Intelligent operation management system for entertainment hall based on multi-source data fusion

Abstract

The invention discloses an entertainment hall intelligent operation management system based on multi-source data fusion, which relates to the technical field of multi-source data analysis, and comprises a data acquisition component, an operation data processing component and an operation optimization component, and the operation optimization component comprises an interaction judgment module, a space value judgment module and a pricing judgment module. Relevant data, including user exploration data, spatial value data and pricing influence data, during operation of an entertainment hall are acquired through a data acquisition component, the acquired data are input into an operation data processing component, the data are cleaned through the operation data processing component, and the cleaned data are input into an operation optimization component. According to the invention, deep fusion and value mining of multi-source data are realized, experience-driven operation of an entertainment hall is changed into data-driven operation, user experience and operation efficiency are effectively improved, and a set of landing technical scheme is provided for intelligent operation of an offline entertainment scene.

Description

Technical Field

[0001] This invention relates to the field of multi-source data analysis technology, specifically to an intelligent operation and management system for entertainment halls based on multi-source data fusion. Background Technology

[0002] With the refined development of offline entertainment consumption scenarios, entertainment halls have become composite consumption carriers that integrate digital interaction, spatial experience, and social needs. Their operation logic has shifted from "resource supply" to "user demand-driven". The system management logic of such scenarios involves the cross-correlation of multi-source heterogeneous data. At the user level, it is necessary to integrate operational behavior, preference characteristics, etc. At the regional level, it is also necessary to collect relevant parameters such as user density and equipment utilization rate in real time. At the same time, at the scenario level, it is also necessary to synchronize dynamic information such as time fluctuations, activity popularity, and real-time operational pressure. These data are stored in different systems, such as user management system, equipment monitoring system, and customer flow statistics system, which are difficult to unify and integrate, making it difficult to deeply mine the value of the data.

[0003] Furthermore, with the continuous development of society, the operational needs of entertainment venues are also showing dynamic, personalized, and precise characteristics. On the one hand, user needs change rapidly with time and scenarios, requiring operational decisions to have real-time responsiveness. On the other hand, the value of regional resources also needs to be dynamically evaluated in conjunction with user behavior, rather than relying on static spatial layout or equipment configuration. Traditional operating models often rely on manual experience or single-dimensional data, such as adjusting equipment positions solely through people flow statistics. This makes it difficult to achieve collaborative analysis of multi-source data and to match the complex operational needs of entertainment venues. Therefore, it is urgent to build an intelligent operation management system based on multi-source data fusion to achieve data-driven precise decision-making and efficient operation. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent operation and management system for entertainment halls based on multi-source data fusion, which solves the problems mentioned in the background art above.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent operation and management system for entertainment halls based on multi-source data fusion, comprising a data acquisition component, an operation data processing component, and an operation optimization component. The data acquisition component acquires relevant data during the operation of the entertainment hall, including user exploration data, space value data, and pricing impact data, and inputs the acquired data into the operation data processing component. The operation data processing component cleans the data, and after cleaning, the data is input into the operation optimization component.

[0006] The operation optimization components:

[0007] Based on the frequency of user device i usage, device operation depth coefficient, average device operation depth coefficient of all users, and total number of devices used by users in the user exploration data, the effective user exploration index is output, and the real high-value users are identified through the effective user exploration index.

[0008] Based on the real-time number of users in region j, the area of ​​region j, the average dwell time of users in region j, the average dwell time of all regions, the content update response rate of region j, the average content response rate of all regions, and the total number of entertainment hall regions in the spatial value data, the spatial value gradient is output, and the regional operation value is evaluated through the spatial value gradient.

[0009] Based on the maximum effective exploration index of all users on the day, the maximum spatial value gradient of all regions on the day, the impact coefficient of holidays and major events, and dynamic weights in the pricing impact data, a dynamic pricing coefficient is output, which provides a basis for the final pricing.

