A data interactive display method and system for a digital exhibition hall

Abstract

The present invention provides a data interactive display method and system for a digital exhibition hall, and relates to the technical field of display content recommendation. The method includes: collecting historical display interactive behavior data of multiple display devices in the exhibition hall, extracting user browsing behavior preference characteristics to determine multiple browsing behavior preference patterns; extracting browsing time period preference characteristics and determining time period distribution patterns, generating a first interactive display strategy; collecting historical display environment data of multiple display devices and performing content environment association analysis, constructing a content environment analysis model and generating a second interactive display strategy; fusing and generating a target interactive display strategy and performing interactive display optimization; obtaining real-time display interactive behavior data of display devices and performing user behavior collaborative impact analysis, generating multiple interactive collaborative optimization strategies to perform interactive display collaborative optimization. The present invention realizes dynamic and personalized display content optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of display content recommendation, and in particular to a data interactive display method and system for a digital exhibition hall. Background Art

[0002] With the continuous development of digital technology, digital exhibition halls, as a new type of exhibition method, have gradually become an important carrier for various exhibitions, product displays and brand promotion. With the help of computer graphics, virtual reality and other technologies, digital exhibition halls can not only provide a richer and more interactive display experience, but also optimize the display content and improve the display effect through real-time data analysis.

[0003] In the process of displaying and promoting exhibits through the digital exhibition hall system, different audience groups may correspond to different time periods during the exhibition, resulting in time-varying demand and focus. For example, more corporate users may be interested in a certain type of high-tech product in the morning, while in the afternoon, there may be more interactive experience needs for ordinary consumers. If the preferences of the audience at different time periods cannot be effectively identified and analyzed, it is easy to lead to low adaptability of the display content.

[0004] In addition, in the process of displaying through display interfaces and interactive modes, the spatial layout of the display equipment has been pre-set and it may be troublesome to change it. The display environment has an impact on the display effect to a certain extent. For example, some exhibits may be more attractive in a bright environment, while other exhibits are better displayed in a darker environment. Therefore, in the process of displaying products through multiple display devices, it is necessary to effectively combine different environmental factors to improve the display effect of different exhibits.

[0005] In actual display scenarios, there are also behavioral influences between different users, that is, the browsing behavior of previous users may cause changes in the tendencies of observing users during the actual browsing process of exhibits. Mining the collaborative behaviors between different users in local scenarios can further enhance the user's browsing interaction experience. Summary of the invention

[0006] In order to solve the above technical problems, the present invention proposes a data interactive display method and system for a digital exhibition hall. By collecting and deeply analyzing the historical display data of the digital exhibition hall, data analysis is performed from multiple dimensions such as user behavior patterns, display time periods and display environments, and the influence of multiple factors such as time period changes, user behavior patterns, environmental characteristics and behavioral synergy on the display effect is fully considered, so as to achieve dynamic and personalized display content optimization and improve the display effect and user experience of the digital exhibition hall.

[0007] To achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides a data interactive display method for a digital exhibition hall, comprising: Collect historical display interaction behavior data of multiple display devices in the exhibition hall, extract browsing behavior preference features corresponding to multiple users from the historical display interaction behavior data, perform browsing behavior preference analysis on multiple users according to the browsing behavior preference features, and determine multiple browsing behavior preference patterns; Determine a browsing behavior tag corresponding to each user according to the browsing behavior preference pattern to which the user belongs, extract browsing time period preference features of multiple browsing behavior preference patterns according to the browsing behavior tags, perform browsing time period distribution analysis on the multiple browsing behavior preference patterns according to the browsing time period preference features, and determine a time period distribution pattern for the multiple browsing behavior preference patterns; Performing a time period behavior association analysis on the historical display interaction behavior data according to the time period distribution patterns corresponding to the multiple browsing behavior preference patterns, extracting the preferred interactive operation features under each browsing behavior preference pattern, and determining the first interactive display strategy for multiple display devices in the exhibition hall according to the time period distribution patterns corresponding to the multiple browsing behavior preference patterns; Collect historical display environment data of multiple display devices, perform content environment association analysis on historical display interaction behavior data based on the historical display environment data, extract content environment association features and construct a content environment analysis model, and generate a second interactive display strategy for multiple display devices in the exhibition hall through the content environment analysis model; The first interactive display strategy and the second interactive display strategy are integrated to generate a target interactive display strategy, and the interactive display of multiple display devices in the exhibition hall is optimized based on the target interactive display strategy; Acquire the real-time display interaction behavior data of each display device, perform user behavior collaborative impact analysis on the real-time display interaction behavior data, extract the behavioral collaborative impact characteristics of each browsing behavior preference pattern, generate an interactive collaborative optimization strategy for each browsing behavior preference pattern, and perform interactive display collaborative optimization on multiple display devices based on the interactive collaborative optimization strategy.

