An interactive display control method

Abstract

The present application relates to the technical field of data processing, in particular to an interactive display control method, comprising: extracting user behaviors on each exhibition node to form a user behavior mapping set, performing interactive type identification on the user behavior mapping set, identifying interactive proportions under direct interaction and indirect interaction, determining commonality metrics between adjacent exhibition nodes according to the interactive proportions in different time periods, and generating an interactive frequency spectrum graph; filtering information of the exhibition nodes according to the exhibition nodes corresponding to the interactive frequency spectrum graph to generate an associated behavior set; in response to a region range where the current associated behavior set is located, performing aggregation analysis on each exhibition event with exhibition events of the exhibition nodes to obtain target event information; and performing parameter updating on the result of the reduction between the current target event information and adjacent target event information, and taking the updated target event information as output. The dynamicization and display efficiency of the display content are realized.

Description

An interactive display control method Technical Field

[0001] This invention relates to the field of data processing technology, specifically an interactive display control method. Background Technology

[0002] With the continuous development of digital technology, digital exhibition halls have gradually become an important platform for various exhibitions, product displays, and brand promotions. By leveraging technologies such as computer graphics and virtual reality, digital exhibition halls not only provide richer and more interactive display experiences but also optimize display content and improve display effectiveness through real-time data analysis. In particular, they require adaptive adjustments to pedestrian flow and user behavior, and the identification of spatial relationships under various interaction types to achieve interactive control of the digital exhibition hall.

[0003] For example, Chinese Patent Publication No. CN119781891A discloses a data interactive display method and system for digital exhibition halls, relating to the field of display content recommendation technology. The method includes: collecting historical display interaction behavior data from multiple display devices in the exhibition hall; extracting user browsing behavior preference features to determine multiple browsing behavior preference patterns; extracting browsing time period preference features and determining time period distribution patterns to generate a first interactive display strategy; collecting historical display environment data from multiple display devices and performing content environment correlation analysis to construct a content environment analysis model and generate a second interactive display strategy; fusing and generating a target interactive display strategy and optimizing the interactive display; acquiring real-time display interaction behavior data from display devices and performing user behavior collaborative impact analysis to generate multiple interactive collaborative optimization strategies for interactive display collaborative optimization.

[0004] For example, Chinese Patent Publication No. CN118838720A discloses a method for constructing a digital twin for data interactive display. This method includes: acquiring node data for each display node; obtaining the suppressed attention level, process information content, process simulation quantity, and system contribution ratio of each display node based on the node data; obtaining the model completion level of each display node based on the process information content, process simulation quantity, and system contribution ratio; obtaining the computational resource allocation weight of each display node based on the suppressed attention level and model completion level; and allocating computational resources to each display node based on the processor's computational resource configuration and the computational resource allocation weight of each display node.

[0005] Existing technologies describe using browsing preference time periods for interaction analysis, using the analyzed data as the basis for interaction display strategies, and further explain the interaction display process by describing the amount of information displayed at each node after assessing the attention level for each type of interaction. Existing technologies focus on the number of times users operate a device and the actual operation time, which tends to describe direct user interaction. However, in most exhibition hall scenarios, the main interaction is indirect, through direct observation. Therefore, it is necessary to identify and process user interest areas and display hotspots based on user data from both direct and indirect interactions to reduce the problem of abnormal behavior analysis caused by data bias. Summary of the Invention

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: an interactive display control method, including: S1, taking the location of the exhibition equipment as the exhibition node, extracting user behavior on each exhibition node, and forming a user behavior mapping set according to the time length corresponding to the user behavior, the user behavior mapping set including: number of times the equipment is used, single usage duration, user dwell time and walking time.

[0007] S2 identifies the interaction type of the user behavior mapping set, which includes direct interaction and indirect interaction. It identifies the proportion of interaction under direct and indirect interaction, determines the commonality metric between adjacent exhibition nodes based on the proportion of interaction in different time periods, and generates an interaction spectrum map.

[0008] S3. Based on the exhibition nodes corresponding to the interaction spectrum, information filtering is performed on the exhibition nodes using the commonality measure of each exhibition node at each time point to extract at least one target behavior information to generate a set of related behaviors.

[0009] S4 responds to the area range where the current set of associated behaviors is located, takes the content implemented by the exhibition node as the exhibition event, performs aggregate analysis on each exhibition event, and obtains the target event information.

[0010] S5 updates parameters based on the target event information for each time period, using the reduction result between the current target event information and the adjacent target event information, and outputs the updated target event information to complete the interactive adjustment of each exhibition node.

[0011] The beneficial effects of this invention are as follows: First, this invention maps extracted user behavior to the location of the exhibit, and performs quantitative analysis of user behavior through device usage frequency, single usage duration, user dwell time, and walking time, thereby identifying the relative quantitative values ​​of direct interaction and handover interaction for users in different locations, and realizing the processing of different interaction types.

