Intelligent business business travel data acquisition and analysis system based on Internet of Things

Abstract

The invention relates to the technical field of Internet of Things, and discloses an intelligent business business travel data acquisition and analysis system based on Internet of Things, which comprises a data acquisition module, a data transmission module, a data processing and analysis module and a service module, through the multi-mode sensor network, the whole scene data acquisition of the business business travel is realized, and the limitation of a traditional single field is broken through; by means of edge computing and NB-IoT hierarchical data transmission, the delay problem of cloud centralized processing is solved, and meanwhile, the credibility of key data is guaranteed by means of block chain evidence storage; and then a digital twin module is combined with an AI algorithm, a dynamic virtual mirror image is constructed, intelligent analysis such as passenger flow prediction and anomaly detection is realized, and the defects of traditional visualization and decision support are made up.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and more specifically discloses an intelligent cultural, commercial, sports and tourism data collection and analysis system based on the Internet of Things. Background Art

[0002] The IoT-based intelligent cultural, commercial, sports and tourism data collection and analysis system integrates data collection, integration, analysis and application. It aims to collect and process data covering multiple fields such as culture, commerce, sports and tourism through IoT technology. IoT technology uses sensors, smart devices and network connections to collect and process various data in real time to support decision-making and optimize services.

[0003] The collection and analysis systems in existing technologies have a single scenario coverage and focus on a single field. They lack the ability to coordinate collection across all scenarios. In addition, they rely on centralized cloud processing, resulting in high latency and a lack of anti-tampering mechanisms. They also lack user experience data, and have weak visualization and simulation capabilities. There is a lack of spatial dynamic visualization tools, insufficient intelligent decision support, and reliance on manual experience and a lack of dynamic simulation predictions. Finally, the system is closed, cross-domain data is difficult to share, and the application of emerging technologies is fragmented, without forming a complete technical closed loop. Summary of the Invention

[0004] The main technical problem solved by the present invention is to provide an intelligent cultural, commercial, sports and tourism data collection and analysis system based on the Internet of Things, which can solve the problems of incomplete data collection, insufficient intelligence, inefficient cross-departmental collaboration and poor user experience in the current intelligent cultural, commercial, sports and tourism data collection and analysis system based on the Internet of Things.

[0005] To solve the above technical problems, according to one aspect of the present invention, more specifically, an intelligent data collection and analysis system for culture, commerce, sports, and tourism based on the Internet of Things (IoT), comprising: first, a data collection module, utilizing a multimodal sensor network, comprehensively collects data in four major scenarios: cultural and sports venues, commercial blocks, sports events, and rural tourism; then, the collected data is transmitted to a data transmission module. In the data transmission module, real-time, high-frequency data is sent to an edge computing node module for local cleaning and feature extraction to reduce the amount of data transmitted, while non-real-time, low-frequency data is transmitted via the NB-IoT network. Simultaneously, key data is stored on the blockchain by a blockchain evidence storage module to ensure data credibility. After transmission, the data enters a data processing and analysis module; within this module, a digital twin module is responsible for constructing virtual images, while an AI analysis module utilizes collaborative filtering, isolation forest, and reinforcement learning algorithms to conduct in-depth data analysis, specifically including predicting passenger flow, recommending services, and detecting anomalies; finally, a service module presents the results obtained by the data processing and analysis module to management, user, and developer terminals, thereby enabling intelligent decision-making, personalized services, and the construction of a data collaborative ecosystem.

[0006] Furthermore, the data transmission module includes: an edge computing node module and a blockchain evidence storage module;

[0007] The edge computing node module receives real-time high-frequency data, performs local cleaning and feature extraction, and uploads only key information to the cloud via the 5G network to reduce transmission delays and traffic costs;

[0008] The blockchain evidence storage module stores key data on the chain to ensure that the data cannot be tampered with, providing a credible basis for consumer rights protection and cross-departmental supervision.