[0008] Preferably, determining the time period distribution patterns of the plurality of browsing behavior preference patterns further includes: The browsing behavior preference feature includes a plurality of browsing behavior characteristic parameters. After constructing a browsing behavior preference characteristic vector of each user through the plurality of browsing behavior characteristic parameters, clustering processing is performed on the plurality of users according to the browsing behavior preference characteristic vectors to determine a plurality of browsing behavior preference patterns. The global display period of the exhibition hall is divided into time periods to determine multiple local display periods, and multiple groups of user distribution data about multiple browsing behavior preference patterns are extracted for each local display period, wherein the user distribution data includes the user quantity characteristics of each browsing behavior preference pattern in the local display period, and the browsing period preference characteristics about the multiple browsing behavior preference patterns are obtained; Analyze multiple groups of user distribution data in each local display time period, determine the traffic characteristic parameters of each group of user distribution data in the local display time period, fuse multiple user quantity characteristics of each browsing behavior preference pattern in the local display time period according to the traffic characteristic parameters, and obtain the behavior preference distribution data of each local display time period; generate a time period distribution pattern for multiple browsing behavior preference patterns based on the behavior preference distribution data of multiple local display time periods.

[0009] Preferably, extracting the preferred interactive operation features in each browsing behavior preference mode, and determining the first interactive display strategy for multiple display devices in the exhibition hall according to the time period distribution patterns corresponding to the multiple browsing behavior preference modes, including: Extracting browsing interaction data of multiple users in each browsing behavior preference mode from the historical display interaction behavior data, determining the user's interactive operation preference data according to the user's browsing interaction data, and generating the preferred interactive operation features in the browsing behavior preference mode according to the interactive operation preference data of multiple users in the browsing behavior preference mode, including a group preference list for multiple display interactive operations; The browsing interaction data includes individual preference scores corresponding to multiple display interaction operations, and the group preference score of each display interaction operation is generated by fusing the interaction operation preference data of multiple users, and the multiple display interaction operations are sorted according to the group preference scores to obtain a group preference list; According to the time period distribution pattern of multiple browsing behavior preference patterns, the preferred interactive operation characteristics under each browsing behavior preference pattern are analyzed, and the allocation parameters of each browsing behavior preference pattern in the local display time period are determined according to the behavior preference distribution data of the local display time period. According to the allocation parameters, an allocation strategy for a preset number of interactive operations is determined, including the allocated number of interactive operations for each browsing behavior preference pattern. According to the preferred interactive operation characteristics and the allocated number of interactive operations under the browsing behavior preference patterns, local display interactive operation recommendation schemes for multiple browsing behavior preference patterns in the local display time period are determined, and according to the local display interactive operation recommendation schemes under the multiple local display time periods, a first interactive display strategy under the global display time period is generated.

[0010] Preferably, the content environment association features are extracted and a content environment analysis model is constructed, and a second interactive display strategy for multiple display devices in the exhibition hall is generated by the content environment analysis model, including: Extract multiple groups of display environment features of each display device from the historical display environment data, extract the exhibit type heat data of each group of display environment features from the historical display interaction behavior data according to the time information of the display environment features, associate each group of display environment features with the corresponding exhibit type heat data to construct a sample data set, use the multiple groups of display environment features in the sample data set as the input of the content environment analysis model, use the exhibit type heat data corresponding to the multiple groups of display environment features in the sample data set as the training target of the content environment analysis model, and obtain the content environment analysis model through training with the sample data set; After acquiring target display environment data about multiple display devices, the target display environment data is input into the trained content environment analysis model to generate exhibit type heat distribution data about the multiple display devices, and a second interactive display strategy for the multiple display devices is generated according to the target heat distribution data.

[0011] Preferably, the first interactive display strategy and the second interactive display strategy are integrated to generate a target interactive display strategy, and the interactive display optimization of multiple display devices in the exhibition hall is performed based on the target interactive display strategy, including: According to the heat distribution data of exhibit types about multiple display devices, multiple recommended exhibit types are set for each display device. For the target display device, after determining the target exhibit currently browsed by the user, according to the local display time period to which the current moment belongs, a local display interactive operation recommendation scheme for the target display device is determined, and according to the local display interactive operation recommendation scheme, multiple recommended display interactive operations for the target display device are set.

