Cinema real-time box office income monitoring system and method based on multi-source data fusion

Abstract

The invention discloses a cinema real-time box office income monitoring system and method based on multi-source data fusion, and the system comprises a multi-source data collection module which is used for collecting the internal and external multi-dimensional data of a cinema; the data fusion processing module is used for carrying out fusion and feature extraction on the collected multi-source data, and quantifying a synergistic effect of a movie hall on time intervals and physical distances through a cross-modal association network; and the intelligent decision analysis module is used for generating a box office monitoring and optimization strategy based on the fusion data, including calculating an overflow coefficient between movie halls, analyzing audience concentration and balancing energy consumption and the box office through a multi-objective optimization model. The cinema internal sensor data, ticket business data, external parking lot data and user authorized behavior data are integrated through the multi-source data acquisition module, the limitation of traditional single ticket business data is broken through, and all-around coverage of box office influence factors is realized.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of cinema operation management, and particularly relates to a cinema real-time box office income monitoring system and method based on multi-source data fusion. BACKGROUND

[0002] The existing cinema box office monitoring technology mainly relies on a ticket management system to realize core functions, collects basic data such as show information, seat sold data, and box office amount in real time through ticket selling terminals, self-service ticket kiosks and other devices, and generates statistical reports for managers to refer to. At the same time, some cinemas introduce basic sensing devices, such as infrared counters for counting the number of entries and cameras for monitoring the occupancy rate of the theater, to assist in verifying the accuracy of ticket data. In terms of film scheduling management, the existing system is mostly based on historical box office data and manual experience to develop a film scheduling plan, which allocates film shows and time intervals for each theater through a fixed template. In addition, energy consumption management mostly adopts a preset mode, such as setting fixed air conditioning temperature and lighting brightness, and adjusting the running state of the equipment through manual inspection. These technologies have realized the basic statistics and management of box office data in daily operation of the cinema, providing a basis for film scheduling, revenue accounting and other aspects.

[0003] However, the existing technology has obvious limitations in actual application and is difficult to meet the needs of fine operation of the cinema. In terms of data coverage dimension, the existing system mostly focuses on ticket data and lacks integrated analysis of audience behavior (such as viewing concentration, off-site dynamic line), external environment (such as surrounding parking lot traffic) and equipment energy consumption, resulting in one-sided interpretation of box office influencing factors. In terms of collaborative optimization capability, the traditional film scheduling strategy does not consider the space-time correlation effect between theaters, and it is difficult to improve the overall occupancy rate through multi-theater linkage; the fixed parameter mode of energy consumption management is easy to cause energy waste due to the lack of correlation with real-time occupancy rate. In terms of decision-making intelligence, the existing system relies on manual judgment to adjust the strategy, the response speed is lagging, and it is impossible to predict the film reputation through audience implicit feedback (such as mobile phone usage behavior). In addition, the balance between data security and privacy protection is insufficient, there is a lack of desensitization and traceability mechanism for sensitive data, which restricts the deep mining of data value. These deficiencies make the existing technology have a large room for improvement in box office promotion, cost control and operation efficiency optimization. SUMMARY

[0004] In order to overcome the above-mentioned defects of the existing technology, the embodiments of the present application provide a cinema real-time box office income monitoring system and method based on multi-source data fusion, which solves the problems of single data, rigid decision-making, high energy consumption, and insufficient security and responsiveness in the prior art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions: A cinema real-time box office income monitoring system and method based on multi-source data fusion, comprising: A multi-source data collection module is configured to collect multi-dimensional data inside and outside the cinema; A data fusion processing module is configured to fuse and extract features of the collected multi-source data, and quantify the synergistic effect of the cinema in the time interval and the physical distance through a cross-modal correlation network; An intelligent decision analysis module is configured to generate a box office monitoring and optimization strategy based on the fused data, including calculating the overflow coefficient between the cinemas, analyzing the audience concentration, and balancing the energy consumption and the box office through a multi-objective optimization model; An execution and feedback module is configured to execute the optimization strategy, realize data blockchain tracing, and visually display the data.

[0006] Preferably, the multi-source data collection module comprises: An Internet of Things sensor network is deployed in each cinema and public area of the cinema, including an infrared passenger flow counter for collecting the occupancy rate, a seat pressure sensor for detecting the seating state, an acoustic collection device for collecting environmental sound features, a positioning device for recording the audience's movement line, a smart meter for counting energy consumption, and an ambient light sensor for monitoring the light; A ticket data interface is connected to the cinema ticketing system to obtain the show time, seat sold information, film type, and ticket price; A third-party data interface is connected to the parking lot management system around the cinema to obtain the parking space occupancy data, vehicle type distribution, and parking duration; A user authorization data unit obtains the user-authorized mobile sensor data through a lightweight applet, and the data collection period is limited to the viewing process, including the gyroscope swing frequency, screen lighting frequency, and ambient light sensor change data.