[0012] Second, this invention analyzes the consistency of adjacent exhibition nodes by measuring the proportion of different types of interactions, thereby obtaining commonalities between adjacent exhibition nodes. This allows for the determination of the ratio of each exhibition node in the same time period of trend change and the frequency probability of user behavior parameters under the paths connected by different interaction types. These probability values ​​and the ratio of the same time period of trend change are used to form the temporal correlation of various direct and indirect interaction sub-type combination paths.

[0013] Third, this invention restricts direct and indirect interactions related to user behavior by adding spatial dimensions and information filtering to all exhibition nodes in the interaction spectrum diagram, using the content displayed on the exhibition nodes as a constraint, and updating the target event information after traversing the specification results under each time period, thereby obtaining the adjustment methods for multiple target event information and the relative content that can be adjusted and displayed subsequently. Attached Figure Description

[0014] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0015] Figure 1 is a flowchart illustrating an interactive display control method.

[0016] Figure 2 is a flowchart illustrating step S1 of an interactive display control method.

[0017] Figure 3 is a flowchart illustrating step S2 of an interactive display control method.

[0018] Figure 4 is a flowchart illustrating step S3 of an interactive display control method.

[0019] Figure 5 is a flowchart illustrating step S4 of an interactive display control method.

[0020] Figure 6 is a flowchart illustrating step S5 of an interactive display control method. Detailed Implementation

[0021] The embodiments of the present invention are described in detail below. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. Where specific techniques or conditions are not specified in the embodiments, they shall be performed in accordance with the techniques or conditions described in the literature in the art or in accordance with the product manual.

[0022] Referring to Figure 1, an interactive display control method includes: S1, taking the location of the exhibition equipment as the exhibition node, extracting user behavior on each exhibition node, and forming a user behavior mapping set according to the time length corresponding to the user behavior. The user behavior mapping set includes: number of times the device is used, duration of a single use, duration of user stay, and duration of walking.

[0023] S2 identifies the interaction type of the user behavior mapping set, which includes direct interaction and indirect interaction. It identifies the proportion of interaction under direct and indirect interaction, determines the commonality metric between adjacent exhibition nodes based on the proportion of interaction in different time periods, and generates an interaction spectrum map.

[0024] S3. Based on the exhibition nodes corresponding to the interaction spectrum, information filtering is performed on the exhibition nodes using the commonality measure of each exhibition node at each time point to extract at least one target behavior information to generate a set of related behaviors.

[0025] S4 responds to the area range where the current associated behavior set is located, takes the content implemented by the exhibition node as the exhibition event, performs aggregate analysis on each exhibition event to obtain target event information, and makes each target event information correspond to a target behavior information.

[0026] S5 updates parameters based on the target event information for each time period, using the reduction result between the current target event information and the adjacent target event information, and outputs the updated target event information to complete the interactive adjustment of each exhibition node.

[0027] The user behavior set includes multiple time length values, such as device usage count, single usage duration, user dwell time, and walking time. By using the described time lengths, the content of the acquired user behavior is mapped and associated with multiple sets of behavior time lengths corresponding to the interactive display, so as to complete the degree of user interaction with the current exhibition content in different exhibition methods and areas.

[0028] When generating aggregated data related to user behavior, it is also necessary to identify the location coordinates of the user's exhibition node, as well as the content and relative position of the content displayed on that exhibition node, in order to identify the real-time distance between the user and the actual exhibition content and the user's orientation, and to identify which exhibition method is more inclined to direct user interaction under each exhibition method.

[0029] As shown in Figure 2, step S1 is implemented by: using the device usage status at any exhibition node as the basic behavioral condition, mapping the basic behavioral condition to the number of times the device is used, the duration of a single use, the user's dwell time, and the walking time, and determining the changing trend of the user's corresponding time under various device usage conditions. The aforementioned device usage status indicates the corresponding state of the device being exhibited, idle, or faulty.

[0030] Set the ratio of single usage duration to device usage frequency as the first row of values; set the ratio of user dwell time to walking time as the second row of values.

[0031] By using the ratio of changes in the first and second row values ​​over different time periods, the first row values ​​are mapped to the second row values ​​to form a user behavior mapping set.

[0032] The first row of numerical values ​​quantifies the average level of user engagement with a single device by comparing the duration of a single use to the number of times the device is used. A high ratio indicates that the user tends to engage deeply, such as operating a single device for an extended period of time, while a low ratio may reflect superficial browsing or rapid switching behavior.

[0033] The first row of values ​​is then correlated with the device usage to determine the relative status of the current exhibition node in exhibition, idle and fault scenarios, identify whether some devices are poorly positioned or the content is outdated, and whether there are any sudden changes in the ratio of the final output value.