[0012] Preferably, the behavior synergy influence feature of each browsing behavior preference pattern is extracted, and the interactive synergy optimization strategy of each browsing behavior preference pattern is generated, including: Segment multiple groups of candidate collaborative interaction data from real-time display interaction behavior data, perform time-series correlation screening on multiple groups of candidate collaborative interaction data to determine multiple groups of target collaborative interaction data, extract the interaction operation evolution path of each user in each group of target collaborative interaction data, perform collaborative impact analysis on multiple interaction operation evolution paths of each group of target collaborative interaction data, and calculate the behavioral collaborative impact parameters corresponding to multiple display interaction operations under each group of target collaborative interaction data; Browsing behavior preference pattern matching is performed on each interactive operation evolution path, pattern matching results are generated, and an operation synergy influence set under each browsing behavior preference pattern is constructed, including multiple behavior synergy influence parameters corresponding to each display interactive operation, multiple interactive synergy influence operations of the browsing behavior preference pattern are determined according to the operation synergy influence set, and an interactive synergy optimization strategy for the browsing behavior preference pattern is generated according to the multiple interactive synergy influence operations.

[0013] In a second aspect, the present invention provides a data interactive display system for a digital exhibition hall, the system being used to implement the data interactive display method for a digital exhibition hall, comprising: A browsing behavior preference analysis module is used to collect historical display interaction behavior data of multiple display devices in the exhibition hall, extract browsing behavior preference features corresponding to multiple users from the historical display interaction behavior data, perform browsing behavior preference analysis on multiple users based on the browsing behavior preference features, and determine multiple browsing behavior preference patterns; A time period distribution analysis module is used to determine the browsing behavior label corresponding to each user according to the browsing behavior preference pattern to which the user belongs, extract browsing time period preference features of multiple browsing behavior preference patterns according to the browsing behavior label, perform browsing time period distribution analysis on multiple browsing behavior preference patterns according to the browsing time period preference features, and determine the time period distribution pattern of multiple browsing behavior preference patterns; A first interactive display optimization module is used to perform time period behavior association analysis on historical display interactive behavior data according to time period distribution patterns corresponding to multiple browsing behavior preference patterns, extract the preferred interactive operation features under each browsing behavior preference pattern, and determine a first interactive display strategy for multiple display devices in the exhibition hall according to the time period distribution patterns corresponding to the multiple browsing behavior preference patterns; A second interactive display optimization module is used to collect historical display environment data of multiple display devices, perform content environment association analysis on historical display interaction behavior data based on the historical display environment data, extract content environment association features and construct a content environment analysis model, and generate a second interactive display strategy for multiple display devices in the exhibition hall through the content environment analysis model; An interactive display management module, used to merge the first interactive display strategy and the second interactive display strategy to generate a target interactive display strategy, and optimize the interactive display of multiple display devices in the exhibition hall based on the target interactive display strategy; The interactive collaborative optimization module is used to obtain the real-time display interactive behavior data of each display device, perform user behavior collaborative impact analysis on the real-time display interactive behavior data, extract the behavioral collaborative impact characteristics of each browsing behavior preference pattern, generate an interactive collaborative optimization strategy for each browsing behavior preference pattern, and perform interactive display collaborative optimization on multiple display devices based on the interactive collaborative optimization strategy.

[0014] Preferably, the time period distribution analysis module includes: A browsing period preference analysis unit, used to determine the browsing behavior label corresponding to each user according to the browsing behavior preference pattern to which the user belongs, and to extract browsing period preference features related to multiple browsing behavior preference patterns according to the browsing behavior labels; The browsing period distribution analysis unit is used to perform browsing period distribution analysis on multiple browsing behavior preference patterns according to the browsing period preference characteristics, and determine the period distribution patterns of the multiple browsing behavior preference patterns.

[0015] The present invention has the following beneficial effects: The present invention determines the browsing behavior preference information of different users in the process of browsing exhibits by analyzing the historical display interactive behavior data of the exhibition hall to determine multiple browsing behavior preference patterns, performs time period feature analysis based on the browsing time period division on the multiple browsing behavior preference patterns to obtain the time period distribution pattern of the multiple browsing behavior preference patterns, generates a first interactive display strategy by comprehensively considering the user behavior pattern characteristics and the time period change characteristics, further considers the influence of environmental characteristics on the display effect, models the correlation between the display environment and the popularity of the exhibit type, and generates a second interactive display strategy according to the actual display environment data of the exhibition hall, and finally integrates to obtain the target interactive display strategy to realize the interactive display optimization of the exhibition hall, and analyzes the synergistic influence characteristics of the browsing behaviors between different users in local scenes according to the real-time display record data, introduces short-term interactive collaborative optimization strategies in different display scenes, improves the short-term diversified experience of users, and generates targeted interactive operation recommendations to ensure that each display device can provide better display effects in different time periods, while maximizing the user experience and realizing dynamic and personalized display content optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic flow chart of a data interactive display method for a digital exhibition hall provided as one of the implementation processes of the present invention.