[0009] Furthermore, the data processing and analysis module includes: a digital twin module and an AI analysis module;

[0010] The digital twin module builds a virtual mirror of culture, commerce, sports and tourism based on multi-source data integrated by the data center, and maps the flow of people and facility status elements in real time;

[0011] The AI ​​analysis module uses algorithms to conduct in-depth analysis of data to achieve prediction, recommendation, and detection functions.

[0012] Furthermore, the AI ​​analysis module includes: collaborative filtering algorithm, isolation forest algorithm, and reinforcement learning algorithm;

[0013] Collaborative filtering algorithm, based on user behavior data, builds tourist profiles and generates personalized recommendations;

[0014] The Isolation Forest algorithm uses computer vision algorithms to analyze scenic area surveillance videos and identify dangerous behaviors such as climbing over guardrails within the scenic area.

[0015] The reinforcement learning algorithm dynamically adjusts the resource scheduling strategy with "tourist satisfaction" and "operating cost" as the objective functions.

[0016] Furthermore, the data acquisition module adopts a multimodal sensor network to collect data in cultural and sports venues, commercial blocks, sports events, and rural tourism scenarios using visual cameras, environmental sensors, Wi-Fi probes, smart POS machines, GPS trackers, pressure sensors, and NB-IoT sensor devices.

[0017] Furthermore, the service module includes: a user side, a management side, and a developer side;

[0018] The user side provides users with an immersive experience of restoring the historical scenes of the Ming Dynasty City Wall, while collecting user behavior data for system optimization;

[0019] On the management side, the digital twin large screen dynamically displays global data, supports simulation of different policy effects, and generates emergency instructions;

[0020] On the developer side, open API interface for third-party calls.

[0021] Furthermore, the data collected by the data collection module is aggregated through the Internet of Things protocol to form an original data set of culture, commerce, sports and tourism.

[0022] The beneficial effects of the IoT-based intelligent data collection and analysis system for culture, commerce, sports, and tourism are as follows: It enables data collection across all scenarios of culture, commerce, sports, and tourism through a multimodal sensor network, breaking through the limitations of traditional single fields; it leverages edge computing and NB-IoT to transmit data in layers, resolving latency issues associated with centralized cloud processing, while also utilizing blockchain evidence storage to ensure the credibility of key data; and it combines digital twin modules with AI algorithms to construct dynamic virtual images and implement intelligent analysis such as passenger flow prediction and anomaly detection, addressing the shortcomings of traditional visualization and decision support.

[0023] Through the multi-end design of the service module (user end, management end, developer end), comprehensive coverage of scene implementation is achieved: the user end provides immersive experience and personalized services to improve tour satisfaction; the management end relies on the digital twin large screen to realize dynamic resource scheduling and emergency response, and improve operational efficiency; the developer end opens the API interface to promote cross-departmental and cross-enterprise data collaboration, activate integrated business formats such as "events + tourism + consumption", and ultimately form a closed-loop ecology of "collection-analysis-application-feedback", and promote the cultural, commercial, sports and tourism industries to upgrade towards intelligence and collaboration. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention will be further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0025] Figure 1 Schematic diagram of the system principle;

[0026] Figure 2 Schematic diagram of the method flow chart. DETAILED DESCRIPTION

[0027] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0028] According to one aspect of the present invention, Figure 1-Figure 2 As shown, an intelligent cultural, commercial, sports and tourism data collection and analysis system based on the Internet of Things is provided, including: first, the data collection module uses a multimodal sensor network to comprehensively collect data in four major scenarios: cultural and sports venues, commercial blocks, sports events, and rural tourism;

[0029] Data collection is carried out through a multimodal sensor network in cultural and sports venues, commercial blocks, sports events, and rural tourism scenarios using the following devices:

[0030] Cultural and sports venues: Deploy visual cameras to count visitor stays and analyze popular exhibits. Use environmental sensors (temperature, humidity, and illumination) to monitor the microenvironment of cultural relics display cabinets in real time, triggering warnings when anomalies occur. For example, when holding a concert at a basketball stadium, historical crowd flow data can be used to predict entrance and exit congestion points and facilitate early traffic flow.

