification and the accompanying drawings, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present application.

Claims

1. A property intelligent customer service system based on AI big model, used for front-end and back-end server interaction, characterized in that: The front end includes: an owner end, a customer service end, and a management end, and the back end server includes: an intelligent customer service module, a big data analysis module, a personalized service module, and an information management module; The owner terminal is used to input the owner's request information and send the request information to the intelligent customer service module, conduct at least one round of interactive dialogue with the customer service terminal or the intelligent customer service module, and receive and view the request response results and personalized service plans generated by manual answers or by the AI large model; The client is used to receive the demand task request and the owner's intention information, conduct at least one round of manual interactive dialogue with the owner, and generate and store the demand response result answered manually; The management terminal is used to view multi-dimensional customer service data, statistical analysis reports and analysis results through the management terminal interface, and adjust customer service strategies and service plans based on the analysis results; The intelligent customer service module is used to receive demand information, identify the demand information using voice recognition and natural language processing technology, generate owner intention information, and determine whether the owner intention information belongs to the property service scope type. According to the property service scope type to which the owner intention information belongs, a demand task request of a corresponding type is generated, and according to the complexity of the owner intention information, the demand task request of the corresponding type and the owner intention information are sent to the customer service end or the business sub-module of the corresponding type. The AI big model is called through the business sub-module to conduct at least one round of AI interactive dialogue with the owner end, generate and store the demand response results generated by the AI big model, and receive personalized service plans, and send the demand response results and personalized service plans generated by the AI big model to the owner end; The big data analysis module is used to collect multi-dimensional customer service data and owner data, generate analysis results based on data mining algorithms, interpret and analyze the analysis results, extract key information related to property management, generate statistical analysis reports, and send the multi-dimensional customer service data, statistical analysis reports, and analysis results to the management end; wherein, the owner data includes the owner's historical request information, historical request response results generated by manual answers, and historical request response results generated by the AI large model; The personalized service module is used to combine the owner's historical demand information and preferences to generate a personalized service plan and send it to the owner's end; The information management module is used to encrypt and store owner data, adopt a multi-level access control mechanism, allow authorized personnel to access and quickly retrieve owner data, and publish privacy policy information including the scope of use and protection measures of owner data.

2. The property intelligent customer service system based on AI big model as claimed in claim 1, characterized in that: The owner side includes: A data input module is used to input the owner's demand information through the owner-side interface and send the demand information to the intelligent customer service module. The data input module includes one or more of a client APP, a mobile terminal, and a property self-service terminal. The types of the demand information include voice information, text information, and / or video information; The dialogue interaction module is used to conduct at least one round of interactive dialogue with the customer service end or the intelligent customer service module, and to receive and view the response results and personalized service plans generated by manual answers or by the AI big model.

3. The property intelligent customer service system based on AI big model as claimed in claim 2, characterized in that: The client terminal includes: The data receiving module is used to receive appeal task requests and owner intention information targeting the owner's personalized needs or questions that the system cannot answer; The manual service module is used to respond to the demand task request, combine the owner's intention information, and conduct at least one round of manual interaction with the dialogue interaction module to manually answer the customer's personalized needs and specific needs, generate a manual response result, and send it to the dialogue interaction module; The data storage module is used to store the response results of the manually answered questions in the Hadoop distributed file system.

4. The property intelligent customer service system based on AI big model as claimed in claim 3, characterized in that: The intelligent customer service module includes: An information receiving submodule is configured to receive a request message and, if the request message is of the voice message and / or video message type, convert the voice information contained in the voice message and / or video message into computer-readable text information through a voice recognition engine, and receive a personalized service plan; The natural language processing submodule is used to identify the appeal information through natural language processing technology to obtain the owner's intention information; The task allocation submodule is used to determine whether the owner's intention information falls within the scope of property services. If the owner's intention information falls within the scope of property services, a demand task request of the corresponding type is generated. Based on the complexity of the owner's intention information, if the owner's intention information is for the owner's personalized needs or a question that the system cannot answer, the demand task request of the corresponding type and the owner's intention information are sent to the manual service module; otherwise, the demand task request of the corresponding type and the owner's intention information are sent to the business submodule of the corresponding type. The business submodule is used to receive the appeal task request and the owner's intention information, conduct at least one round of AI interactive dialogue with the dialogue interaction module, call multiple AI large models to generate multiple initial appeal response results of text type, fuse the multiple initial appeal response results of text type to obtain the appeal response result of text type, and convert the text type appeal response result into the voice type appeal response result through text-to-speech technology. The voice type appeal response result is used as the appeal response result generated by the AI large model and sent to the dialogue interaction module, and a personalized service plan is sent; wherein, the type of the business submodule corresponds to the type of property service scope; The data storage submodule is used to store the demand response results generated by the AI large model in the Hadoop distributed file system.