[0017] Figure 2 A structural schematic diagram of a data interactive display system for a digital exhibition hall provided in one of the implementation processes of the present invention. DETAILED DESCRIPTION

[0018] In order to make those skilled in the art better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0019] Figure 1 The flowchart of a data interactive display method for a digital exhibition hall provided by one of the implementation processes of the present invention is shown. Specifically, the data interactive display method for a digital exhibition hall provided by an embodiment of the present invention includes the following steps: Step S1, collecting historical display interaction behavior data of multiple display devices in the exhibition hall, extracting browsing behavior preference features corresponding to multiple users from the historical display interaction behavior data, performing browsing behavior preference analysis on multiple users according to the browsing behavior preference features, and determining multiple browsing behavior preference patterns.

[0020] Specifically, the historical display interaction behavior data of multiple display devices specifically includes the interactive operation data of different users on different display devices, such as dwell time, number of clicks, interaction methods, etc., reflecting the user's browsing behavior of the various exhibits provided by the display devices. By analyzing the historical display interaction behavior data, it is possible to extract the browsing behavior preference characteristics corresponding to multiple users, which include multiple browsing behavior characteristic parameters of the user during a certain browsing process, such as browsing speed, interaction frequency, interaction duration, etc. For example, some users may prefer to browse quickly and skip exhibits quickly; while some users may browse in depth and view the content of each exhibit in detail.

[0021] Based on the extracted browsing behavior preference features of each user, browsing behavior preference analysis is further performed on multiple users, and different browsing behavior preference patterns are identified, so as to facilitate understanding of the behavioral trends and habits of different types of users in the display of exhibits during the exhibition in the exhibition hall. In this process, through the multiple browsing behavior feature parameters contained in the browsing behavior preference features of the user, a browsing behavior preference feature vector of the user can be constructed to fully reflect the interactive mode and behavioral preferences of the user in the exhibition hall. A clustering algorithm such as a distance or density clustering scheme is adopted. In this embodiment, the DBSCAN clustering algorithm is taken as an example, and the DBSCAN clustering algorithm is adopted to analyze and process the browsing behavior preference feature vectors of different users, and multiple users are clustered according to the browsing behavior preference feature vectors, so that users with similar browsing behaviors are classified into one category, thereby forming multiple browsing behavior preference patterns, which are used to reflect the common characteristics or behavior patterns exhibited by different users during the browsing process.

[0022] Step S2: determine the browsing behavior label corresponding to each user according to the browsing behavior preference pattern to which the user belongs, extract browsing time period preference features related to multiple browsing behavior preference patterns according to the browsing behavior label, perform browsing time period distribution analysis on the multiple browsing behavior preference patterns according to the browsing time period preference features, and determine the time period distribution pattern related to the multiple browsing behavior preference patterns.

[0023] Specifically, after determining the different browsing behavior preference patterns of multiple users, the browsing behavior characteristics of users in different time periods are further analyzed. In this process, the browsing behavior labels of users are first determined according to the browsing behavior preference patterns to which the users belong. These labels can be, for example, quick browsing type, deep browsing type, interest-oriented type, etc., so as to distinguish the behavior preferences corresponding to different user groups. Then, the historical display interaction behavior data is divided into time periods, and the activity of different users in different time periods is analyzed. For example, some browsing behavior patterns are more frequent in the morning, while some patterns are more common in the afternoon or evening. By analyzing the distribution of personnel in browsing behavior preference patterns in different time periods, the browsing time period preference characteristics of multiple browsing behavior preference patterns are extracted to characterize the time period distribution of users in each browsing behavior preference pattern in the actual browsing of exhibits. Finally, the time period distribution patterns of multiple browsing behavior preference patterns are analyzed to identify and quantify the time period distribution characteristics of each browsing behavior preference pattern, clarify the concentration of a certain behavior pattern in different time periods, and provide a basis for subsequent display content adjustment.

[0024] Step S3, performing time period behavior association analysis on historical display interaction behavior data according to the time period distribution patterns corresponding to multiple browsing behavior preference patterns, extracting the preferred interactive operation features under each browsing behavior preference pattern, and determining the first interactive display strategy for multiple display devices in the exhibition hall according to the time period distribution patterns corresponding to the multiple browsing behavior preference patterns.