[0031] Commercial districts: Wi-Fi probes are used to capture mobile phone MAC addresses, anonymously analyzing customer flow patterns and hotspots. Smart POS machines (with integrated NFC modules and consumer product tags) are used to collect transaction data in real time and generate best-selling product lists.

[0032] Sports events: Athletes wear GPS trackers and heart rate monitors to transmit their movement trajectories and physiological indicators (such as heart rate and speed) in real time; pressure sensors are deployed in the audience seats to monitor crowd density and identify abnormal gathering risks.

[0033] Rural tourism: B&Bs install NB-IoT smart door locks to collect data such as check-in time and length of stay;

[0034] All types of data are aggregated through the Internet of Things protocol (MQTT) to form an original data set of culture, commerce, sports and tourism that includes elements such as passenger flow distribution, environmental status, consumption behavior, and movement trajectory.

[0035] The collected data is then transmitted to the data transmission module. In the data transmission module, real-time high-frequency data is sent to the edge computing node module for local cleaning and feature extraction to reduce the amount of data transmitted. Non-real-time low-frequency data is transmitted via the NB-IoT network. At the same time, key data is stored on the blockchain by the blockchain evidence module to ensure the credibility of the data. After transmission, the data enters the data processing and analysis module.

[0036] Among them, the edge computing node module receives real-time high-frequency data (such as scenic spot video streams and real-time trajectory data of sports events), performs local cleaning and feature extraction (such as filtering invalid images, identifying abnormal behaviors, and removing redundant motion trajectory points), and only uploads key information (such as crowd density thresholds, consumption amounts, and athlete heart rate abnormality indicators) to the cloud via the 5G network to reduce transmission delays and traffic costs (response speed is increased by 90% and traffic consumption is reduced by 70%).

[0037] Local cleaning and feature extraction uses computer vision algorithms (such as YOLOv5 object detection) to filter invalid images from scenic area video streams, identify abnormal behaviors such as climbing over guardrails and crowd gatherings, and use data feature extraction techniques (such as spatial clustering of motion trajectory points) to remove redundant points in sports event trajectory data, retaining only key coordinates and speed parameters.

[0038] The blockchain evidence storage module stores key data (such as intangible cultural heritage product transaction records, event ticketing information, and core homestay occupancy data) on the chain. It builds a distributed ledger based on alliance chain technology (such as Hyperledger Fabric). Through encryption algorithms and consensus mechanisms, it ensures that data cannot be tampered with, providing a full-link trusted traceability record for consumer rights protection, supporting cross-departmental regulatory data collaborative verification, ensuring that data cannot be tampered with, and providing a credible basis for consumer rights protection and cross-departmental supervision;

[0039] The on-chain evidence storage is based on the construction of a distributed ledger based on alliance chain technology (such as Hyperledger Fabric), and is jointly maintained by multiple nodes such as the Cultural and Tourism Bureau, the Market Supervision Bureau, and enterprises. The data is hashed using the SHA-256 encryption algorithm to generate a unique and tamper-proof digital fingerprint. The PBFT consensus mechanism is used to ensure the consistency and reliability of data on the chain, and to achieve full-chain traceability of intangible cultural heritage products from production to sales (such as Yunjin's silk breeding, dyeing process, quality inspection report and other data are uploaded to the chain in real time). At the same time, it supports cross-departmental collaborative verification of regulatory data (such as the Cultural and Tourism Bureau and the Public Security Bureau sharing scenic spot carrying capacity evidence data), providing trusted timestamps and operation log audits for consumer rights protection, and ensuring that the data cannot be tampered with and is traceable.

[0040] Then, within this module, the digital twin module is responsible for building a virtual image, while the AI ​​analysis module uses collaborative filtering, isolation forest, and reinforcement learning algorithms to conduct in-depth data analysis, including predicting passenger flow, recommending services, and detecting anomalies.

