Intelligent listing system based on AI demand perception and accurate information reaching method

Abstract

The invention discloses an intelligent listing system based on AI demand perception and an information accurate reaching method, and relates to the technical field of multi-modal interaction. Comprising an intelligent listing terminal, a multi-mode interaction module, a behavior data acquisition module, an AI demand perception engine module, a data storage and analysis center, an information accurate reaching module and an information display and feedback module. Touch control, voice, face recognition, mobile phone linkage and other multi-mode interaction are integrated, different resident requirements are met, the threshold is lowered, and inclusiveness is improved; behavior data collection is combined with an association model and quantitative evaluation of an AI engine, resident demands are mined, information on-demand pushing is achieved through space, crowd, scene and behavior quadruple orientation, and redundancy is reduced. A real-time feedback closed-loop mechanism promotes dynamic iteration of the model and the strategy to form a service optimization virtuous cycle; the community operation efficiency is improved through data-driven management decision and intelligent appeal processing, and the resident trust is enhanced through privacy protection design.

Description

Technical Field

[0001] This invention relates to the field of multimodal interaction technology, specifically to an AI-based intelligent tagging system and a method for precise information delivery. Background Technology

[0002] With the rapid development of information technology, intelligent and digital community management systems have gradually become an inevitable trend in modern community construction. However, existing community service facilities rely on relatively simple methods for information disclosure and interaction, lacking the ability for real-time updates, two-way interaction, and precise services.

[0003] Community electronic signage, as an intelligent device integrating display, broadcasting, and monitoring functions, is increasingly favored by community managers and residents.

[0004] Therefore, developing a comprehensive and stable community electronic tagging device is not only in line with the development trend of modern community intelligence and digitalization, but also of great significance for improving residents' quality of life and enhancing community cohesion.

[0005] The present invention aims to overcome the shortcomings of the prior art and provide an intelligent listing system and a precise information delivery solution based on AI demand perception.

[0006] The system can dynamically sense residents' interests and hot topics, and realize interactive information query and feedback through various interaction methods (voice, touch, facial recognition, mobile phone linkage, etc.). Based on AI analysis results, it can realize personalized and accurate information push, effectively improving the intelligence level of community information services and residents' satisfaction. Summary of the Invention

[0007] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides an AI-based intelligent listing system and a method for precise information delivery, solving the problems mentioned in the background section.

[0008] (II) Technical Solution To achieve the above objectives, the present invention is implemented through the following technical solution: an AI-based intelligent tagging system, comprising an intelligent tagging terminal, a multimodal interaction module, a behavior data acquisition module, an AI demand perception engine module, a data storage and analysis center, an information precision delivery module, and an information display and feedback module; The intelligent tagging terminal is deployed in the public areas of the community; The multimodal interaction module is integrated into the intelligent tagging terminal to enable multi-mode interaction; The behavior data collection module collects residents' behavior data; The AI ​​demand perception engine module performs demand analysis based on data. The data storage and analysis center stores and processes data; The precise information delivery module enables targeted information push; The information display and feedback module presents information and receives feedback.

[0009] As a further aspect of the present invention: the intelligent tagging terminal is a vertical or wall-mounted device, including a touch interaction submodule, a voice interaction submodule, a face recognition interaction submodule and a mobile phone linkage interaction module; The touch interaction submodule responds to touch operations to display information or trigger functions, specifically: responding to click and swipe operations on the touch screen to display menus, information details or trigger functions; The voice interaction submodule collects voice commands through a microphone array and, after speech recognition and semantic understanding, feeds them back through a speaker. The facial recognition interaction submodule uses a camera to identify the user's identity with the user's authorization in order to provide personalized services. Specifically, under the premise of user authorization and privacy protection, the camera identifies the resident's identity, and then the user can view their personal information after identity authentication. The mobile phone linkage interaction module connects with residents' mobile terminals via QR code, NFC, or Bluetooth, and supports information scanning and downloading, remote control via mobile terminal, QR code evaluation and feedback, and receiving personalized push messages.