[0010] The operation management system also includes an optimization component, which adjusts the dynamic weights based on the pricing effect coefficient and the standardized correlation coefficient, and recalculates the dynamic pricing coefficients based on the iterative dynamic weights until the iterative convergence condition is met, thereby achieving an upgrade from static weights to adaptive weights.

[0011] Optionally, the operation optimization components include an interaction judgment module, a space value judgment module, and a pricing judgment module.

[0012] Optionally, the processing logic of the interaction judgment module is as follows: by multiplying the frequency ratio by the logarithm of the frequency ratio to base 2, summing them, and taking the absolute value, the diversity of user behavior is quantified, avoiding the use of the number of times of use in the traditional way to judge value. The ratio of the device operation depth coefficient to the average value of all users is multiplied by 1.2, and then 0.8 is added to convert the user's operation depth into an adjustable weight, ensuring that the user's effective exploration index is preferentially tilted towards users with deep operations. Finally, the result is magnified by 100 times for easy system storage, comparison, and display.

[0013] Optionally, the processing logic of the spatial value judgment module is as follows: for region j, the relative values ​​of user density and dwell time are multiplied by the relative value of content response rate, and the scale difference between regions is eliminated by the ratio of the relative average value to ensure that regions of different sizes can be directly compared. The average value of all regions is magnified by 100 times to standardize the regional value, which is convenient for sorting and resource allocation.

[0014] Optionally, the processing logic of the pricing judgment module is as follows: by multiplying the normalized user value by the weight coefficient, and then adding the normalized regional value by the weight coefficient, pricing deviations are avoided, such as users with high effective exploration index in low-value areas or users with low effective exploration index in high-value areas. The weight allocation makes the pricing more in line with the comprehensive value of the user and the scenario. The calculation result is then multiplied with the impact coefficient of holidays and major events and the weight coefficient to match the actual attributes of the scenario and avoid pricing from deviating from reality.

[0015] Optionally, the optimization component includes an iterative update module and a convergence module.

[0016] Optionally, the processing logic of the iterative update module is as follows: after taking the Pearson correlation coefficient between the user effective exploration index and the pricing effect coefficient as non-negative, normalize it to obtain the weight coefficient of the user effective exploration index; after taking the Pearson correlation coefficient between the spatial value gradient and the pricing effect coefficient as non-negative, normalize it to obtain the weight coefficient of the spatial value gradient; after taking the Pearson correlation coefficient between the impact coefficient of holidays and major events and the pricing effect coefficient as non-negative, normalize it to obtain the weight coefficient of the impact coefficient of holidays and major events. By adjusting the weight coefficients, negatively correlated factors are avoided from interfering with pricing, ensuring that the weights are tilted only towards effective factors.

[0017] Optionally, the processing logic of the convergence module is as follows:

[0018] The iteration stops when any of the following conditions are met:

[0019] A1: The number of iterations has reached 5;

[0020] A2: The rate of change of weights in adjacent iterations is less than 5%.

[0021] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0022] I. This invention achieves multi-dimensional quantification of user value by integrating information entropy and device operation depth coefficient correction terms. The information entropy parameter reflects the breadth and diversity of user behavior, while the device operation depth coefficient, combined with the number of devices, operation depth, and correction coefficient, accurately captures the user's exploration depth on a single device. The synergistic effect of the two avoids the limitations of single-dimensional evaluation and can comprehensively identify the value of different types of users, providing a precise basis for subsequent personalized operation decisions.

[0023] Second, this invention dynamically quantifies the real-time value of a region by comprehensively calculating the relative values ​​of real-time user density, dwell time, and content response rate. The user density parameter reflects the degree of user concentration in the region, the relative value of dwell time reflects the stickiness of users in the region, and the relative value of content response rate measures the attractiveness of the region's content to users. Under the combined effect of these three factors, the assessment of regional value shifts from static physical attributes to dynamic user behavior attributes, capturing the fluctuations and changes in regional value in real time, and providing data support for the optimal allocation of regional resources.