[0041] The digital twin module, based on multi-source data (such as passenger flow, facility status, and environmental parameters) integrated by the data center, uses 3D modeling technology (such as oblique photography and BIM integration) to build 1:1 virtual mirrors of cultural, commercial, sports, and tourism scenes (such as the Confucius Temple block and the Youth Olympic Sports Park). It maps factors such as the heat distribution of pedestrian flow, the business status of shops, and equipment operating parameters in real time with an accuracy of 95%. Then, through dynamic rendering technology (such as the Three.js engine), managers can intuitively view the distribution of resources in different time periods and different areas, and simulate the effects of strategies such as lighting adjustment and route optimization (such as predicting consumption time in the night economy scenario). Finally, the digital twin screen displays the results.

[0042] The AI ​​analysis module relies on three algorithms to achieve in-depth analysis:

[0043] The collaborative filtering algorithm builds a tourist profile based on user behavior data and generates personalized recommendations. The specific implementation method is as follows: first, user browsing and consumption behavior data is collected and cleaned and standardized, and a sparse matrix of user-item interactions is constructed. Then, the similarity between items is calculated using adjusted cosine similarity. Then, a recommendation list is generated for the target user. At the same time, it handles the cold start problem, balances diversity and novelty, and optimizes real-time performance.

[0044] The cosine similarity calculation formula is as follows:

[0045]

[0046] Where U represents the set of users who rate items i and j at the same time, R u Represents the average rating of user u. For the target user u, the K items with the highest similarity to his / her historical preference items are recommended. The specific formula is as follows:

[0047]

[0048] Where N(u) represents the set of items rated by user u, and sim(i,j) represents the similarity between items i and j;

[0049] The Isolation Forest Algorithm uses computer vision algorithms to analyze scenic area surveillance videos and identify dangerous behaviors such as climbing over guardrails in scenic areas. In specific implementation, the data points are recursively segmented by constructing multiple isolation trees. Due to sparsity, abnormal points will be quickly isolated in the shallow layers of the tree. In this system, the algorithm first extracts features from multi-source data such as scenic area videos and passenger flow density and standardizes them (Z-score). Then, 100-200 isolation trees with a maximum depth (usually log2(n), where n is the number of samples) are randomly sampled to form a forest. For each data point, the average path length in all trees is calculated and converted into an anomaly score of 0-1. If the threshold is exceeded (such as 0.6), an alert is triggered. The rule engine is combined to identify anomaly types such as climbing over guardrails and equipment failures. Flink is used to distribute and process tens of thousands of event streams per second. The model is updated every hour using an online learning mechanism. Finally, the abnormal location is visualized on the digital twin platform, achieving a real-time alert response of less than 1 minute in cultural, commercial, sports and tourism scenarios.

[0050] For each data point x, the average path length h(x) across all trees is calculated and converted into an anomaly score s. The technical formula for the anomaly score is as follows:

[0051]

[0052] Where E[h(x)] is the average path length of x on all trees, and c(n) is the expected average path length of isolated trees with sample size n, calculated as:

[0053]

[0054] Where H(n) is the harmonic number;

[0055] The reinforcement learning algorithm uses "tourist satisfaction" and "operating cost" as objective functions to dynamically adjust resource scheduling strategies. Specifically, based on the Markov decision process, it models the cultural, commercial, sports and tourism scenarios as a multidimensional state space that includes passenger flow density, facility status, etc., defines action spaces such as sightseeing vehicle scheduling and staffing, and designs a weighted reward function centered on tourist satisfaction (based on stay time and complaint rate) and operating costs (energy consumption and labor costs). It uses the proximal policy optimization (PPO) algorithm combined with experience replay and distributed training to dynamically generate the optimal strategy by maximizing cumulative rewards. After deployment to the edge server, it adjusts resource allocation (such as sightseeing vehicle frequency and cleaning schedule) according to real-time status and continuously optimizes through online learning. In scenic area applications, it shortens tourist waiting time by 40% and reduces vehicle idle rate by 25%. At the same time, it introduces a barrier function to ensure safety constraints and explains the decision logic through SHAP values ​​to achieve dynamic resource scheduling with multi-objective balance.