[0010] As a further aspect of the present invention: the multimodal interaction module supports voice input / output, touch screen operation, face recognition, NFC / RFID reading and writing, QR code generation and scanning, and supports intelligent service robot interaction, face recognition linkage and seamless mobile phone collaboration; The intelligent service robot is based on process configuration and large model technology to answer questions and guide residents to submit their requests. Specifically, users wake up the terminal intelligent service robot through the voice module. The intelligent service robot is based on process configuration and large model technology to answer questions and guide residents to submit their service requests through dialogue with the robot. The service requests include consultation, suggestions and service needs. The facial recognition linkage displays matching services based on resident profiles after identity verification. Specifically, after facial recognition of a resident's identity, service resources are matched to that resident based on their profile and behavioral data. The seamless collaboration between the mobile phone and the terminal supports scanning QR codes to obtain extended information or synchronize the operation interface. Specifically, it allows users to obtain extended information or synchronize the current operation interface to the mobile phone by scanning the terminal's QR code.

[0011] As a further aspect of the present invention: the data collected by the behavior data acquisition module includes: People flow data: acquired through cameras or infrared sensors; Dwell time data: Records the time residents spend in front of the terminal, specifically the time residents spend watching or operating the terminal; Interaction event data: includes the click location, dwell time, voice request content, QR code scanning type and frequency, and number of de-identified face recognition operations; Environmental data: Temperature, humidity, light, and noise collected by additional sensors.

[0012] As a further aspect of the present invention: the AI ​​demand perception engine module is deployed in the cloud or edge computing node, receives behavioral data and historical interaction data, applies machine learning or deep learning algorithms, establishes a data model through the Apriori algorithm and FP-Growth algorithm to identify the correlation between population characteristics and service needs, and calculates the degree of matching between residents and service resources through a quantitative evaluation model to achieve accurate recommendations.

[0013] As a further aspect of the present invention: the quantitative evaluation model considers factors at multiple levels, including basic attributes, behavioral preferences, and social influence. The quantitative evaluation model includes the standardization of basic attributes, the standardization of behavioral preferences, the standardization of social influence, and the calculation of a comprehensive score.

[0014] As a further aspect of the present invention: the standardized formula for the basic attributes is:

[0015] in, Indicates population In terms of service resource types basic attributes The original value on, This is the standardized value.

[0016] As a further aspect of the present invention: the standardized formula for behavioral preferences is: ,in, Indicates population In terms of service resource types behavioral preferences The original value on, This is the standardized value.

[0017] As a further aspect of the present invention: the standardized formula for social influence is: ,in, Indicates population The original value of social influence. This is the standardized value.

[0018] As a further aspect of the present invention: wherein, Refers to different types of service resources, and each type of service resource All Basic attributes At the same time, each service resource There are also kinds behavioral preferences , It indicates social influence.

[0019] As a further aspect of the present invention: the comprehensive score calculation formula is as follows:

[0020] in, Indicates population With service resource type The matching degree score, i.e., the overall score, For basic attributes Corresponding to the preset weighting coefficients, To target behavioral preferences Corresponding to the preset weighting coefficients, In response to social impact Corresponding to preset weighting coefficients; where, As a moderating factor, it increases the scores of users with higher social influence.

[0021] As a further aspect of the present invention: the service resources are pushed to the top N people with the highest comprehensive scores based on the comprehensive scores of each population.

[0022] As a further aspect of the present invention: the precise information delivery module, based on the analysis results of the AI ​​demand perception engine module, achieves information push through spatial orientation, crowd orientation, scene orientation, and behavior triggering; The spatial orientation filters information by building, unit, or designated public area; The targeted audience push is based on resident tag profiles; The scenario-based push notifications are triggered by time, weather, or special events. The behavior trigger pushes relevant information when a specified behavior is detected.

[0023] As a further aspect of the present invention: the information display and feedback module includes a dynamic content arrangement engine, a multimodal fusion presentation unit, and a real-time feedback closed-loop processing unit; The content dynamic arrangement engine arranges the information layout and style according to terminal characteristics, information priority, resident interaction status and demand profile; specifically, it receives instructions and content from the precise reach decision module, and dynamically arranges the layout and visual style of the information in real time according to the characteristics of the terminal display area, information priority, resident's current interaction status and personal demand profile. The multimodal fusion presentation unit includes a main visual area displaying high-priority information, a personalized recommendation area displaying customized content, an interactive guidance area guiding interaction, and a voice broadcast subunit that selects the broadcast method based on urgency and scenario. in: The main visual area specifically refers to the intelligent display of high-priority, universally applicable graphic / video information, such as emergency notices and important events, in the core area of ​​the display screen.

[0024] The personalized recommendation area dynamically displays customized content in designated areas based on AI-predicted individual or small group needs; designated areas include sidebars and floating windows, and customized content includes designated interest activities and personalized reminders.

[0025] The interactive guidance area clearly displays the voice wake-up word, touch buttons, or QR code icons to guide residents to make inquiries or provide feedback.