[0007] Preferably, the data fusion processing module comprises: A feature preprocessing unit uses a feature selection algorithm to clean and filter the original data, generating a high-dimensional feature vector containing time dimension, space dimension, film type, audience behavior, and energy consumption; A multi-modal fusion unit dynamically calculates the cinema correlation matrix through a cross-modal correlation network, and the matrix node weight is dynamically updated based on the real-time occupancy rate of the cinema, time interval, and physical distance; A behavior feature extraction unit verifies the actual number of seated people by analyzing the environmental sound features through acoustic fingerprinting, generates a heat map of the audience's departure movement line combined with positioning data, and extracts the departure stay duration and secondary consumption correlation degree features.

[0008] Preferably, the intelligent decision analysis module comprises: A cinema synergy analysis unit calculates the overflow coefficient to quantify the driving effect of the audience leaving theater A on the occupancy rate of theater B, and generates a film scheduling combination suggestion combined with the film type similarity; The attention analysis unit builds an attention-box office correlation model based on mobile phone sensor data. It infers the audience's attention by using the gyroscope swing frequency and the number of times the screen is turned on, and predicts the word-of-mouth spread of the film. The energy consumption optimization unit uses a multi-objective optimization model to link occupancy rate and energy consumption data, and dynamically adjusts equipment operating parameters based on occupancy rate to optimize energy consumption per unit of box office.

[0009] Preferably, the execution and feedback module includes: The edge control unit adjusts the parameters of the theater lighting, air conditioning and projection equipment in real time based on the dynamic equipment control strategy output by the intelligent decision analysis module. The data traceability unit uses blockchain technology to store real-time box office data hash values, forming an immutable audit chain that supports multi-party queries; The visual interactive unit generates box office monitoring heatmaps and film scheduling optimization suggestions reports, allowing managers to adjust strategies and receive feedback on execution results via natural language commands.

[0010] Preferably, a method for real-time cinema box office revenue monitoring based on multi-source data fusion includes the following steps: (1) Multi-source data acquisition steps: Collect internal and external data of the cinema through the multi-source data acquisition module, including data such as occupancy rate and energy consumption collected by IoT sensor network, ticketing information obtained by ticketing data interface, parking lot data obtained by third-party data interface, and mobile phone sensor data authorized by users, and de-identify all data; (2) Data fusion and feature extraction steps: The collected data is fused through the data fusion processing module, and the theater association matrix is ​​calculated using the cross-modal association network to quantify the synergistic effect and extract audience behavior features and energy consumption features; (3) Intelligent decision-making and execution steps: The intelligent decision analysis module generates scheduling optimization and energy consumption control strategies, the execution and feedback module adjusts equipment parameters, the box office data is simultaneously uploaded to the blockchain for traceability, and the results are displayed through a visual interface.

[0011] Preferably, in step (2), the node weight update frequency of the cross-modal association network is a preset time interval, and the update basis includes the real-time occupancy rate of the cinema, the time interval between screenings, and the physical distance of the cinema.

[0012] Preferably, in step (3), the overflow coefficient is the ratio of the number of viewers entering the B hall within a preset time after the A hall ends to the total number of viewers in the A hall. When the ratio exceeds a preset threshold, a suggestion for binding the filming time slots of the A and B halls is generated.

[0013] Preferably, in step (3), the multi-objective optimization model achieves a balance between energy consumption and box office revenue in the following ways: when the occupancy rate is lower than a preset ratio, the power of unnecessary equipment is reduced; when the occupancy rate is higher than a preset ratio, the efficiency of environmental control equipment is improved, while ensuring the viewing experience.

[0014] Preferably, in step (3), the audience focus determination index includes: the gyroscope swing frequency is lower than a preset threshold, the number of times the screen is lit is less than a preset number, and the ambient light sensor data is stable. When the above indexes are met, the audience is determined to be highly focused.

[0015] The technical effects and advantages of the real-time box office revenue monitoring system and method for cinemas based on multi-source data fusion of the present invention are as follows: 1. This invention enhances the comprehensiveness and accuracy of box office monitoring. By integrating data from internal cinema sensors, ticketing data, external parking lot data, and user-authorized behavioral data through a multi-source data acquisition module, it overcomes the limitations of traditional single-source ticketing data and achieves comprehensive coverage of factors influencing box office revenue. Combined with the cross-modal correlation network of the data fusion processing module, it quantifies the spatiotemporal synergy of the cinema and audience behavioral characteristics, upgrading box office monitoring from "outcome statistics" to "process tracing + dynamic prediction," thus improving the accuracy of data interpretation.