[0056] The weighted reward function is mainly to balance the visitor experience and operating costs. The specific formula is as follows:

[0057] R=w1R 满意度 +w2R 成本 +w3R 公平性

[0058] Tourist satisfaction is calculated based on length of stay, waiting time, and complaint rate. Operating costs include energy consumption, labor costs, and equipment wear and tear. Fairness is the degree of balance in the tourist experience across regions. The Proximal Policy Optimization (PPO) algorithm is used due to its high sample efficiency and strong stability. Its formula is as follows:

[0059] max θ E t [r t (θ)A t ,clip(r t (θ),1-∈,1+∈)A t ]

[0060] Among them A t is the advantage function, r t (θ) is the probability ratio of the new and old strategies, clip() is the clipping function, which t (θ) is restricted to the interval [1-∈, 1+∈], where ∈ is a hyperparameter (usually 0.1 or 0.2).

[0061] Finally, the service module presents the results obtained by the data processing and analysis module to the management, user, and developer ends, thereby realizing intelligent decision-making, personalized services, and the construction of a data collaborative ecosystem.

[0062] The user side provides users with a variety of immersive experiences, including restoration of historical scenes of the Ming City Wall, simulation of urban life in ancient commercial districts, virtual viewing of sports events, and immersive roaming of rural scenery. For example, VR technology is used to restore interactive scenes of the construction process of the Ming City Wall, AR technology is used to allow tourists to "travel" to the Qinhuai River during the Ming and Qing Dynasties to participate in a virtual market, and 360° panoramic live broadcast technology is used to create a "cloud VIP viewing seat" for sports events. At the same time, user behavior data (such as experience duration, interaction preferences, path selection, etc.) is collected to optimize the system, forming a closed loop of "experience-data feedback-service upgrade";

[0063] On the management side, the digital twin large screen dynamically displays global data, supports simulation of different policy effects, and generates emergency instructions;

[0064] On the developer side, open API interfaces for third-party calls, such as opening real-time passenger flow data of scenic spots to tourism platforms for them to design "off-peak tour routes"; opening consumption preference data of sports event spectators to sports brands to support "precise delivery of event-related products"; opening occupancy rate data of rural tourism homestays to local governments to assist in the formulation of "cultural tourism support policies"; through cross-departmental data sharing (such as the joint optimization of scenic area shuttle bus scheduling by the Cultural and Tourism Bureau and the Transportation Bureau) and cross-enterprise ecological cooperation (such as the joint launch of "consumption-tourism packages" by merchants in commercial blocks and travel agencies), activate integration scenarios such as "converting event spectators into tourist groups" and "linking commercial district consumption with scenic area traffic", such as holding an event at the Youth Olympic Sports Park. During concerts, the system pushes ticketing data to surrounding hotels and catering companies. Merchants can allocate resources in advance and launch "performance + accommodation + food" packages to achieve efficient conversion of "event traffic-tourism consumption", and ultimately form a closed-loop ecosystem of "collection (multi-scene data)-analysis (intelligent algorithm modeling)-application (multi-terminal decision support)-feedback (user behavior and business data return optimization model)", promoting the cultural, commercial, sports and tourism industries from a single business to a "data-driven integrated ecology" upgrade. For example, in rural tourism scenarios, homestay occupancy data is fed back to agricultural cooperatives after analysis to guide them in developing "customized agricultural product picking for tourists" projects, forming a synergistic growth chain of "accommodation-agricultural experience-consumption".

[0065] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by ordinary technicians in this technical field within the essential scope of the present invention also fall within the scope of protection of the present invention.