[0026] The voice broadcasting subunit uses a speaker to intelligently select to broadcast the full text, summary, or key prompts based on the urgency of the information and the context, avoiding excessive interference.

[0027] The real-time feedback closed-loop processing unit receives feedback through multiple channels and, after semantic analysis, sends it back to the data storage and analysis center; the real-time feedback closed-loop processing unit includes: Multi-channel feedback receiving submodule: integrates multiple feedback channels corresponding to touch, voice, and QR code / NFC; The feedback semantic analysis submodule performs real-time sentiment analysis and key opinion extraction on voice feedback; and performs topic classification and sentiment judgment on text feedback. The real-time feedback data feedback submodule transmits structured and unstructured feedback data, along with the interaction context, to the data storage and analysis center and the AI ​​demand awareness engine in real time for service optimization and model iteration.

[0028] A method for accurately reaching residents about their needs based on AI, implemented through an AI-based intelligent signage system, is characterized by the following steps: Step 1: Collect residents' behavior data around the smart tag terminal in real time through the behavior data collection module, including pedestrian flow, dwell time, interaction events and environmental data, and transmit it to the data storage and analysis center; Step 2: The AI ​​demand perception engine module retrieves data from the data storage and analysis center, applies the Apriori algorithm and FP-Growth algorithm to establish a correlation model, and calculates the matching degree score between residents and service resources through the quantitative evaluation model; Step 3: The precise information delivery module generates an information push strategy based on the matching degree score and targeting rules; Step 4: The information display and feedback module dynamically arranges information according to the push strategy and displays it through the smart signage terminal, while receiving feedback from residents; Step 5: Feedback data flows back to the data storage and analysis center. The AI ​​demand perception engine module optimizes the model based on the feedback data, and the information precision delivery module iterates the push strategy.

[0029] (III) Beneficial Effects This invention provides an AI-based intelligent tagging system and a method for precise information delivery. Compared with existing technologies, it has the following advantages: This invention integrates multiple interaction methods, including touch, voice, facial recognition, and mobile phone linkage (QR code / NFC / Bluetooth), to cover the needs of residents of different ages and with different usage habits. For example, the elderly can wake up the intelligent service robot with voice to ask questions, young people can scan a code with their mobile phones to operate the interface or control it remotely, and those with poor eyesight can rely on voice feedback to achieve personalized interaction, lower the technical threshold for use, and improve the inclusiveness of community services.

[0030] This invention, based on facial recognition and linking resident profiles and behavioral data, automatically displays matching service resources (such as recommendations for health lectures for the elderly and reminders for parent-child activities for parents) after identity authentication; the seamless mobile phone collaboration function supports scanning codes to obtain extended information or synchronized operation interfaces, further extending service scenarios and allowing residents to experience "personalized" services.

[0031] This invention utilizes a behavior data collection module to comprehensively capture data such as pedestrian traffic, dwell time, and interaction events (e.g., touch location, voice request content). Combined with machine learning algorithms (Apriori, FP-Growth) and quantitative evaluation models from an AI demand perception engine, it accurately identifies the correlation between population characteristics and service needs. By weighting the matching degree across multiple dimensions—basic attributes, behavioral preferences, and social influence—it ensures that service resource recommendations are more closely aligned with residents' actual needs, reducing interference from "invalid information."

[0032] This invention features a real-time feedback closed-loop processing unit that feeds residents' touch feedback, voice evaluation, and QR code scanning feedback back to the data center. The AI ​​engine continuously optimizes model parameters and push strategies based on the feedback, forming a virtuous cycle of "data collection - demand analysis - service push - feedback iteration". This allows community services to be dynamically adjusted according to changes in residents' needs, continuously improving the accuracy of service matching.

[0033] This invention's precise information delivery module achieves "on-demand" information delivery through four redirection logics: spatial targeting (building / unit), demographic targeting (resident tagging and profiling), scenario targeting (time / weather / special events), and behavioral triggering (such as pushing relevant services if the device's dwell time exceeds a threshold). For example, it can push elevator maintenance notices only to specific buildings or push parent-child activity information to families with children, avoiding the information overload problem caused by the "one-size-fits-all" notifications of traditional bulletin boards.

[0034] This invention's information display and feedback module uses a dynamic layout of a main visual area (high-priority information), a personalized recommendation area (customized content), and an interactive guidance area (operation instructions), combined with urgency-based adaptation of the voice broadcast sub-unit (full text / summary / prompt tone), to ensure efficient delivery of urgent notifications (such as fire warnings) and important events, reducing the risk of missing key information.