[0016] 2. This invention optimizes screening strategies and resource utilization efficiency. The intelligent decision analysis module calculates the spillover coefficient of the theater and the similarity of film types to explore the synergistic effect of multiple theaters, generate the optimal screening combination, and improve the overall occupancy rate. At the same time, it analyzes audience attention to infer film reputation and dynamically adjusts screening priorities to achieve precise matching between film resources and audience needs, solving the problems of traditional screening relying on experience and rigid resource allocation.

[0017] 3. This invention achieves a dynamic balance between energy consumption and box office revenue. The energy consumption optimization unit uses a multi-objective optimization model to link occupancy rate with equipment energy consumption, dynamically adjusting equipment parameters such as lighting and air conditioning to reduce energy consumption per unit of box office revenue while ensuring a good viewing experience. Compared to the traditional operating model with fixed equipment parameters, this significantly reduces energy waste, lowers cinema operating costs, and achieves the dual benefits of "increased revenue + cost reduction".

[0018] 4. This invention enhances data security and traceability. The execution and feedback module uses blockchain technology to store the hash value of box office data, forming an immutable audit chain to ensure the authenticity and reliability of box office data and facilitate multi-party auditing. At the same time, it protects user privacy through data anonymization and makes full use of behavioral data under the premise of compliance, balancing data value and privacy protection.

[0019] 5. This invention improves operational response speed and decision-making intelligence. The edge control unit executes optimization strategies in real time, and combined with a visual interactive interface, it realizes the instant presentation of monitoring results and adjustment suggestions, supporting managers to respond quickly to market changes. Compared with the traditional manual decision-making mode, it reduces human intervention errors and improves the timeliness and scientific nature of decisions such as scheduling and energy consumption control.

[0020] 6. This invention expands the application boundaries of box office monitoring. The system not only realizes real-time box office statistics, but also explores the potential of non-ticket revenue through functions such as audience movement analysis and secondary consumption correlation extraction. It provides data support for value-added businesses such as derivative sales and advertising, and promotes the transformation of cinemas from "single screening" to "comprehensive entertainment services". Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall system architecture of a cinema real-time box office revenue monitoring system and method based on multi-source data fusion proposed in this invention. Figure 2 This is a flowchart illustrating the steps of a real-time cinema box office revenue monitoring system and method based on multi-source data fusion proposed in this invention. Figure 3 This is a schematic diagram of the multi-source data acquisition sub-process of a cinema real-time box office revenue monitoring system and method based on multi-source data fusion proposed in this invention; Figure 4 This is a schematic diagram of the intelligent decision analysis sub-process of a cinema real-time box office revenue monitoring system and method based on multi-source data fusion proposed in this invention. Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0023] 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 a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. refer to Figures 1-4 This invention discloses a real-time box office revenue monitoring system and method for cinemas based on multi-source data fusion. It aims to achieve collaborative management of real-time box office monitoring, film scheduling optimization, energy consumption control, and audience feedback analysis by integrating multi-dimensional data from both inside and outside the cinema. The following detailed description, along with specific embodiments and comparative examples, illustrates the system composition, implementation steps, and technical effects. Example

[0024] This embodiment provides a real-time cinema box office revenue monitoring system and method based on multi-source data fusion, used for the deployment and operation of the multi-source data acquisition module. Specific implementation details include: Purpose of implementation: The test verifies the comprehensiveness and real-time acquisition capabilities of the multi-source data acquisition module for collecting data from both inside and outside the cinema, providing a foundation for subsequent data fusion and decision-making.

[0025] System Architecture: Multi-source data acquisition module: Consistent with the original deployment, it includes an IoT sensor network, ticketing data interface, third-party data interface, and user-authorized data unit. Specific device models and data types are as follows: IoT sensors: Infrared passenger flow counter (HX-IR200), seat pressure sensor (FSR402), omnidirectional microphone array (MA-8), UWB positioning base station (DW1000), smart meter (DL / T645), ambient light sensor (BH1750); Interface data includes: showtime / ticket information from the ticketing system (China Film V5.0), parking space data from the parking management platform (Smart Parking V2.0), and data from user-authorized mobile phone sensors (gyroscope / screen status / ambient light).