[0034] The second set of numerical values ​​distinguishes user behavior types by the ratio of dwell time to walking distance. For example, high dwell time and low walking distance indicate focused exploration, while low dwell time and high walking distance indicate fast browsing. The identified situations are used to evaluate the relative situation of each exhibition node. After mapping the two values ​​to the exhibition nodes, a three-dimensional profile of user behavior can be obtained to describe the user's behavior in different scenarios.

[0035] By analyzing the changes in the two values ​​over time series, we can obtain the shift in user interest between different exhibition nodes at different time periods. Then, based on the data obtained from the user behavior mapping set, we can set the exhibition format to provide exhibition methods tailored to different users.

[0036] Preferably, since large crowds are common in scenarios such as exhibitions, it is difficult to calculate the duration of user stay and walking. In this case, the number of people at each exhibition node can be counted in a unit time period. The number of people staying at the current exhibition node in a unit time period is used as their stay time, and the number of people moving from the current exhibition node in a unit time period is used to identify their walking time. That is, the stay time and walking time are obtained by dividing the unit time by the number of people at the current exhibition node and the number of people moving from the current exhibition node, respectively.

[0037] When completing the user behavior mapping set, the implementation method of mapping the first behavior value to the second behavior value also includes: based on the obtained first behavior value and second behavior value, cross-grouping the user behavior, setting the user behavior matrix, and obtaining at least one user type.

[0038] The user behavior matrix is ​​divided according to the dimension corresponding to the time period, and a trend behavior label is set for each element in the user behavior matrix.

[0039] Based on the current equipment usage at the exhibition nodes, the user behavior matrix is ​​updated, and the updated data is output as a set of user behavior maps.

[0040] The aforementioned cross-clustering method uses the current first and second row values ​​as cross-clustering methods. It shuffles and combines these two values, comparing them with relevant data in the historical database to determine the user type corresponding to each value. When setting up the user behavior matrix, the sorted first and second row values ​​are used as rows and columns respectively, in time period order, to form the user behavior matrix. When there are multiple exhibition nodes being identified and processed, the resulting user behavior matrix can be formed by combining the first and second row values ​​using pre-labeled mapping, or it can be divided into multiple user behavior matrices for each exhibition node.

[0041] Preferably, the method of obtaining user types also includes hierarchical clustering. First, clustering is performed using the first row of numerical values. Then, the part after clustering the first row of numerical values ​​is clustered again using the second row of numerical values. That is, the first row of numerical values ​​and the second row of numerical values ​​are cross-combined and clustered sequentially to divide into multiple smaller clusters. The user types of the multiple small clusters after the final clustering are used to complete the quantitative processing of each user behavior in the current exhibition node.

[0042] When dividing the data by time, the ratio of changes in the values ​​in the first and second rows over multiple time periods is used as the trend behavior label. The exhibition nodes represented by each element in the user behavior matrix are further explained. For example, if some user behaviors only occur during peak user periods, the ratio of changes identified at this time is converted into a label and embedded into the user behavior matrix to further restrict and explain the corresponding elements in the user behavior matrix.

[0043] As for combining equipment usage, the status of the corresponding exhibition node in exhibition, idle and faulty is obtained in real time and synchronized to the user behavior matrix to illustrate the update of the user behavior matrix under different conditions, and the updated data is output as a user behavior mapping set.

[0044] Preferably, in order to obtain more user behavior data, the processing unit time can be 30 minutes to 1 hour to obtain a relatively sufficient set of behavioral data; if the current period is a peak time for browsing, a time interval of 5 to 15 minutes is used to update the user behavior data in real time within the corresponding time period.

[0045] In one embodiment of the invention, direct interaction and indirect interaction are used to indicate whether the current user is directly experiencing the interactive exhibition device or interacting through a model of special performances and scene displays.

[0046] Direct interaction includes interactive methods such as operational displays, interactive displays, participatory displays, demonstration displays, and experimental displays, while indirect interaction is suitable for model scene displays and special performance displays.

[0047] Among these, hands-on demonstrations involve direct contact with and disassembly of exhibits; interactive demonstrations exchange information through dialogue, actions, and choices between visitors and exhibits, providing video viewing, information, and results based on visitor selections; participatory demonstrations guide visitors through the process, often involving multiple choices and guided by visitor responses or participation; demonstrations are typically in the form of experiential learning activities such as crafts and workshops; and experimental demonstrations showcase the process of demonstrating principles through experiments, primarily illustrating interactive methods of scientific experience.

[0048] As for the model scene display, it uses small-scale model scenes or other model-making methods to achieve interaction through observation of the model's performance; special performance displays explain the expression forms of various media such as videos and models, allowing direct observation of their performance forms. The indirect interaction content, compared to the direct interaction content, only includes the observation part and does not include the part of making operation choices to achieve interaction. The exhibition content is set up in this way by distinguishing the frequency of user interaction under different methods through direct and indirect forms.