[0035] This invention utilizes a data storage and analysis center to integrate and analyze data such as pedestrian traffic, terminal usage frequency, and service request types, providing quantitative decision-making support for community management. For example, by analyzing terminal dwell time and interaction hotspots, terminal deployment locations can be optimized; by statistically analyzing high-frequency request types (such as facility maintenance and property service inquiries), property resources can be rationally allocated, improving problem response efficiency.

[0036] This invention enables intelligent service robots to answer questions intelligently based on large-scale model technology, and guide residents to submit service requests such as inquiries, demands, and suggestions, reducing the pressure on manual reception. The request data and feedback information are synchronized to the management end in real time, helping property management to quickly locate problems, track the progress of processing, shorten the service response cycle, and improve the level of precision in community governance.

[0037] This invention strictly adheres to the principle of "user authorization + privacy protection" during data collection and use: facial recognition employs de-identification processing and only performs identity verification with the resident's authorization; interactive data collection focuses on behavioral characteristics rather than sensitive personal information, balancing data value and privacy security through technical means, enhancing residents' trust in the intelligent system, and laying the foundation for the technology's implementation.

[0038] This invention, through the deep integration of terminal hardware, AI algorithms, and data platforms, breaks through the pain points of "information asymmetry" and "passive service" in traditional community services, realizing the transformation from "residents seeking services" to "services seeking residents." Simultaneously, functions such as mobile phone linkage and intelligent service robot collaboration promote seamless integration of online and offline service scenarios, providing a core interactive platform for the construction of a smart community ecosystem encompassing medical care, elderly care, and convenience services, thus helping community services upgrade towards digitalization, intelligence, and humanization.

[0039] In summary, this invention significantly improves residents' service experience, community management efficiency, and information dissemination quality through a full-link design that integrates multimodal interaction, AI demand perception, precise reach, and closed-loop feedback, providing a feasible technical solution and practical path for the construction of smart communities. Attached Figure Description

[0040] Figure 1 This is a system block diagram of the intelligent listing system based on AI demand perception of the present invention.

[0041] Figure 2 This is a flowchart illustrating the method for precise information delivery based on AI demand perception, as described in this invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Please see Figure 1 and Figure 2 As shown, this invention is an AI-based intelligent listing system for demand perception, comprising: Intelligent sign-on terminal: Deployed at the unit entrance and corresponding public areas of the activity center, it realizes touch, voice, face recognition and mobile phone linkage interaction through touch interaction sub-module, voice interaction sub-module, face recognition interaction sub-module and mobile phone linkage interaction module, providing residents with diversified operation access; In this embodiment, the intelligent sign-on terminal is a vertical or wall-mounted terminal device; Multimodal interaction module: Supports voice input / output, touch screen operation, and includes a face recognition camera, NFC / RFID reader, QR code generation and scanning unit to assist residents in submitting requests, provide personalized services through face recognition linkage, and expand information acquisition channels through mobile phone collaboration; Behavioral data collection module: Deployed on the smart signage terminal, it is used to collect real-time data on pedestrian traffic, dwell time, interaction events, and environmental data around the smart signage terminal. Among them, the interaction event data records the details of residents' operations, and the environmental data assists in the analysis of behavioral patterns. AI Demand Perception Engine Module: The core intelligent module, deployed in the cloud or edge computing node, is used to receive data collected from the behavior data collection module and apply machine learning or deep learning algorithms for analysis and processing. It establishes data associations through machine learning or deep learning algorithms, performs standardization processing through a quantitative evaluation model to eliminate the influence of data dimensions, and calculates the matching degree score by integrating multiple factors through a comprehensive scoring formula to ensure accurate recommendation of service resources. Data storage and analysis center: used to store all raw data, including behavioral data, interaction logs, information content, and environmental data generated during system operation, as well as the analysis results of the AI ​​demand perception engine module; Precise Information Delivery Module: Utilizing resident interaction feedback data collected from the information display and feedback module, as well as AI analysis of the delivery effect, the module continuously iterates and optimizes information matching algorithms and push strategies through reinforcement learning or online learning algorithms to improve the relevance, acceptability, and resident satisfaction of information, effectively solve the problem of information overload, and ensure that "useful information" is accurately delivered to "those who need it". Information display and feedback module: Deployed in the smart signage terminal, it receives instructions from the precise reach decision module, dynamically arranges the information presentation style, and displays different types of information through multiple areas; at the same time, it integrates the multimodal interaction module to receive interactive feedback from residents, such as voice evaluation, touch rating, and QR code evaluation.