[0024] Third, this invention achieves dynamic pricing decisions based on the weighted fusion of user value index, regional value index, and scenario timeliness coefficient. The user value index reflects the comprehensive value of users, the regional value index reflects the real-time value of regions, and the scenario timeliness coefficient considers the impact of scenario timeliness. Furthermore, the dynamic weights can be iteratively optimized based on operational data. Under the overall effect, the value quantification results of multi-source data are transformed into directly applicable operational decisions, enabling pricing strategies to simultaneously take into account user needs, regional efficiency, and scenario specificity, thereby improving the accuracy and real-time nature of operational decisions. Attached Figure Description

[0025] Figure 1 This is a schematic diagram illustrating the computational principle of the operation optimization component of this invention;

[0026] Figure 2 This is a flowchart of the present invention. Detailed Implementation

[0027] 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.

[0028] Example:

[0029] Please see Figure 1 and Figure 2 This invention provides an intelligent operation management system for entertainment halls based on multi-source data fusion, including a data acquisition component, an operation data processing component, and an operation optimization component. The data acquisition component acquires relevant data during the operation of the entertainment hall and inputs the acquired data into the operation data processing component. The operation data processing component cleans the data and then inputs it into the operation optimization component.

[0030] The data acquisition component obtains relevant data on the operation of the entertainment hall, including user exploration data, space value data, and pricing impact data. User exploration data includes the frequency p of user interaction with device i, obtained by dividing the number of times a user interacts with a single device (number of game levels) by the total number of possible interactions with device i. i That is, p i =O i / T i ;

[0031] Among them O i T represents the number of operations performed by device i, specifically the number of checkpoints cleared. i This refers to the total number of operations performed on device i, which is a factory-preset parameter for the device, specifically indicating the total number of game levels.

[0032] It also includes the Device Operation Depth Coefficient (DOC), which quantifies the depth of user operations on a single device, taking into account both operation coverage and duration coverage. The calculation method is as follows:

[0033]

[0034] Where S i The operation duration of device i is collected by a timing sensor built into the device, such as an embedded clock module, S. max,i The longest operation time for device i, such as the longest time to complete a game, is the factory default parameter for the device. n is the total number of devices used by the user. It is obtained by associating the user identity through the user identification module of the device, reflecting the breadth of the user's exploration of the entertainment hall equipment, helping the system to determine whether the user is a wide-area exploration user, and supplementing the evaluation dimension of user value, through the device operation depth coefficient DOC.

[0035] It also includes the average device operation depth coefficient (DOC) for all users. avg ,pass The calculation is performed, where K is the total number of users within the statistical period, derived from store counts. (DOC) k Device operation depth coefficient for an individual user, and average device operation depth coefficient (DOC) for all users. avg As a baseline for user operation depth, it determines whether the operation depth of an individual user is higher or lower than the average level, providing a reference for subsequent calculations.

[0036] Spatial value data includes the real-time number of users N in region j, collected in real time by infrared sensors deployed within the region. j The physical footprint A of area j, obtained from the architectural floor plan of the entertainment hall, is a fixed parameter. j The duration of each user's stay is collected by RFID readers at the area's entrance and exit, and then... The calculated average dwell time T for users in region j j Through formula The average length of stay T in all areas was calculated. avg Where m is the total number of entertainment areas, determined by the entertainment venue operator based on fixed parameters set according to the site plan, using the formula CRR. j = (Number of user visits to region j after content update - Number of user visits to region j before content update) / Number of user visits to region j before content update. The content update response rate (CRR) for region j is calculated as follows: j User access volume is collected by user counters within the region, using the formula... The calculated average content response rate (CRR) for all regions avg .

[0037] The pricing impact data includes the impact coefficient HAC for holidays and major events, which is automatically matched by the system backend according to preset time period type rules.