[0026] Data fusion processing module: Hardware deployment: Edge computing server (model: Dell PowerEdge R750) is used, equipped with NVIDIA A30 GPU for parallel computing; Feature preprocessing unit: Runs the Python sklearn library to implement the ReliefF feature selection algorithm, selects 20 key features (such as screen spacing, occupancy rate, film type encoding, etc.) from 200+ original features, and generates high-dimensional feature vectors; Multimodal fusion unit: Deploys a cross-modal association network (built on the PyTorch framework, including 3 graph attention layers), and dynamically outputs a theater association matrix after inputting multi-source data (matrix dimension = number of theaters × number of theaters), with node weights updated every 5 minutes; Behavioral Feature Extraction Unit: MFCC features are extracted from acoustic data using the librosa library, and a departure movement heatmap is generated by combining UWB positioning coordinates (visualized using Matplotlib). Features such as "areas with a stay duration of >5 minutes" and "secondary consumption point arrival rate" are extracted.

[0027] Intelligent Decision Analysis Module: Theater Collaborative Analysis Unit: Runs a C++-written overflow coefficient calculation module to count the number of viewers entering theater B within 15 minutes after theater A ends, and automatically generates screening binding suggestions (triggered when overflow coefficient > 10%). Attention Analysis Unit: Based on TensorFlow, an LSTM model is built. Input features such as the oscillation frequency of the phone's gyroscope (<0.5 times / second) and the number of times the screen is turned on (<3 times / scene) are input. The output is an attention score of 0-100 points. When the score is >80 points, it is marked as a "high-potential film with good word-of-mouth". Energy consumption optimization unit: Deploys a multi-objective optimization model (NSGA-II algorithm) with the objectives of "lowest energy consumption per unit of box office" and "audience satisfaction > 4.5 points", and outputs dynamic adjustment parameters for air conditioning temperature (22-26℃) and lighting power (60%-100%).

[0028] Execution and Feedback Module: Edge control unit: Connects to the cinema PLC controller via Modbus protocol to adjust projector brightness (100-300cd / m²), air conditioning fan speed (level 1-3), and lighting switch status in real time; Data traceability unit: Adopting the Hyperledger Fabric blockchain framework, the box office data hash value is written to a block every 30 seconds (block size ≤ 1MB), supporting cinemas, film producers, and regulators to query and audit via API; Visual Interaction Unit: Develop a web-based monitoring platform (Vue.js + ECharts) to display real-time box office heatmaps and a list of film scheduling optimization suggestions. It also supports managers in adjusting strategies via voice commands, with a command response time of less than 2 seconds.

[0029] Implementation steps: (1) Equipment deployment: Install sensors and interfaces at the above locations and complete network configuration (LAN bandwidth 100Mbps, latency ≤50ms). (2) Data acquisition: After the system starts, each device automatically collects data and uploads it to the local server (using the MQTT protocol, with the transmission frequency consistent with the sampling frequency). (3) Data anonymization: Mobile phone sensor data is associated with anonymous IDs (unbound from ticket purchase accounts), and location tag data is automatically cleared after the audience leaves, retaining only the movement characteristics. Implementation effect

[0030] Data integrity: 98.7% of sensor data can be uploaded in real time (daily average loss rate <1.3%). Timeliness: Ticketing data and energy consumption data latency ≤ 10 seconds, location data latency ≤ 1 second; Coverage: Data collection is achieved across 6 cinemas, public areas, and 3 surrounding parking lots. Example

[0031] This embodiment provides a cinema real-time box office revenue monitoring system and method based on multi-source data fusion, used for cross-modal correlation analysis in the data fusion processing module. Specific implementation details include: Purpose of implementation: The data fusion processing module was verified to have the ability to fuse multi-source data, and the synergistic effect of the cinema was quantified through cross-modal association networks.

[0032] System Architecture: Data fusion processing module: Hardware: Dell PowerEdge R750 server (equipped with NVIDIA A30 GPU); Feature preprocessing unit: Runs the sklearn library to implement the ReliefF algorithm, selects 20 key features (screen room spacing, occupancy rate, etc.) from 200+ original features, and generates a 128-dimensional feature vector; Multimodal fusion unit: A 3-layer graph attention network is built based on PyTorch. After inputting multi-source data, it outputs a theater association matrix (dimension = 6×6, number of theaters is 6). The node weights are updated every 5 minutes (based on occupancy rate, time interval, and physical distance). Behavioral feature extraction unit: Extracts acoustic MFCC features using the librosa library, generates an exit movement heatmap by combining it with UWB positioning data, and extracts 5 types of behavioral features such as "stay time at secondary consumption points".

[0033] Collaboration with other modules: Multi-source data acquisition module: provides raw data input; Intelligent decision analysis module: Receives correlation matrix and behavioral characteristics; Execution and Feedback Module: Visualizes the fusion results (such as a heat map of theater associations).