[0044] This invention also provides a method for precise information delivery based on AI demand perception, which includes the following steps: Step 1: Collect residents' behavior data around the smart tag terminal in real time through the behavior data collection module, including pedestrian flow, dwell time, interaction events and environmental data, and transmit it to the data storage and analysis center; Step 2: The AI ​​demand perception engine module retrieves data from the data storage and analysis center, applies the Apriori algorithm and FP-Growth algorithm to establish a correlation model, and calculates the matching degree score between residents and service resources through the quantitative evaluation model; Step 3: The precise information delivery module generates an information push strategy based on the matching degree score and targeting rules; Step 4: The information display and feedback module dynamically arranges information according to the push strategy and displays it through the smart signage terminal, while receiving feedback from residents; Step 5: Feedback data flows back to the data storage and analysis center. The AI ​​demand perception engine module optimizes the model based on the feedback data, and the information precision delivery module iterates the push strategy.

[0045] The embodiments of the present invention provide the following technical solutions: As an embodiment of the present invention: In its implementation, this application will provide targeted services to disabled individuals facing difficulties, based on feedback from community residents. The specific methods are as follows: Step A1, Data Acquisition: The behavioral data collection module records the resident's interaction data with the smart sign-on terminal, including voice requests for "help for disabled people" and touch clicks on the "policy consultation" option, with a dwell time of 8 minutes; at the same time, it collects the ambient noise level of the terminal's surrounding environment at 55dB.

[0046] Step A2, AI Requirements Analysis: The AI ​​demand perception engine module retrieves data and uses the Apriori algorithm to discover a strong correlation between the "disability" label and services such as "hardship subsidies" and "barrier-free renovations".

[0047] Based on the quantitative assessment model, the "disability status" attribute in this resident profile corresponds to... =0.9, weight =0.3; The "frequency of policy consultation" in behavioral preferences corresponds to... =0.8, weight =0.4; Social influence =0.2, =0.3.

[0048] The calculation yields: .

[0049] Step A3, Information Push: The precise information delivery module generates push strategies based on population targeting (“disabled person” tag) and scenario targeting (after the request is submitted).

[0050] Step A4, Information Display: The information display and feedback module arranges policy information such as "living allowance for disabled people in difficulty" and "home barrier-free renovation subsidy" in the personalized recommendation area, with the font size enlarged to 20pt and the colors using high contrast. Due to the ambient noise level of 55dB, the voice broadcast summary is "We have pushed the disability assistance policy to you. Click to view details."

[0051] Step A5, Feedback and Optimization: Residents click the "Helpful" rating button via touch, and feedback data is returned. The AI ​​demand perception engine module increases the weight of "policy consultation" behavior preference to 0.45, and the information precision delivery module increases the frequency of pushing similar policies.

[0052] This embodiment focuses on the precise delivery of services to people with disabilities facing difficulties. A behavior data collection module comprehensively records residents' interactions with smart terminals and environmental noise data, providing solid data support for needs analysis. The AI-powered needs perception engine leverages the Apriori algorithm to accurately identify the strong correlation between the "disability" label and core support services, and scientifically calculates needs weights through a quantitative evaluation model, achieving precise quantification of needs. The information delivery process combines population-oriented and scenario-oriented strategies to ensure the targeted reach of policy information; information display employs measures such as enlarged fonts, high-contrast color schemes, and voice broadcasting to fully consider the ease of use for people with disabilities. The feedback mechanism optimizes behavioral preference weights and delivery frequency in real time through user ratings, forming a closed loop of "data collection - needs analysis - precise delivery - feedback optimization," effectively improving the efficiency of access to support policies for people with disabilities facing difficulties and enhancing the accuracy and applicability of services.

[0053] As a second embodiment of the present invention: In its specific implementation, compared to Embodiment 1, the only difference between the technical solution of this embodiment and Embodiment 1 is that, in this embodiment, for communities with a large number of elderly residents, community meal assistance services for the elderly are promoted, and the specific method is as follows: Step B1, Data Collection: The behavioral data collection module statistics show that residents aged 60 and above spend an average of 5 minutes in front of the smart signage terminal between 7:00-9:00 and 11:00-13:00 every day. They frequently click on the "Life Services" module and make 12 voice requests for "Canteen Information". Environmental data shows that the light intensity is 300 lux at 11:00 every day.