[0038] The operation optimization component includes an interaction judgment module. This module outputs a user effective exploration index based on the frequency percentage of user device i, the device operation depth coefficient, the average device operation depth coefficient for all users, and the total number of devices used by users, all derived from user exploration data. This index identifies truly high-value users. Specifically, the processing logic of the interaction judgment module is as follows:

[0039]

[0040] UEI stands for User Effective Exploration Index, pi represents the frequency percentage of user interaction with device i, reflecting the distribution of user preferences across different devices. This helps the system identify users' core interest devices, providing a basis for subsequent precise user-interest-based operations. DOC is the Device Operation Depth Coefficient, identifying whether users only briefly try out devices, filtering out invalid trial behaviors, ensuring the system focuses on genuine high-value users, and avoiding wasting resources on users who only briefly interact with devices. avg The average device operation depth coefficient for all users serves as a benchmark for user operation depth, determining whether an individual user's operation depth is higher or lower than the average level. This makes the User Effective Exploration Index (UEI) more comparable, quantifies the user's true exploration depth of the device through the UEI, filters out invalid trial behaviors, and accurately identifies high-value user groups.

[0041] More specifically, the information entropy is calculated by multiplying the frequency percentage by the logarithm of the frequency percentage to base 2, summing the results, and taking the absolute value. Taking the absolute value avoids negative values ​​affecting subsequent weighting, quantifying the diversity of user behavior, and avoiding judging value solely by the number of uses in traditional methods. For example, users who frequently use a single device may have a lower exploration breadth than users who use multiple devices. The ratio of the device operation depth coefficient to the average of all users is multiplied by 1.2 to positively amplify the value of "deep operation users." Adding 0.8 constrains the lower limit of the correction term, avoiding excessive punishment for shallow users. This process corrects the user's operation depth, transforming it into an adjustable weight, ensuring that the effective exploration index prioritizes deep operation users. Finally, the result is amplified 100 times for easy system storage, comparison, and display.

[0042] By introducing information entropy, the distribution of user operations on different devices is transformed into a behavioral diversity indicator, avoiding the limitations of a single indicator. By using the device operation depth coefficient (DOC) as a correction term, the user's exploration depth on a single device is incorporated into the evaluation, achieving a dual quantification of breadth and depth, and enabling the accurate identification of the value of high-potential users.

[0043] The operation optimization component also includes a spatial value judgment module. This module outputs a spatial value gradient based on the spatial value data, including the real-time number of users in region j, the area of ​​region j, the average dwell time of users in region j, the average dwell time of all regions, the content update response rate of region j, the average content response rate of all regions, and the total number of entertainment hall areas. This spatial value gradient is used to evaluate the operational value of a region. Specifically, the processing logic of the spatial value judgment module is as follows:

[0044]

[0045] Where N j Let A be the real-time number of users in region j, representing the current number of users in region j at the statistical time. By quantifying the user density of a region, it reflects the region's immediate attractiveness to users. This can also provide a basis for the allocation of regional resources (such as cleaning and equipment maintenance priorities). j Let T be the physical area of ​​region j. By combining this with the number of users to calculate user density, the impact of regional area differences on the number of users can be eliminated, making the user concentration of regions of different sizes comparable. j The average dwell time of users in region j reflects the stickiness of the region to users. The longer the dwell time, the stronger the attraction of the region's content or devices to users. This can also provide a reference for updating regional content or adjusting devices.

[0046] T avgThe average dwell time across all regions is used as a baseline for regional stickiness. This helps determine whether the user dwell time in a single region is higher or lower than the average, providing a reference for adjusting the Spatial Value Gradient (SVG). (CRR) j The content update response rate (CRR) for region j quantifies the effectiveness of content updates in that region, reflecting user responsiveness to new content. This also provides a basis for regional content operation strategies (such as update frequency and content type). avg The average content response rate of all regions is used as a baseline for the regional content response rate. This is used to determine whether the content update effect of a single region is higher or lower than the average level, providing a reference for the correction of the spatial value gradient SVG. The total number of entertainment hall regions m ensures the completeness of the value assessment of the spatial value gradient SVG.

[0047] More specifically, for region j, the user density N j / A j Relative value of stay duration T j / T avg CRR (Content Response Rate) j / CRR avg By multiplying the values ​​and using the ratio of the relative average values, the size differences between regions are eliminated, ensuring that regions of different sizes can be directly compared. The value items of all regions are averaged and magnified by 100 times to standardize the regional values, which facilitates sorting and resource allocation.