5. The property intelligent customer service system based on AI big model as claimed in claim 4, characterized in that: The types of property services include: consultation, problem guidance and confirmation, automatic problem reporting and repair promotion, emergency event appeasement and emergency linkage, one-click rescue, notification of the recipient of the problem reporting and repair, AI dialogue to gently remind customers of overdue bills or express gratitude for not having overdue bills, AI automatic telephone collection of overdue bills and collection of opinions from owners with overdue bills, AI-assisted office work, AI-exclusive training and AI service drawing.

6. The property intelligent customer service system based on AI big model according to claim 1, characterized in that: The management terminal includes: The visualization module is used to view multi-dimensional customer service data, statistical analysis reports and analysis results in a visual form through the management interface; The application and feedback module is used to adjust customer service strategies and service plans based on the analysis results, and apply them in actual property management work, collect feedback information on application effects, and send it to the big data analysis module.

7. The property intelligent customer service system based on AI big model according to claim 6, characterized in that: The big data analysis module includes: The data collection and integration submodule is used to call multi-dimensional customer service data through the property fee collection API, integrate the called multi-dimensional customer service data, build a data analysis library, and store it in the Hadoop distributed file system. The multi-dimensional customer service data includes property management data, owner property fee payment records, parking space usage data, payable fee details, owner satisfaction survey data, historical payment history data, community financial revenue and expenditure data, incident and repair report data, address book information, announcement and notification information, and management details; The data cleaning and preprocessing submodule is used to call the owner data through the information management module, clean and preprocess the owner data, obtain preprocessed owner data, and store it in the Hadoop distributed file system; The first model building and training submodule is used to build multiple data mining models through the Hadoop platform, and use the multi-dimensional customer service data and pre-processed owner data in the data analysis library to train and optimize the multiple data mining models to obtain multiple trained data mining models; The result analysis submodule is used to generate multiple analysis results based on the trained multiple data mining models, interpret and analyze the analysis results, extract key information related to property management work, generate statistical analysis reports, and send multi-dimensional customer service data, statistical analysis reports and analysis results to the visualization module; The continuous optimization sub-module is used to receive feedback on application effects, call the latest multi-dimensional customer service data and the latest owner data, and continuously optimize the data mining model.

8. The property intelligent customer service system based on AI big model as claimed in claim 7, characterized in that: The result analysis submodule also includes: The interpretation and analysis submodule is used to extract keywords from the analysis results using NLP technology, obtain common problem types, keywords and their frequencies, build and train risk warning models, and predict potential risks based on the trained risk warning models.

9. The property intelligent customer service system based on AI big model as claimed in claim 4, characterized in that: The personalized service module includes: The data collection submodule is used to call multi-dimensional customer service data related to owners through the property fee API, and collect owners' historical demand information and preferences; The second model construction and training sub-module is used to build and train the owner portrait model and multiple recommendation algorithms through the Hadoop platform, generate owner behavior label results based on the trained owner portrait model, combine the recommendation algorithm to generate personalized service plans, and send them to the information receiving sub-module.

10. The property intelligent customer service system based on AI big model according to claim 1, characterized in that: The information management module includes: The data encryption storage submodule is used to encrypt the owner data using data encryption technology and store the encrypted owner data in the Hadoop distributed file system; Access control submodule, which is used to adopt a multi-level access control mechanism to allow authorized personnel to access and quickly retrieve encrypted owner data; The privacy policy publishing submodule is used to publish privacy policy information including the scope of use and protection measures of owner data.

Images