[0054] Step B2, AI Demand Analysis: The AI ​​demand perception engine module uses the FP-Growth algorithm to mine the association rules between the "elderly" tag, "meal time," and "community meal assistance" services. For resident A, aged 65, the attribute is "age." 0.95, weight =0.35; Behavioral preference "Frequency of clicks on lifestyle services" =0.85, weight =0.3; Social influence =0.1, =0.2.

[0055] The calculation yields: .

[0056] Step B3, Information Push: The precise information delivery module pushes information based on demographic targeting (“elderly group” tag) and time targeting (peak dining hours).

[0057] Step B4, Information Display: The information display and feedback module shows pictures of the canteen environment in the main visual area, displays the menu for the next day in the personalized recommendation area, and displays a "scan to reserve" QR code in the interactive guidance area; due to the light intensity of 300 lux, the screen brightness is adjusted to 80%, and the voice broadcasts "The community canteen will serve braised pork and green vegetables and tofu tomorrow. Scan the code to reserve."

[0058] Step B5, Feedback and Optimization: Five elderly people filled out feedback by scanning a QR code, expressing their desire for "soft and sticky dishes." After semantic analysis, the data was fed back. The AI ​​demand perception engine module updated the behavioral profiles of the elderly, the information display and feedback module added a "soft and sticky dishes" tag to the menu, and the information precision delivery module moved the push notification time to 10:30 every day.

[0059] This embodiment focuses on the delivery of community meal assistance services for the elderly. A behavioral data collection module captures key behavioral data such as the duration of elderly people's stay during peak dining hours, module click frequency, and voice requests. Combined with ambient lighting data, this accurately identifies the elderly's meal service needs. An AI demand perception engine utilizes the FP-Growth algorithm to efficiently mine the association rules between the "elderly" tag, "meal time," and "community meal assistance." A quantitative model assesses factors such as age and behavioral preferences to improve the accuracy of demand identification. The push strategy is based on targeting the elderly population and peak dining times, ensuring service information reaches the target group at key points. Information is presented intuitively through canteen photos, next-day menus, and reservation QR codes, with screen brightness adjusted according to light intensity and voice announcements of dishes, catering to the cognitive and operational habits of the elderly. The addition of a "soft and chewy dishes" tag driven by user feedback and the optimization of push times further enhance the personalization and timeliness of the meal assistance service, effectively solving the problem of accessing meal services for the elderly.

[0060] As an embodiment of the present invention: In its specific implementation, compared to Embodiment 1 and Embodiment 2, the only difference between this embodiment and Embodiment 1 and Embodiment 2 is that in this embodiment, a welcome service message is pushed to newly moved-in residents of the community. The specific method is as follows: Step C1, Data Acquisition: The behavior data collection module identifies the new resident, B, through the facial recognition interaction sub-module, records the duration of B's ​​first stay in front of the terminal as 10 minutes, and shows that B interacts with modules such as "Community Introduction" and "Property Management Process" via touch, and the interface is synchronized by scanning the terminal's QR code with a mobile phone.

[0061] Step C2, AI Requirements Analysis: The AI-powered demand perception engine module, based on the "new occupancy" tag, identifies strong correlations with services such as "property registration," "surrounding facilities," and "garbage disposal points" through an association model. Resident B's attribute is "occupancy duration." =0.9, weight =0.4; Behavioral preference "Frequency of exploring new features" =0.95, weight =0.3; Social influence =0.15, =0.25. The calculation yields: .

[0062] Step C3, Information Push: The precise information delivery module pushes information based on spatial orientation (resident B's unit) and behavioral trigger (first interaction).

[0063] Step C4, Information Display: The information display and feedback module dynamically arranges information. The main visual area displays a "Welcome to the Community" video, the personalized recommendation area lists information such as "Property Registration Process" and "Location of Nearby Supermarkets", and the interactive guidance area prompts "Wake up 'Xiao Bo' with your voice to ask more questions"; the voice broadcast reads the full text "Welcome to this community. The following is the service information you may need".

[0064] Step C5, Feedback and Optimization: Resident B provided voice feedback stating "very useful," and semantic analysis of the feedback data indicated a positive sentiment. The AI ​​demand perception engine module strengthened the correlation between the "new resident" tag and various services, while the information precision delivery module simultaneously pushed the service package to other new residents.