[0048] By user density N j / A j Relative value of stay duration T j / T avg CRR (Content Response Rate) j / CRR avg Collaborative computing, with user density N j / A j Instead of static area metrics, this method can better reflect the user concentration in a region in real time compared to traditional methods. It also introduces the relative values ​​of dwell time and content response rate, incorporating user stickiness and interaction quality within the region into the evaluation. This shifts the quantification of regional value from static physical attributes to dynamic user behavior attributes, capturing regional value fluctuations in real time and solving the problem of the lack of dynamism in regional value quantification.

[0049] The maximum value of the User Effective Exploration Index (UEI) for all users on that day is obtained as UEEI. max The maximum value of the spatial value gradient (SVG) for all regions on that day is used to obtain the SVG. max , will UEI max With SVG maxThe operations optimization component also includes a pricing judgment module in the input pricing impact data. This module is based on the maximum effective exploration index (UEI) of all users on that day from the pricing impact data. max The maximum spatial value gradient (SVG) for all regions on that day. max The impact coefficients of holidays and major events (HAC) and dynamic weights α, β, and γ output a dynamic pricing coefficient (DPC). The DPC provides the basis for final pricing. Specifically, the processing logic of the pricing judgment module is as follows:

[0050]

[0051] HAC is the impact coefficient of holidays and major events, which is matched to the scenario of the day in real time. It is 1.0 for weekdays, 1.2 for ordinary holidays, 1.5 for major event periods, and 0.8 for special low periods. By using the impact coefficient of holidays and major events (HAC), dynamic pricing can be adapted to different time periods, so as to increase revenue during peak periods and drive traffic during off-peak periods, and balance revenue and user experience. α, β, and γ represent the weight ratio of User Effective Exploration Index (UEI), Spatial Value Gradient (SVG), and impact coefficient of holidays and major events (HAC) in the calculation of dynamic pricing coefficient (DPC), respectively, and satisfy the condition that the sum of α, β, and γ is 1. The initial values ​​are set to 0.4, 0.4, and 0.2, respectively.

[0052] Considering that users are the main consumers, the User Effective Exploration Index (UEI) quantifies the depth of user exploration and value potential, so it is used as one of the core weights to ensure that pricing is tilted towards high-value users. Region is the physical scenario of user consumption, and the Spatial Value Gradient (SVG) quantifies the user density, dwell time stickiness, and content attractiveness of the region. It is equally important as user value and together constitutes the "foundation" of pricing. Therefore, the initial weights of the two are equal. The impact coefficient of holidays and major events (HAC) is a time-sensitive supplementary factor. Its impact has short-term and fluctuating characteristics, so its initial weight is lower than that of users and regions to avoid time-sensitive factors from excessively interfering with the core pricing logic.

[0053] After normalizing the user value and multiplying it by the weight coefficient α, we add the product of normalizing the regional value and the weight coefficient β. This avoids pricing discrepancies between users with high User Effective Exploration Index (UEI) in low-value regions and users with low UEI in high-value regions. By allocating weights, we make pricing more in line with the comprehensive value of users and scenarios. The calculation result is then multiplied by the impact coefficient of holidays and major events (HAC) and the weight coefficient γ to match the actual attributes of the scenario and prevent pricing from deviating from reality.

[0054] The operations management system also includes optimization-focused components, including an iterative update module based on the pricing effectiveness coefficient (PEC). t and standardized correlation coefficient ωUEI ω SVG With ω HAC The dynamic weights α, β, and γ are adjusted. Specifically, the processing logic of the iterative update module is as follows:

[0055]