[0049] The proportion of direct interaction can be obtained from the number of times the device is used in the user's actual operation record. The ratio of the number of times the device is used in the current time period to the total number of times the device is used in the total time period is used to illustrate the proportion of interaction. The proportion of indirect interaction can be described based on the walking time and dwell time of the user at the exhibition node of indirect interaction. It is judged by the average walking time and dwell time of the user in a unit time period. The average walking time and dwell time of the current user are divided by the sum of the average walking time and dwell time of the user in all time periods. The sum of the ratios is used as the proportion of indirect interaction.

[0050] Compared to the user type obtained in step S1, the main focus of the type analysis in step S2 is whether the current user behavior is in direct or indirect interaction. This type can be directly obtained based on the relevant devices and exhibition content set for the exhibition nodes. Then, for exhibition nodes with direct and indirect interactions, the user behaviors that have been initially processed in the user behavior mapping set are displayed. The proportion of the corresponding user behavior-related interactions is further shown in the form of stacked bar charts or area charts. The ratio of each time period to the total is used to illustrate the relative situation of direct and indirect interactions in the overall exhibition nodes.

[0051] As shown in Figure 3, the implementation of step S2 includes: S21, identifying the interaction type of the user behavior mapping set, which includes various interaction methods under direct and indirect interaction, marking the user behaviors of direct and indirect interaction, and sorting each user behavior in chronological order to form a behavior interaction sequence; this behavior interaction sequence will identify multiple exhibition nodes in a certain area, and the interaction ratio and relative user behavior collected by multiple exhibition nodes in each time period during the exhibition will be segmented and stored in the form of a time series.

[0052] S22, connect the exhibition nodes according to their locations to form a flow path, use the behavioral interaction sequence to perform path analysis on the flow path, identify the common metrics between adjacent exhibition nodes; generate an interaction spectrum map based on the common metrics between adjacent exhibition nodes.

[0053] When performing path analysis to obtain commonality metrics for the flow path, the locations of the exhibition nodes are connected to form a flow path to illustrate the path when the exhibition space moves. Then, the probability value and time interval are calculated based on the interaction ratio of two adjacent exhibition nodes. The calculated probability value and time interval are used as the commonality metric, and then the commonality metric is used as the content of the judgment. After that, according to the value of the commonality metric, an interaction spectrum diagram related to interaction analysis is generated.

[0054] When implementing step S22, users' movement and interaction tendencies can be tracked using devices such as cameras placed in the venue. Alternatively, when calculating the interaction ratio of each exhibition node, each exhibition node can be used as the processing object. After forming an interaction ratio time series based on the changing trend of the interaction ratio over time, Gaussian distributions can be fitted to the interaction ratio time series of two adjacent exhibition nodes respectively. The probability values ​​after Gaussian distribution are calculated, and the relevant conditional probabilities are obtained. The time interval between the two exhibition nodes showing the same trend change is regarded as the time interval, which illustrates the commonality of two adjacent exhibition nodes in the same time period. This can reflect in which time period, user behavior at certain locations will change significantly.

[0055] The commonality measurement between adjacent exhibition nodes in step S22 is implemented by taking the interaction ratio of each exhibition node as the processing object, obtaining the interaction ratio time series of the corresponding interaction type, fitting a Gaussian distribution to the interaction ratio time series, calculating the probability value of each exhibition node after the Gaussian distribution, and obtaining the conditional probability of two adjacent exhibition nodes.

[0056] Determine whether the conditional probability values ​​of two adjacent exhibition nodes are greater than a preset conditional probability threshold. If they are, obtain the ratio of the time periods in which the two exhibition nodes are in the same trend change and output it as a commonality measure between adjacent exhibition nodes.

[0057] When calculating the conditional probability, assuming there are two exhibition nodes, A and B, their probability values ​​P(A) and P(B) are calculated using the Gaussian distribution formula based on the interaction proportion values ​​in the corresponding interaction proportion time series. For the combined probability value P(AB) of the two exhibition nodes, they need to be treated as a single node. When the two nodes are of the same type, the data of the two nodes are simply added together to describe the interaction proportion, and the corresponding probability value is calculated. If one of the two exhibition nodes belongs to a direct interaction type and the other to an indirect interaction type, the two exhibition nodes are treated as an object. The frequency of the data formed by the intersection of the first and second row values ​​in the user behavior mapping set is used. A time series is formed based on the proportion of this frequency to the historical data. The Gaussian distribution value of this time series is then used to represent the probability value represented by P(AB), thus obtaining the conditional probabilities of P(A|B) and P(B|A). Simultaneously, when determining the commonality measure, the conditional probability values ​​corresponding to P(A|B) and P(B|A) must both be greater than a preset conditional probability threshold. The preset conditional probability threshold is used to indicate the consistency of the calculation of the two probability values ​​and whether there is significant commonality. The value of the preset conditional probability threshold can be described based on the average value of the conditional probabilities when measuring commonality in historical data. Subsequently, a consistent trend indicates that two adjacent exhibition nodes are both in periods of increase and decrease in the interaction proportion time series. Dividing the length of this period by the total time indicates that the two exhibition nodes are experiencing the same change.