[0065] This embodiment focuses on push notifications for welcome services to newly moved-in residents. It leverages facial recognition technology to quickly identify new residents and records their initial interaction duration, operation modules, and QR code scanning behavior through a behavioral data collection module, accurately capturing their exploration needs for basic community services. An AI-powered demand perception engine, based on the "new resident" tag, efficiently identifies strong associations with core services such as property registration and surrounding facilities through a correlation model. Combined with quantitative evaluation indicators such as occupancy duration and function exploration frequency, it achieves precise demand positioning. Information pushes employ spatial orientation and behavioral triggering mechanisms to ensure a high degree of matching between service information and the resident's current scenario. Information display uses "welcome" videos, process lists, and voice prompts to intuitively present key information needed by new residents, lowering the information access barrier. User positive feedback-driven enhanced relevance and optimized synchronous service package pushes quickly form a standardized service model, helping new residents integrate into the community and improving their initial satisfaction with community services.

[0066] As an embodiment of the present invention: In specific implementation, compared with Embodiment 1, Embodiment 2, and Embodiment 3, the technical solution of this embodiment is to combine the solutions of Embodiment 1, Embodiment 2, and Embodiment 3.

[0067] This embodiment combines the technical solutions of the first three embodiments to achieve full coverage of the differentiated needs of different groups within the community, such as the disabled, the elderly, and newly moved-in residents. The solution integrates precise data collection dimensions for specific groups, diverse association rule mining algorithms (Apriori, FP-Growth, association models), multi-dimensional targeted push strategies (people, scenarios, time, space), and personalized display adaptation mechanisms, forming a highly versatile and scalable community service precision push system. Through the integrated analysis and service push of cross-group needs, the limitations of serving a single group are avoided, the comprehensive service efficiency of community smart terminals is improved, and precise service coverage of "one terminal, multiple scenarios" is achieved. This effectively enhances the sense of gain and satisfaction of different resident groups with community services and provides efficient technical support for refined community management.

[0068] It should be stated that all user data collected in this application was collected with the user's consent and authorization, and the use of user data is legal and compliant, and the use and processing of user data comply with the relevant laws, regulations and standards of the relevant regions.

[0069] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0070] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0071] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

[0072] It should also be noted that each block in a block diagram and / or flowchart, as well as combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0073] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. An AI-based intelligent listing system for demand perception, characterized in that: include: Intelligent tagging terminal: Deployed at unit entrances and corresponding public areas of the activity center, it realizes touch, voice, face recognition and mobile phone linkage interaction through touch interaction sub-module, voice interaction sub-module, face recognition interaction sub-module and mobile phone linkage interaction module; Multimodal interaction module: Supports voice input / output, touch screen operation, and includes a face recognition camera, NFC / RFID reader, QR code generation and scanning unit to assist residents in submitting requests, provide personalized services through face recognition linkage, and expand information acquisition channels through mobile phone collaboration; Behavioral data acquisition module: used to collect real-time data on pedestrian traffic, dwell time, interaction events, and environment around the smart signage terminal. Interaction event data records residents' operational details, and environmental data assists in analyzing behavioral patterns. AI Demand Perception Engine Module: Used to receive data collected from the behavior data collection module and apply machine learning or deep learning algorithms for analysis and processing. It establishes data associations through machine learning or deep learning algorithms, performs standardization processing through a quantitative evaluation model, and calculates the matching degree score by integrating multiple factors through a comprehensive scoring formula. Data storage and analysis center: used to store all raw data, including behavioral data, interaction logs, information content, and environmental data generated during system operation, as well as the analysis results of the AI ​​demand perception engine module; Precise Information Reaching Module: Utilizing resident interaction feedback data collected from the information display and feedback module, as well as AI analysis of the reach effect, the information matching algorithm and push strategy are continuously iterated and optimized through reinforcement learning or online learning algorithms. Information display and feedback module: used to receive instructions from the precise reach decision module, dynamically arrange the information presentation style, and display different types of information through multiple areas; at the same time, it integrates the multimodal interaction module to receive interactive feedback from residents.

2. The AI-based demand-aware intelligent listing system according to claim 1, characterized in that: The intelligent tagging terminal is a vertical or wall-mounted device, including a touch interaction submodule, a voice interaction submodule, a face recognition interaction submodule, and a mobile phone linkage interaction module; The touch interaction submodule responds to touch operations to display information or trigger functions; The voice interaction submodule collects voice commands through a microphone array, recognizes them, and then feeds them back through a speaker. The facial recognition interaction submodule uses a camera to identify the user's identity with authorization in order to provide personalized services; The mobile phone linkage interaction module connects with residents' mobile terminals via QR code, NFC, or Bluetooth.