[0034] Implementation steps: (1) Feature preprocessing: The raw data collected in Example 1 is cleaned (outliers, such as negative energy consumption data are removed), and the feature weights are calculated by the ReliefF algorithm, retaining the top 20 features by weight; (2) Calculation of cross-modal correlation matrix: Node weight formula: Weight = 0.4 × Occupancy rate coordination coefficient + 0.3 × Time interval coefficient + 0.3 × Distance coefficient; (Occupancy rate coordination coefficient = product of real-time occupancy rates of the two theaters, time interval coefficient = 1 / (1 + interval in minutes), distance coefficient = 1 / (1 + distance in meters)). Update frequency: once every 5 minutes; (3) Behavioral feature extraction: Spectral analysis of acoustic data was performed, and the SVM model (accuracy 92%) was used to identify areas with people. Combined with UWB positioning data, the average dwell time of the audience from the theater exit to the snack bar was statistically analyzed (3.2 minutes). Implementation effect

[0035] Association matrix precision: The association weight (0.72) of adjacent screening rooms (e.g., screening room 1 and screening room 2, 10 meters apart) is significantly higher than that of non-adjacent screening rooms (e.g., screening room 1 and screening room 6, 50 meters apart, weight 0.31). Seating capacity verification: The deviation rate between acoustic fingerprint analysis results and ticketing data is less than 3%, effectively identifying "ghost shows" (fake ticket sales). Example

[0036] This embodiment provides a real-time cinema box office revenue monitoring system and method based on multi-source data fusion, used for theater collaborative scheduling optimization in the intelligent decision analysis module. Specific implementation details include: Purpose of implementation: The effectiveness of the intelligent decision analysis module in optimizing film scheduling combinations through spillover coefficients was verified, thereby improving the box office revenue of multi-screen collaboration.

[0037] System Architecture: Intelligent Decision Analysis Module - Cinema Collaborative Analysis Unit: Hardware: Embedded computing unit (ARM Cortex-A72), supporting real-time computing; Algorithm layer: A C++ module is used to calculate the overflow coefficient, with the formula being "number of audience members entering Hall B within 15 minutes after the end of Hall A / total number of people in Hall A", with a preset threshold of 10%. Screening suggestion generation: When the overflow coefficient is greater than 10%, the system automatically generates a binding suggestion of "A Hall's end time + 15 minutes = B Hall's start time" based on the similarity of film types (comedy-animation similarity 0.7).

[0038] Collaboration with other modules: Multi-source data acquisition module: provides passenger flow and ticketing data for Halls A and B; Data fusion processing module: Outputs a cinema association matrix (enhanced collaborative analysis when weight ≥ 0.6); Execution and Feedback Module: Pushes scheduling suggestions to the Web platform, allowing administrators to confirm and implement them.

[0039] Implementation steps: (1) Overflow coefficient calculation: Set the preset duration = 15 minutes, the preset threshold = 10%, and the number of viewers who entered the theater 2 (animation) within 15 minutes after the end of the movie in theater 1 (comedy) is 42. The total number of people in theater 1 is 200. The overflow coefficient = 42 / 200 = 21% (>10%). (2) Screening suggestion generation: The system recommends that the end time of screening 1 in theater 1 is 19:00 and the start time of screening 2 in theater 2 is 19:15 (15 minutes apart), taking advantage of the close proximity of the two theaters (10 meters) to improve the convenience of the movement. (3) Execution feedback: The film scheduling suggestion is pushed to the cinema management backend, and after the administrator confirms it, it is synchronized to the ticketing system. Implementation effect

[0040] After the film schedule was adjusted, the occupancy rate of Theater 2 increased from 65% to 87%, and the combined daily box office revenue increased by 18,000 yuan. The proportion of viewers watching movies in other theaters increased from 8% to 21%, verifying the effectiveness of the spillover coefficient. Example

[0041] This embodiment provides a cinema real-time box office revenue monitoring system and method based on multi-source data fusion, used for energy consumption-box office balance control of the intelligent decision analysis module. Specific implementation details include: Purpose of implementation: The study verified the effectiveness of the multi-objective optimization model in reducing energy consumption per unit of box office revenue while ensuring the viewing experience.

[0042] System Architecture: Intelligent Decision Analysis Module - Energy Consumption Optimization Unit: Hardware: Edge controller (Siemens S7-1200); Algorithm layer: Deploy the NSGA-II multi-objective optimization model, with the objective function being "min unit box office energy consumption + max audience satisfaction"; Control rules: If occupancy is less than 30%, reduce lighting power to 60% and set air conditioning to 26°C. Occupancy rate 30%-80%: lighting power 80%, air conditioning 25℃; Occupancy rate >80%: Lighting 100%, air conditioning 24℃, projector heat dissipation power increased by 15%.