[0058] Preferably, when obtaining the behavioral interaction sequence, it is also necessary to use the subtypes of direct and indirect interactions as data when connecting the sequences to form a multi-level sequence.

[0059] The implementation of step S21 also includes: nesting multiple interaction types using subtypes of direct and indirect interaction, using the occurrence time of the nested subtypes of direct and indirect interaction as the sorting basis, and outputting the sorted multi-level sequence as a behavior interaction sequence.

[0060] For the output behavioral interaction sequence, the exhibition nodes corresponding to the subtypes of direct and indirect interaction are connected and combined to form a flow path.

[0061] At this point, it is necessary to further label the exhibition methods of direct and indirect interaction to determine how to conduct the exhibition in the current scenario and refine user behavior and exhibition events.

[0062] In one embodiment of the present invention, after processing adjacent exhibition nodes in the interactive spectrum graph, it is also necessary to identify and dynamically adjust the spatial influence matrix to optimize user behavior distribution, extract target behavior information, and further quantify the interaction between exhibition nodes to achieve dynamic spatial user behavior analysis. For example, whether adjacent nodes affect the exhibition status of other exhibition nodes, causing differences in parameters such as commonality metrics in their identification output. To minimize prediction error and maximize the consistency of behavior distribution, relative target behavior information is extracted, such as high-frequency paths, abnormal clustering areas, and repeated distributions after different exhibition method combinations when changes are consistent. Finally, the distribution relative to the overall space is completed.

[0063] As shown in Figure 4, the implementation of step S3 also includes: S31, using the commonality measure of any exhibition node in the interaction spectrum to track the distribution position of each exhibition node, forming a spatial influence matrix related to the position of the exhibition node; the spatial influence matrix is ​​used to take the value of the currently identified commonality measure as the first judgment condition. Since the commonality measure needs to obtain a part greater than the preset conditional probability threshold, the exhibition node obtained at this time is biased towards the core node relative to the overall node, mainly realizing the main path identification of user exhibition interaction behavior.

[0064] S32 sets the weights of the spatial influence matrix based on the distance values ​​and commonality metric values ​​between each exhibition node in the spatial influence matrix; and updates the spatial influence matrix according to the weights of the spatial influence matrix.

[0065] Let there be n exhibition nodes being processed at this time. The weights of the spatial influence matrix can be expressed as: ;in, The weights of the spatial influence matrix for exhibition nodes i and j are used to illustrate the weights of their commonality measure and distance value combination under spatial influence, emphasizing the correlation between location and exhibition node related probability values. This represents the distance between exhibition node i and exhibition node j. The distance value can be calculated using Euclidean distance. It is used to describe the distance between the equipment and interactive content deployed in the current exhibition scene. This distance value will be standardized so that the reciprocal of the distance can be kept in the range of 0-1. This represents the commonality metric between exhibition node i and exhibition node j. Since the commonality metric is obtained by representing the time ratio and conditional probability value, these two values ​​are standardized here to eliminate the dimensions. The commonality metric is then used as a separate vector, and its modulus is calculated to obtain the commonality metric at this time. At the same time, this commonality metric is normalized to a value range of 0-1 to prevent the problem of the spatial influence matrix weight ignoring spatial distance due to excessively large data. , These represent the weights of distance and commonality metrics, respectively, and their sum is 1. The weights can be set in the form of 0.5, 0.5, or to emphasize spatial distance in the form of 0.7 and 0.3.

[0066] After calculating the weights of the spatial influence matrix, we can know the weight values ​​of adjacent exhibition nodes after considering their relative spatial distance. This allows the spatial influence matrix to further consider the spatial dimensions after taking into account the correlation between adjacent exhibition nodes at different time periods. By combining these values, the calculated weights are used as the values ​​of the spatial influence matrix to update it. This makes the data dimensions considered by the spatial influence matrix more comprehensive, preventing the limitations of single-dimensional processing in exhibition scenarios and the difficulty in interpreting the magnitude of some numerical values.

[0067] S33, based on the spatial influence matrix, find the parameter combination that satisfies the minimum prediction error and the maximum distribution consistency, and output the corresponding parameter set as target behavior information.

[0068] When seeking the parameter combination that satisfies the minimum prediction error, it is necessary not only to compare the conditional probability and time period ratio in the commonality measure, but also to consider the spatial influence matrix that incorporates the actual spatial distance of the current exhibition node, and use the data that all satisfy the minimum error as a set of parameter combinations.