3. The AI-based demand-aware intelligent listing system according to claim 1, characterized in that: The multimodal interaction module also supports intelligent service robot interaction, facial recognition linkage, and seamless mobile phone collaboration. The intelligent service robot uses process configuration and large model technology to answer questions and guide residents to submit their requests. The facial recognition linkage displays matching services based on resident profiles after identifying the resident's identity; The mobile phone seamless collaboration supports scanning QR codes to obtain extended information or synchronize the operation interface.

4. The AI-based demand-aware intelligent listing system according to claim 1, characterized in that: in, People flow data is acquired through cameras or infrared sensors; dwell time data records the time residents spend in front of the terminal; interaction event data includes touch operation click locations, dwell time, voice request content, QR code scanning type and frequency, and number of de-identified face recognitions; environmental data consists of temperature, humidity, light, and noise collected by additional sensors.

5. The AI-based demand-aware intelligent listing system according to claim 1, characterized in that: The AI ​​demand perception engine module is deployed in the cloud or edge computing nodes. It receives behavioral data and historical interaction data, applies machine learning or deep learning algorithms, and establishes a data model through the Apriori algorithm and FP-Growth algorithm to identify the correlation between population characteristics and service needs. It also calculates the degree of matching between residents and service resources through a quantitative evaluation model.

6. The AI-based demand-aware intelligent listing system according to claim 5, characterized in that: The quantitative assessment model includes basic attribute standardization, behavioral preference standardization, social influence standardization, and comprehensive score calculation; The standardized formula for the basic attributes is:

7. Among them, Indicates population In terms of service resource types basic attributes The original value on, The value is the standardized value; The standardized formula for behavioral preferences is: ,in, Indicates population In terms of service resource types behavioral preferences The original value on, This is the standardized value.

8. Among them, Refers to different types of service resources, and each type of service resource All Basic attributes At the same time, each service resource There are also kinds behavioral preferences .

9. The AI-based demand-aware intelligent listing system according to claim 6, characterized in that: The standardized formula for social influence is: ,in, Indicates population The original value of social influence. The standardized value. Indicates social influence; The formula for calculating the comprehensive score is as follows:

10. Among them, Indicates population With service resource type The matching degree score, i.e., the overall score, For basic attributes Corresponding to the preset weighting coefficients, To target behavioral preferences Corresponding to the preset weighting coefficients, In response to social impact The corresponding preset weighting coefficients.

11. The AI-based demand-aware intelligent listing system according to claim 1, characterized in that: The precise information delivery module is based on the analysis results of the AI ​​demand perception engine module and pushes information through spatial orientation, crowd orientation, scene orientation and behavior triggering. The spatial orientation filters information by building, unit, or designated public area; The targeted audience push is based on resident tag profiles; The scenario-based push notifications are triggered by time, weather, or special events. The behavior trigger pushes relevant information when a specified behavior is detected.

12. The AI-based demand-aware intelligent listing system according to claim 1, characterized in that: The information display and feedback module includes a dynamic content arrangement engine, a multimodal fusion presentation unit, and a real-time feedback closed-loop processing unit; The content dynamic arrangement engine arranges the information layout and style according to terminal characteristics, information priority, resident interaction status and demand profile; The multimodal fusion presentation unit includes a main visual area displaying high-priority information, a personalized recommendation area displaying customized content, an interactive guidance area guiding interaction, and a voice broadcast subunit that selects the broadcast method based on urgency and scenario. The real-time feedback closed-loop processing unit receives feedback through multiple channels and sends it back to the data storage and analysis center after semantic analysis.

13. A method for accurately reaching residents about their needs based on AI, the method being implemented through the intelligent signage system based on AI needs as described in any one of claims 1-9, characterized in that... The method includes the following steps: Step 1: Collect residents' behavior data around the smart tag terminal in real time through the behavior data collection module, including pedestrian flow, dwell time, interaction events and environmental data, and transmit it to the data storage and analysis center; Step 2: The AI ​​demand perception engine module retrieves data from the data storage and analysis center, applies the Apriori algorithm and FP-Growth algorithm to establish a correlation model, and calculates the matching degree score between residents and service resources through the quantitative evaluation model; Step 3: The precise information delivery module generates an information push strategy based on the matching degree score and targeting rules; Step 4: The information display and feedback module dynamically arranges information according to the push strategy and displays it through the smart signage terminal, while receiving feedback from residents; Step 5: Feedback data flows back to the data storage and analysis center. The AI ​​demand perception engine module optimizes the model based on the feedback data, and the information precision delivery module iterates the push strategy.

Images