[0043] Collaboration with other modules: Multi-source data acquisition module: Real-time uploading of occupancy rate and energy consumption data; Data fusion processing module: Outputs the correlation characteristics of "occupancy rate - energy consumption"; Execution and Feedback Module: Controls the device via the Modbus protocol and synchronously uploads the execution results.

[0044] Implementation steps: (1) Parameter settings: preset occupancy rate thresholds of 30% (low) and 80% (high), and movie viewing experience indicators (light ≥ 150 lux, temperature 24±2℃). (2) Strategy generation: When the occupancy rate of Theater 3 is 25% (<30%), the model output is: the lighting power decreases from 100% to 60% (maintaining 180 lux), and the air conditioning temperature increases from 24℃ to 26℃; When the occupancy rate of Theater 3 is 85% (>80%), the model outputs: the air conditioner fan speed is increased from level 2 to level 3, and the cooling efficiency is increased by 20%; (3) Execution control: The edge control unit sends instructions via the Modbus protocol, and the response time for adjusting device parameters is <200ms. Implementation effect

[0045] Low attendance screenings: Energy consumption per unit of box office revenue decreased from 0.8 kWh / yuan to 0.64 kWh / yuan (a 20% reduction), and audience satisfaction rating was 4.6 / 5 (no significant decrease). High attendance rates: 100% environmental comfort rate (no audience complaints), equipment failure rate <0.5%. Example

[0046] This embodiment provides a cinema real-time box office revenue monitoring system and method based on multi-source data fusion, used for audience attention analysis in the intelligent decision analysis module. Specific implementation details include: Purpose of implementation: Verify the accuracy of inferring audience attention from mobile phone sensor data to provide a basis for film scheduling priorities.

[0047] System Architecture: Intelligent Decision Analysis Module - Attention Analysis Unit: Hardware: Cloud server (Alibaba Cloud ECS); Model layer: An LSTM model is built based on TensorFlow, with input features being "gyroscope oscillation frequency (<0.5 times / second), screen illumination frequency (<3 times / field), and ambient light fluctuation (<50 lux)"; Output: Focus score from 0 to 100. Films scoring ≥80 are marked as "high-potential films with positive word-of-mouth" and automatically recommended for increased screenings during prime time.

[0048] Collaboration with other modules: Multi-source data acquisition module: Provides user-authorized mobile phone sensor data; Data fusion processing module: Extracts behavioral feature vectors; Execution and Feedback Module: Displays ratings and scheduling suggestions on a visualization platform. Implementation effect

[0049] After adding two prime-time screenings (7:00 PM - 9:00 PM), the film's weekend box office increased by 32,000 yuan, and its Douban rating rose from 7.2 to 7.8, validating the accuracy of word-of-mouth predictions.

[0050] It provides a traditional cinema box office monitoring system, including the following: Traditional system components: Data collection: Relies solely on ticketing system data, without sensor or third-party data; Data processing: Daily box office figures are manually compiled; no data fusion analysis is performed. Decision-making method: Manual scheduling based on historical box office figures, without considering theater coordination; Energy consumption control: Equipment parameters are fixed (e.g., air conditioning at 24°C, lighting at 100%), and are unrelated to occupancy rate; Feedback mechanism: Relies on a questionnaire survey after the event (sample size <5%), with data lag of more than 24 hours.

[0051] Traditional methods and steps (1) Data collection: The ticketing system exports box office data after closing each day; (2) Screening decision: The manager manually adjusts the screening schedule based on the box office revenue of the previous 3 days; (3) Equipment control: The electrician switches the equipment on and off at fixed times every day, with the parameters remaining unchanged; (4) Problem investigation: If abnormalities are found in the box office (such as ticket theft), the surveillance video needs to be manually checked, which takes ≥24 hours.

[0052] Comparison results: See Table 1 for comparison details.

[0053] Table 1. Comparison of parameters between the present invention and the conventional solution of Comparative Example 1.

[0054] Compared with Examples 1-5 and Comparative Example 1, the present invention and the conventional solution are compared as follows: Data acquisition capabilities: The systems in Examples 1-5 achieve comprehensive data coverage through multi-source data acquisition modules: an IoT sensor network (infrared counters, pressure sensors, etc.) collects real-time data such as theater occupancy and energy consumption; a ticketing interface obtains ticket sales information; a third-party interface accesses parking lot data; and a user-authorized unit collects mobile phone sensor data (gyroscope, screen status, etc.), forming a three-dimensional data system of "internal operation + external environment + user behavior." In contrast, the traditional system in Example 1 relies solely on ticket sales data from the ticketing system, resulting in a single data dimension and an inability to capture implicit information such as audience movement and equipment energy consumption.