[0069] When finding the parameter set with the largest consistent distribution, it is necessary to use the intersection of the spatial influence matrices calculated at different time periods. The largest common subset after multiple intersections is used as the part to determine the consistency of the distribution. The exhibition nodes in this part obviously have high-frequency distance and abnormal clustering. Based on this part of the data, we can directly obtain the exhibition nodes with high-frequency similarity, and which subtypes of direct or indirect interaction these similar nodes belong to. This is convenient for filtering the area range where the set of related behaviors is located in subsequent exhibition event aggregation analysis.

[0070] The implementation of step S33 includes: obtaining the maximum common subset of the spatial influence matrix in multiple time periods, and using the data corresponding to the maximum common subset as the first parameter set; the output first parameter set will include user behavior on the corresponding exhibition node, as well as values ​​such as device usage times, single usage duration, user dwell time, and walking time.

[0071] Determine whether the user behavior of the first parameter set is a high-frequency user behavior; if it is a high-frequency user behavior, extract behavioral features from the first parameter set, generate a behavioral description vector, and perform minimum prediction error statistics on the behavioral description vector to determine multiple target behavioral information greater than a preset number.

[0072] When performing minimum error statistics, data from multiple exhibition nodes are input, and the minimum prediction error is calculated using the first set of input parameters. This error value uses the commonality metric and spatial influence matrix values ​​corresponding to each exhibition node as the benchmark for calculating the error. When the calculated error value is less than the minimum allowable error, the user behavior of the corresponding exhibition node is statistically analyzed and output as target behavior information. The minimum allowable error can be based on the average error value of the target behavior information filtered from historical data, and the part less than this error value is used as the output target behavior information.

[0073] The preset quantity describes the number of elements in the current output set of related behaviors that must reach a certain value. This prevents the analysis from being biased towards low-probability events due to an insufficient number of elements, thus failing to address the issue of analyzing high-frequency events in the exhibition scenario. The preset quantity is calculated using the minimum sample size.

[0074] For example The minimum sample size can be expressed as: ;in, The Z-value, representing the confidence level, is based on a 95% confidence level. The data describing the confidence level is based on the number of times the device is used, the duration of a single use, the user's dwell time, and the walking time. After calculating the confidence interval using these as inputs, the data is extracted from these four dimensions at the current exhibition node in turn, and the average of the minimum sample size under the four dimensions is taken as the preset number of devices currently in use. This indicates the expected percentage of behavior. This percentage will be differentiated based on the interaction type of the current exhibition node to illustrate the percentage value under the corresponding interaction type. This indicates the allowable error, which is typically set to 5%.

[0075] If it is not a high-frequency user behavior, then based on the exhibition nodes associated with the first parameter set, check the update time of the corresponding exhibition nodes, and update the first parameter set based on the update time of the exhibition nodes. The updated user behavior associated with the current exhibition node is used as the target behavior information for output.

[0076] When the information is not frequently accessed, the target behavioral information processed is more closely related to the current user behavior characteristics to avoid analysis bias caused by outdated data.

[0077] In one embodiment of the present invention, after extracting the location and target behavior of the user's main interaction during the exhibition in step S4, the content displayed at this time is identified, such as obtaining the triggering conditions of the display, as well as the efficiency and content of the display, and is adjusted in real time according to the content displayed to improve the user's interaction with each exhibition event in the exhibition node.

[0078] Preferably, the activity range of each extracted target behavior browsing exhibition node is represented by the regional range where the associated behavior set is located.

[0079] As shown in Figure 5, the implementation of step S4 includes: S41, identifying exhibition events based on the area range where the associated behavior set is located, and obtaining at least one point to be adjusted based on the triggering conditions of the exhibition events; the point to be adjusted represents the influence of the position, angle and playback speed of the displayed images on each exhibition node on the user's dwell time, and is used to identify the decision basis or element specification of multiple user behaviors under the current associated behavior set, so as to identify the exhibition content and relative form that users are interested in in high-frequency scenarios at different exhibition nodes.

[0080] S42, perform aggregation analysis on the points to be adjusted to obtain the aggregation path of each exhibition event, match the positions of each point to be adjusted on the aggregation path with the preset interactive hotspots, determine the matching degree of each point to be adjusted, and sort the exhibition events based on the matching degree of each point to be adjusted, and take the top 10% of the sorted exhibition events as the output target event information.

[0081] Preferably, the output target event information also needs to be examined for corresponding target behavior information, so that the target event information can correspond to a target behavior information. The triggering condition of the exhibition event is represented by the user behavior that triggers the current exhibition event and the corresponding timestamp.