Claims

1. An intelligent culture, commerce, sports and tourism data collection and analysis system based on the Internet of Things, characterized by: include: First, the data acquisition module uses a multimodal sensor network to comprehensively collect data in four major scenarios: cultural and sports venues, commercial blocks, sports events, and rural tourism; Subsequently, the collected data is transmitted to the data transmission module. In the data transmission module, real-time high-frequency data will be sent to the edge computing node module for local cleaning and feature extraction to reduce the amount of data transmitted. Non-real-time low-frequency data is transmitted through the NB-IoT network. At the same time, key data is stored on the chain by the blockchain evidence module to ensure the credibility of the data. After transmission, the data enters the data processing and analysis module; in this module, the digital twin module is responsible for building a virtual mirror, and the AI ​​analysis module uses collaborative filtering, isolation forest, and reinforcement learning algorithms to conduct in-depth analysis of the data, including predicting passenger flow, recommending services, and detecting anomalies; finally, the service module presents the results obtained by the data processing and analysis module on the management side, user side, and developer side, thereby realizing intelligent decision-making, personalized services, and the construction of a data collaborative ecosystem.

2. The intelligent culture, commerce, sports and tourism data collection and analysis system based on the Internet of Things according to claim 1 is characterized by: The data transmission module includes: an edge computing node module and a blockchain evidence storage module; The edge computing node module receives real-time high-frequency data, performs local cleaning and feature extraction, and uploads only key information to the cloud via the 5G network to reduce transmission delays and traffic costs; The blockchain evidence storage module stores key data on the chain to ensure that the data cannot be tampered with, providing a credible basis for consumer rights protection and cross-departmental supervision.

3. The intelligent culture, commerce, sports and tourism data collection and analysis system based on the Internet of Things according to claim 1 is characterized by: The data processing and analysis module includes: a digital twin module and an AI analysis module; The digital twin module builds a virtual mirror of culture, commerce, sports and tourism based on multi-source data integrated by the data center, and maps the flow of people and facility status elements in real time; The AI ​​analysis module uses algorithms to conduct in-depth analysis of data to achieve prediction, recommendation, and detection functions.

4. The intelligent culture, commerce, sports and tourism data collection and analysis system based on the Internet of Things according to claim 3 is characterized by: The AI ​​analysis module includes: collaborative filtering algorithm, isolation forest algorithm, and reinforcement learning algorithm; Collaborative filtering algorithm, based on user behavior data, builds tourist profiles and generates personalized recommendations; The Isolation Forest algorithm uses computer vision algorithms to analyze scenic area surveillance videos and identify dangerous behaviors such as climbing over guardrails within the scenic area. The reinforcement learning algorithm uses "tourist satisfaction" and "operating cost" as objective functions to dynamically adjust resource scheduling strategies.

5. The intelligent culture, commerce, sports and tourism data collection and analysis system based on the Internet of Things according to claim 1 is characterized by: The data acquisition module adopts a multimodal sensor network to collect data in cultural and sports venues, commercial blocks, sports events, and rural tourism scenarios using visual cameras, environmental sensors, Wi-Fi probes, smart POS machines, GPS trackers, pressure sensors, and NB-IoT sensor devices.

6. The intelligent culture, commerce, sports and tourism data collection and analysis system based on the Internet of Things according to claim 1 is characterized by: The service modules include: user side, management side, and developer side; The user side provides users with an immersive experience of restoring the historical scenes of the Ming Dynasty City Wall, while collecting user behavior data for system optimization; On the management side, the digital twin large screen dynamically displays global data, supports simulation of different policy effects, and generates emergency instructions; On the developer side, open API interface for third-party calls.

7. The intelligent culture, commerce, sports and tourism data collection and analysis system based on the Internet of Things according to claim 1 is characterized by: The data collected by the data collection module is aggregated through the Internet of Things protocol to form the original data set of culture, commerce, sports and tourism.

Images