[0055] Data fusion and analysis capabilities: Examples 2-5 utilize the cross-modal association network of the data fusion processing module to transform multi-source data into decision features such as theater association matrices and spillover coefficients, thereby quantifying the synergistic effect of theaters (e.g., the association weight of adjacent theaters increased by 32% in Example 3). The intelligent decision-making module combines the attention model (Example 5) and the energy consumption optimization model (Example 4) to generate dynamic strategies. Comparative Example 1 lacks a data fusion mechanism and only performs simple statistics on ticket sales data, failing to uncover deeper patterns such as theater synergy and audience preferences.

[0056] Decision-making and execution efficiency: The intelligent decision analysis modules in Examples 3-5 can automatically calculate overflow coefficients (e.g., in Example 3, the occupancy rate increased by 22% after binding comedy films and animation films in the theater), dynamically adjust equipment parameters based on the occupancy rate (e.g., energy consumption decreased by 23% in Example 4), and implement strategies in real time through the execution module. Comparative Example 1 relies on manual scheduling, with rigid scheduling logic (e.g., fixed screening intervals) and constant equipment parameters (e.g., air conditioning at 24°C), making it unable to respond to real-time changes in occupancy rate, leading to resource waste.

[0057] Energy consumption and cost control: The energy consumption optimization unit in Example 4 achieves a balance between box office revenue and energy consumption through a multi-objective model. When the occupancy rate is below 30%, it reduces the power of lighting and air conditioning, and when it is above 80%, it improves equipment efficiency. The energy consumption per box office is reduced by 20%-30% compared to Comparative Example 1. The equipment parameters of Comparative Example 1 are fixed, and it maintains full-load operation regardless of the occupancy rate. For example, when the occupancy rate of a certain cinema is 25%, the energy consumption is still the same as when it is full, and the operating cost is significantly higher.

[0058] Feedback and optimization speed: Examples 1-5 utilize real-time sensor data (such as acoustic fingerprints and mobile phone gyroscopes) to achieve dynamic feedback, providing real-time alerts for abnormal situations (such as ticket theft); audience attention data (Example 5) can quickly infer film reviews and support scheduling adjustments. Comparative Example 1 relies on post-event questionnaires, resulting in delayed feedback (usually over 24 hours) and a sample size of less than 10%, making it difficult to guide real-time operations.

[0059] Overall benefits: The systems in Examples 1-5 achieve a dual improvement in box office revenue and efficiency through multi-source fusion and intelligent decision-making: the combined occupancy rate of the theaters increases by 18%-25% compared to Comparative Example 1, and daily box office revenue increases by 18,000-32,000 yuan; energy consumption costs are reduced by more than 20%, and the response time for abnormal issues is shortened from 24 hours to real-time. Comparative Example 1, due to its limited data and delayed decision-making, suffers from significant disadvantages in both box office revenue growth and cost control.

[0060] In summary, the systems in Examples 1-5, through multi-source data fusion, intelligent analysis, and dynamic execution, solve the problems of single data, rigid decision-making, and excessive energy consumption in traditional systems, highlighting the technical advantages of "multi-source fusion + intelligent optimization" in cinema box office monitoring.

[0061] The above embodiments can be implemented in whole or in part by software, hardware, firmware or other arbitrary combinations. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0062] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0063] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0064] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of protection of the claims.

[0065] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A real-time cinema box office revenue monitoring system based on multi-source data fusion, characterized in that, include: Multi-source data acquisition module, used to collect multi-dimensional data from inside and outside the cinema; The data fusion processing module is used to fuse and extract features from the collected multi-source data, and quantify the synergistic effect of the cinema in terms of time interval and physical distance through cross-modal association network; The intelligent decision analysis module is used to generate box office monitoring and optimization strategies based on fused data, including calculating the spillover coefficient between theaters, analyzing audience focus, and balancing energy consumption and box office revenue through a multi-objective optimization model. The execution and feedback module is used to execute optimization strategies and enable data blockchain traceability and visualization.

2. The cinema real-time box office revenue monitoring system based on multi-source data fusion as described in claim 1, characterized in that, The multi-source data acquisition module includes: The Internet of Things (IoT) sensor network is deployed in various theaters and public areas of the cinema, including an infrared passenger flow counter for collecting occupancy data, a seat pressure sensor for detecting seat status, an acoustic acquisition device for collecting ambient sound characteristics, a positioning device for recording audience movement, a smart meter for calculating energy consumption, and an ambient light sensor for monitoring light intensity. The ticketing data interface connects to the cinema's ticketing system to obtain showtimes, seat availability, film types, and ticket prices. Third-party data interfaces are used to connect with the parking management system around the cinema to obtain parking space occupancy data, vehicle type distribution, and parking duration. The user-authorized data unit acquires user-authorized mobile phone sensor data through a lightweight mini-program. The data collection period is limited to the movie viewing process and includes gyroscope oscillation frequency, screen illumination frequency, and ambient light sensor change data.