[0082] Preferably, when processing the area where the set of related behaviors is located, the values ​​of device usage count and user dwell time are used as the judgment criteria. When both the number of device usage counts and user dwell time obtained within a unit of time are greater than a preset threshold, then the number of device usage counts and user dwell time are regarded as the triggering conditions for the current exhibition event. The preset thresholds for device usage counts and user dwell time are set based on the average value of users in historical data during the current period to identify the arrangement of relatively frequent exhibition events. If the value of device usage count and user dwell time of any exhibition node in the set of related behaviors is not greater than the preset threshold in a certain time period, then the exhibition node is not processed. Only when both are greater than the preset threshold are the relevant adjustment points extracted, and the feature characteristics of the element reduction or decision basis expressed by the adjustment points are used for aggregation analysis to complete the connection of the aggregation path.

[0083] Preferably, the area range where the associated behavior set is located represents the location where the current user behavior occurs. When integrating the exhibition data at these locations and performing aggregate analysis on the exhibition events, the implementation method includes: clustering based on the representational features of each point to be adjusted, and clustering based on the cosine similarity between each representational feature, i.e., using the K-means clustering method, treating each point to be adjusted with multiple similarity values ​​close to each other as a cluster, and connecting the clusters in descending order according to the average cosine similarity within the cluster to obtain the aggregation path of each exhibition event; as for the matching degree value with the preset interaction hotspot, it is based on the cosine similarity calculation result between the value of the cluster center of each point to be adjusted after clustering and the preset interaction hotspot, as the matching degree of the corresponding point to be adjusted. Finally, the top 10% of the exhibition events will represent the exhibition content that the current user is interested in in high-frequency scenarios, thereby completing the processing of user interaction.

[0084] Preferably, when extracting the representational features of each point to be adjusted, for example, the number of times the device is used and the duration of user stay at the exhibition node where the point to be adjusted is located are used as its feature values, or the position, angle and playback speed of the displayed image of the device at the exhibition node where the point to be adjusted is located are used as the corresponding labels, and the feature values ​​configured by the labels are used as the basis for clustering. After clustering, the cosine similarity is calculated by using the above feature values ​​and the feature values ​​of the preset interactive hotspots, and finally the matching degree of each point to be adjusted is set.

[0085] In one embodiment of the present invention, as shown in FIG6, step S5 is further implemented by: when there is an inclusion and intersection relationship between the current target event information and the adjacent target event information, determining the reduction result of the current target event information by the union of the device usage count and user dwell time of the current target event information with the adjacent target event information.

[0086] When there is no inclusion or intersection relationship between the current target event information and the adjacent target event information, the reduction result of the current target event information is determined by the single use duration and travel duration of the current target event information.

[0087] Based on the specification results of the current target event information, the parameters of the current target event information are updated, and the updated target event information is configured at each exhibition node.

[0088] The final output specification result is obtained by retrieving multiple sets of data extracted from the current target event information from the database. For example, if the current event information is related to the art exhibition area and its adjacent event is the sculpture exhibition stand, and these two events are related by inclusion and intersection, then the two events are jointly judged, and the average number of device usages and user dwell time of these two events is used as the retrieval content. The database is then used to find out how to adjust the displacement of the corresponding exhibition area and the size of the related video playback to achieve targeted processing of the relevant exhibition nodes. These updated parameters are then synchronized to the device-related information recorded in the exhibition nodes. The specification result represents the method of adjusting the number of device usages and user dwell time.

[0089] As for event information that does not contain or intersect, the approach tends to identify whether the corresponding exhibition node is biased towards a buffer zone for pedestrian flow, that is, to identify whether the corresponding exhibition node needs to be adjusted in terms of pedestrian flow movement based on the single use time and walking time of some equipment and exhibits on the exhibition node. When obtaining the protocol result, it is also necessary to identify whether the event information that does not contain or intersect overlaps, and determine whether it is greater than 5m or other distance values ​​that can describe two exhibits being in the same exhibition area. If they do not overlap, they are processed separately; if they overlap, the target event information is re-identified.

[0090] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention, which are still covered within the protection scope of the present invention.

Claims

1. An interactive display control method, characterized in that, include: S1. Using the location of the exhibited equipment as the exhibit node, extract user behavior at each exhibit node. Based on the corresponding time length of the user behavior, form a user behavior mapping set, which includes: number of times the device is used, duration of a single use, user dwell time, and walking time. S2. Identify the interaction type of the user behavior mapping set. Interaction types include direct and indirect interactions. Identify the proportion of interactions under direct and indirect interactions. Based on the interaction proportions in different time periods, determine the common metrics between adjacent exhibit nodes and generate an interaction spectrum. S3. Based on the exhibit nodes corresponding to the interaction spectrum, filter the exhibit nodes using the common metrics of each exhibit node at each time point, extract at least one target behavior information, and generate a set of associated behaviors. S4. Responding to the area range where the current set of associated behaviors is located, use the content implemented by the exhibit nodes as... For exhibition events, aggregate and analyze each exhibition event to obtain target event information; S5, based on the target event information in each time period, update the parameters using the reduction result between the current target event information and the adjacent target event information, and use the updated target event information as the output to complete the interactive adjustment of each exhibition node; The implementation method of step S1 includes: using the device usage on any exhibition node as the basic behavior condition, mapping the basic behavior condition to the number of device uses, the duration of a single use, the user's dwell time, and the walking time; setting the ratio of the duration of a single use to the number of device uses as the first value; using the ratio of the user's dwell time to the walking time as the second value; using the change ratio of the first value and the second value in different time periods, mapping the first value and the second value to form a user behavior mapping set.