[0056] Actual revenue is collected through the POS system or mobile payment platform. The benchmark revenue is the expected revenue for the time period set by the operator. The high-value user conversion rate is calculated by dividing the number of high-value users in time period t by the total number of users in time period t. The average conversion rate is obtained by averaging the high-value user conversion rates across all time periods within the statistical period. t The pricing effectiveness coefficient is then calculated by combining the User Effective Exploration Index (UEI) with the pricing effectiveness coefficient (PEC). t The Pearson correlation coefficient is taken to be non-negative, and the spatial value gradient SVG and the pricing effect coefficient PEC are compared. t The Pearson correlation coefficient is taken to be non-negative, and the influence coefficient of holidays and major events (HAC) and the pricing effect coefficient (PEC) are compared. t The Pearson correlation coefficient is taken to be non-negative:

[0057] ω UEI =max(0,Corr(UEI t PEC t ))

[0058] ω SVG =max(0,Corr(SVG t PEC t ))

[0059] ω HAC =max(0,Corr(HAC t PEC t ))

[0060] After ensuring that the weights are tilted only towards effective factors, normalization is performed to obtain the weight coefficient ω of the user's effective exploration index. UEI The weighting coefficient ω of the spatial value gradient SVG The weighting coefficient ω of the impact coefficient of holidays and major events HAC ;

[0061] α k＋1 =ω UEI / (ω UEI +ω SVG +ω HAC )

[0062] β k＋1 =ω SVG / (ω UEI +ω SVG +ωHAC )

[0063] γ k＋1 =ω HAC / (ω UEI +ω SVG +ω HAC )

[0064] After normalization, the sum of the weight coefficients is automatically satisfied to be 1. By adjusting the weight coefficients, negatively correlated factors are avoided from interfering with pricing, ensuring that the weights are tilted only towards effective factors. The dynamic weights after iteration are input into the pricing judgment module, and the dynamic pricing coefficient is recalculated until the iteration convergence condition is met. The final dynamic pricing coefficient DPC is then output. The base price is multiplied by the dynamic pricing coefficient DPC. In this way, the system can be transformed from static rules to adaptive rules, which can fit the actual operation scenario without manual intervention.

[0065] It is worth noting that since HAC is a quantitative coefficient representing the attributes of holidays, events, and other time periods, it directly affects users' willingness to consume. For example, users have higher consumption demand during major events, and HAC is necessarily positively correlated with the pricing effectiveness coefficient PEC, which is revenue × conversion rate. Therefore, the weighting coefficient ω of the impact coefficient of holidays and major events is important. HAC At least one positive number is required. In other words, even in extreme cases where the User Effective Exploration Index (UEI) and Spatial Value Gradient (SVG) are not positively correlated with the Pricing Effect Coefficient (PEC), the positive correlation between the Holiday and Major Event Impact Coefficient (HAC) will ensure that ω is positive. HAC >0, meaning that the denominator is not 0 during the above normalization process.

[0066] Furthermore, the optimization-focused components also include a convergence module, whose processing logic stops iteration when any of the following conditions are met:

[0067] A1: The number of iterations has reached 5;

[0068] A2: The rate of change of weights in adjacent iterations is less than 5%, that is:

[0069]

[0070] Dynamic Pricing Coefficient (DPC) balances the impact of user value, regional value, and scenario timeliness through dynamic weights α, β, and γ, solving the problem of multi-factor coordination in traditional operation system decision-making. Furthermore, the weights can be iteratively optimized based on operational data, enabling operational strategies to simultaneously consider user needs, regional efficiency, and scenario specificity, thereby improving the accuracy and real-time nature of decision-making.

[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.