3. The cinema real-time box office revenue monitoring system based on multi-source data fusion as described in claim 1, characterized in that, The data fusion processing module includes: The feature preprocessing unit uses a feature selection algorithm to clean and filter the raw data, generating a high-dimensional feature vector that includes time dimension, spatial dimension, film type, audience behavior and energy consumption. The multimodal fusion unit dynamically calculates the cinema association matrix through a cross-modal association network. The weights of the matrix nodes are dynamically updated based on the real-time occupancy rate of the cinema, time intervals, and physical distances. The behavioral feature extraction unit analyzes environmental sound features through acoustic fingerprinting to verify the actual number of people seated, and generates a heat map of audience departure routes by combining location data, extracting features such as departure dwell time and recurrence correlation.

4. The cinema real-time box office revenue monitoring system based on multi-source data fusion as described in claim 1, characterized in that, The intelligent decision analysis module includes: The theater collaboration analysis unit calculates the spillover coefficient to quantify the impact of audience leaving theater A on the occupancy rate of theater B, and generates screening combination suggestions based on film type similarity. The attention analysis unit builds an attention-box office correlation model based on mobile phone sensor data. It infers the audience's attention by using the gyroscope swing frequency and the number of times the screen is turned on, and predicts the word-of-mouth spread of the film. The energy consumption optimization unit uses a multi-objective optimization model to link occupancy rate and energy consumption data, and dynamically adjusts equipment operating parameters based on occupancy rate to optimize energy consumption per unit of box office.

5. The cinema real-time box office revenue monitoring system based on multi-source data fusion as described in claim 1, characterized in that, The execution and feedback module includes: The edge control unit adjusts the parameters of the theater lighting, air conditioning and projection equipment in real time based on the dynamic equipment control strategy output by the intelligent decision analysis module. The data traceability unit uses blockchain technology to store real-time box office data hash values, forming an immutable audit chain that supports multi-party queries; The visual interactive unit generates box office monitoring heatmaps and film scheduling optimization suggestions reports, allowing managers to adjust strategies and receive feedback on execution results via natural language commands.

6. A method for real-time cinema box office revenue monitoring based on multi-source data fusion, characterized in that, Includes the following steps: (1) Multi-source data acquisition steps: Collect internal and external data of the cinema through the multi-source data acquisition module, including data such as occupancy rate and energy consumption collected by IoT sensor network, ticketing information obtained by ticketing data interface, parking lot data obtained by third-party data interface, and mobile phone sensor data authorized by users, and de-identify all data; (2) Data fusion and feature extraction steps: The collected data is fused through the data fusion processing module, and the theater association matrix is ​​calculated using the cross-modal association network to quantify the synergistic effect and extract audience behavior features and energy consumption features; (3) Intelligent decision-making and execution steps: The intelligent decision analysis module generates scheduling optimization and energy consumption control strategies, the execution and feedback module adjusts equipment parameters, the box office data is simultaneously uploaded to the blockchain for traceability, and the results are displayed through a visual interface.

7. The method for real-time cinema box office revenue monitoring based on multi-source data fusion as described in claim 6, characterized in that, In step (2), the node weight update frequency of the cross-modal association network is a preset time interval, and the update basis includes the real-time occupancy rate of the theater, the time interval between screenings, and the physical distance of the theater.

8. The method for real-time cinema box office revenue monitoring based on multi-source data fusion as described in claim 6, characterized in that, In step (3), the overflow coefficient is the ratio of the number of viewers entering the B hall within a preset time after the A hall ends to the total number of viewers in the A hall. When the ratio exceeds a preset threshold, a suggested time slot binding for the A and B halls is generated.

9. The method for real-time cinema box office revenue monitoring based on multi-source data fusion as described in claim 6, characterized in that, In step (3), the multi-objective optimization model achieves a balance between energy consumption and box office revenue in the following ways: when the occupancy rate is lower than the preset ratio, the power of unnecessary equipment is reduced; when the occupancy rate is higher than the preset ratio, the efficiency of environmental control equipment is improved, while ensuring the viewing experience.

10. The method for real-time cinema box office revenue monitoring based on multi-source data fusion as described in claim 6, characterized in that, In step (3), the audience focus determination index includes: the gyroscope swing frequency is lower than a preset threshold, the number of times the screen is lit is less than a preset number, and the ambient light sensor data is stable. When the above indexes are met, the audience is judged to be highly focused.

Images