2. The interactive display control method according to claim 1, characterized in that, The implementation method of mapping the first row of values ​​to the second row of values ​​also includes: based on the obtained first row of values ​​and second row of values, cross-grouping user behavior, setting a user behavior matrix, and obtaining at least one user type; dividing the user behavior matrix according to the dimension corresponding to the time period, setting a trend behavior label for each element in the user behavior matrix; updating the user behavior matrix in combination with the device usage of the current exhibition node, and outputting the updated data as a user behavior mapping set.

3. The interactive display control method according to claim 1, characterized in that, The implementation of step S2 includes: S21, identifying the interaction type of the user behavior mapping set, which includes various interaction methods under direct and indirect interaction, marking the user behaviors of direct and indirect interaction, and sorting each user behavior in chronological order to form a behavior interaction sequence; S22, connecting the exhibition nodes according to their locations to form a flow path, using the behavior interaction sequence to perform path analysis on the flow path, identifying the common metrics between adjacent exhibition nodes; and generating an interaction spectrum map based on the common metrics between adjacent exhibition nodes.

4. The interactive display control method according to claim 3, characterized in that, The implementation method of the commonality measurement between adjacent exhibition nodes in step S22 includes: taking the interaction ratio of each exhibition node as the processing object, obtaining the interaction ratio time series of the corresponding interaction type, fitting a Gaussian distribution to the interaction ratio time series, calculating the probability value of each exhibition node after the Gaussian distribution, and obtaining the conditional probability of two adjacent exhibition nodes; determining whether the conditional probability value of two adjacent exhibition nodes is greater than the preset conditional probability threshold, and if it is greater, obtaining the ratio of the time periods in which the two exhibition nodes are in the same trend change, and outputting it as the commonality measurement between adjacent exhibition nodes.

5. The interactive display control method according to claim 3, characterized in that, The implementation of step S21 also includes: nesting multiple interaction types with subtypes of direct and indirect interaction, using the occurrence time of the nested subtypes of direct and indirect interaction as the sorting basis, and outputting the sorted multi-level sequence as a behavior interaction sequence.

6. The interactive display control method according to claim 1, characterized in that, The implementation of step S3 also includes: S31, using the commonality measure of any exhibition node in the interactive spectrum to track the distribution location of each exhibition node, forming a spatial influence matrix related to the location of the exhibition node; S32, using the distance value and commonality measure value between each exhibition node in the spatial influence matrix to set the weight of the spatial influence matrix; updating the spatial influence matrix according to the weight of the spatial influence matrix; S33, based on the spatial influence matrix, finding the parameter combination that satisfies the minimum prediction error and the maximum distribution consistency, and outputting the corresponding parameter set as target behavior information.

7. The interactive display control method according to claim 6, characterized in that, The implementation of step S33 includes: obtaining the maximum common subset of the spatial influence matrix in multiple time periods, and using the data corresponding to the maximum common subset as the first parameter set; determining whether the user behavior of the first parameter set is a high-frequency user behavior; if it is a high-frequency user behavior, extracting behavioral features from the first parameter set, generating a behavioral description vector, and performing minimum prediction error statistics on the behavioral description vector to determine multiple target behavioral information greater than a preset number.

8. The interactive display control method according to claim 1, characterized in that, The implementation of step S4 includes: S41, identifying exhibition events based on the area range where the associated behavior set is located, and obtaining at least one point to be adjusted based on the triggering conditions of the exhibition events; S42, performing aggregation analysis on the points to be adjusted to obtain the aggregation path of each exhibition event, matching the position of each point to be adjusted on the aggregation path with the preset interactive hotspot, determining the matching degree of each point to be adjusted, and sorting the exhibition events based on the matching degree of each point to be adjusted, and using the top 10% of the sorted exhibition events as the output target event information.

9. The interactive display control method according to claim 1, characterized in that, The implementation of step S5 further includes: when there is an inclusion and intersection relationship between the current target event information and the adjacent target event information, determining the reduction result of the current target event information by the union of the device usage count and user dwell time of the current target event information with the adjacent target event information; when there is no inclusion and intersection relationship between the current target event information and the adjacent target event information, determining the reduction result of the current target event information by the single usage time and walking time of the current target event information; updating the parameters of the current target event information based on the reduction result of the current target event information, and configuring the updated target event information at each exhibition node.

Images