Claims

1. An intelligent operation and management system for entertainment halls based on multi-source data fusion, characterized in that: It includes a data acquisition component, an operational data processing component, and an operational optimization component. The data acquisition component acquires relevant data during the operation of the entertainment hall, including user exploration data, space value data, and pricing impact data. The collected data is then input into the operational data processing component, where the data is cleaned. After cleaning, the data is input into the operational optimization component. The operation optimization components: Based on the frequency of user device i usage, device operation depth coefficient, average device operation depth coefficient of all users, and total number of devices used by users in the user exploration data, the effective user exploration index is output, and the real high-value users are identified through the effective user exploration index. Based on the real-time number of users in region j, the area of ​​region j, the average dwell time of users in region j, the average dwell time of all regions, the content update response rate of region j, the average content response rate of all regions, and the total number of entertainment hall regions in the spatial value data, the spatial value gradient is output, and the regional operation value is evaluated through the spatial value gradient. Based on the maximum effective exploration index of all users on the day, the maximum spatial value gradient of all regions on the day, the impact coefficient of holidays and major events, and dynamic weights in the pricing impact data, a dynamic pricing coefficient is output, which provides a basis for the final pricing. The operation management system also includes an optimization component, which adjusts the dynamic weights based on the pricing effect coefficient and the standardized correlation coefficient, and recalculates the dynamic pricing coefficients based on the iterative dynamic weights until the iterative convergence condition is met, thereby achieving an upgrade from static weights to adaptive weights.

2. The intelligent operation management system for entertainment halls based on multi-source data fusion according to claim 1, characterized in that: The operation optimization components include an interaction judgment module, a space value judgment module, and a pricing judgment module.

3. The intelligent operation management system for entertainment halls based on multi-source data fusion according to claim 2, characterized in that: The processing logic of the interaction judgment module is as follows: By multiplying the frequency percentage by the logarithm of the frequency percentage to base 2, summing the results, and taking the absolute value, the diversity of user behavior is quantified, avoiding the use of the number of uses alone to judge value. The ratio of the device operation depth coefficient to the average of all users is multiplied by 1.2, and then 0.8 is added to convert the user's operation depth into an adjustable weight, ensuring that the user's effective exploration index is preferentially tilted towards users with deep operations. Finally, the result is magnified 100 times for easy system storage, comparison, and display.

4. The intelligent operation management system for entertainment halls based on multi-source data fusion according to claim 3, characterized in that: The processing logic of the spatial value judgment module is as follows: For region j, the relative values ​​of user density and dwell time are multiplied by the relative value of content response rate. The scale difference between regions is eliminated by the ratio of the relative average value, ensuring that regions of different sizes can be directly compared. The value items of all regions are averaged and magnified by 100 times to standardize the regional value, which facilitates sorting and resource allocation.

5. The intelligent operation management system for entertainment halls based on multi-source data fusion according to claim 4, characterized in that: The processing logic of the pricing determination module is as follows: By producting the normalized user value with the weighting coefficient, and then adding the normalized regional value with the weighting coefficient, pricing discrepancies are avoided, such as users with high effective exploration indices in low-value regions or users with low effective exploration indices in high-value regions. Through weighting, pricing is made more aligned with the comprehensive value of users and scenarios. The calculation result is then multiplied with the impact coefficients and weighting coefficients of holidays and major events to match the actual attributes of the scenario and prevent pricing from being detached from reality.

6. The intelligent operation management system for entertainment halls based on multi-source data fusion according to claim 5, characterized in that: The optimization-focused component includes an iterative update module and a convergence module.

7. The intelligent operation management system for entertainment halls based on multi-source data fusion according to claim 6, characterized in that: The processing logic of the iterative update module is as follows: After taking the non-negative Pearson correlation coefficient between the user effective exploration index and the pricing effect coefficient, normalization is performed to obtain the weight coefficient of the user effective exploration index. After taking the non-negative Pearson correlation coefficient between the spatial value gradient and the pricing effect coefficient, normalization is performed to obtain the weight coefficient of the spatial value gradient. After taking the non-negative Pearson correlation coefficient between the impact coefficient of holidays and major events and the pricing effect coefficient, normalization is performed to obtain the weight coefficient of the impact coefficient of holidays and major events. By adjusting the weight coefficients, negatively correlated factors are avoided from interfering with pricing, ensuring that the weights are tilted only towards effective factors.

8. The intelligent operation management system for entertainment halls based on multi-source data fusion according to claim 7, characterized in that: The processing logic of the convergence module is as follows: The iteration stops when any of the following conditions are met: A1: The number of iterations has reached 5; A2: The rate of change of weights in adjacent iterations is less than 5%.

Images