[0025] Specifically, the purpose of performing time period behavior association analysis on historical display interaction behavior data is to analyze the interactive operation habits preferred by users under different browsing behavior preference modes. For example, when a user prefers to browse quickly, he may simply browse the brief information of the exhibits, and quickly browse the structural information of the exhibits through some simple interactive operations such as zooming and flipping. When a user prefers to browse in depth, he may view the detailed introduction information of the exhibits, and view the structural design of some details of the exhibits through more complex interactive operations. By analyzing the interactive data of different browsing behavior modes in different time periods, the preferred interactive operation features in each browsing behavior preference mode can be extracted. These features include the interactive operations that users prefer in the specific browsing process, and further according to the browsing behavior preference modes to which the user groups belong in different time periods, it is possible to provide users with simple and convenient interactive interfaces in different time periods in a targeted manner, so as to avoid affecting the actual experience of users due to the numerous interactive operations. In this way, suitable interactive strategies can be designed for different display time periods, thereby determining the first interactive display strategy for multiple display devices in the exhibition hall.

[0026] Step S4: collect historical display environment data of multiple display devices, perform content environment association analysis on historical display interaction behavior data based on the historical display environment data, extract content environment association features and construct a content environment analysis model, and generate a second interactive display strategy for multiple display devices in the exhibition hall through the content environment analysis model.

[0027] Specifically, considering that the display environment data will affect the display effect of some exhibits to a certain extent, the historical display environment data of multiple display devices in the exhibition hall are further collected, and the content environment association analysis is performed on the historical display interaction behavior data. Among them, the historical display environment data can be data such as the light intensity and noise level of the area where the display device is located, so as to determine the light conditions, noise conditions, etc. of the area where the display device is located. For example, some exhibits may require sound or voice interaction, and the area near the wall or corner may be less affected by noise than the central area. Through the analysis of these environmental data, the correlation between the heat of different types of exhibits and the display environment can be extracted, so as to extract the corresponding content environment association features, and establish a content environment analysis model based on these features. The content environment analysis model learns the heat change rules of different types of exhibits under different display environments to achieve the inference of the display effects of different exhibits in the new environment, so that the display environment data can be analyzed based on the model to generate a second interactive display strategy that can improve the display effect, which is used to indicate how to adjust the display mode of multiple types of exhibits according to the environmental changes of the display equipment.

[0028] Step S5: The first interactive display strategy and the second interactive display strategy are integrated to generate a target interactive display strategy, and interactive display optimization is performed on multiple display devices in the exhibition hall based on the target interactive display strategy.

[0029] Specifically, the first interactive display strategy determined above optimizes the display from the perspective of time period and user behavior mode, and the second interactive display strategy further considers the changes in display effects caused by differences in display environment, optimizes the display process, and finally integrates the two strategies to achieve more accurate display content optimization, that is, according to factors such as light and noise in the environment where the display equipment is located, determine different types of exhibits suitable for display, and further determine the corresponding user groups according to the display time period, so as to provide a personalized interactive operation interface according to the preferences of the user group, and intelligently adjust the display content of each display device in the exhibition hall. Exemplarily, in the morning period, the light in a certain display area is strong and because the current user group prefers to browse quickly, it is preferred to recommend the display of some exhibits with strong visual impact and simplicity, and provide a relatively simple interactive operation interface, such as providing some operation options commonly used by the current group to facilitate users to browse exhibits quickly; while in the evening period, the environment of a certain display area is dark and quiet, and the user group prefers to browse deeply, it is preferred to recommend the display of some exhibits suitable for this environment, such as electronic audio equipment, and provide more detailed operation options such as sound quality experience to actually display the playback effect. Based on the integrated target interactive display strategy, a more intelligent, flexible and accurate interactive display solution is provided for the display equipment in the exhibition hall, thereby improving the display effect and audience experience.

[0030] Step S6: acquiring real-time display interaction behavior data of each display device, performing user behavior collaborative impact analysis on the real-time display interaction behavior data, extracting the behavioral collaborative impact features of each browsing behavior preference pattern, generating an interactive collaborative optimization strategy for each browsing behavior preference pattern, and performing interactive display collaborative optimization on multiple display devices based on the interactive collaborative optimization strategy.

[0031] Specifically, the target interactive display strategy provides a strategy for optimizing the interactive operation interface during the display process from the perspective of the preferences and habits of different user groups, and provides exhibits suitable for display in different areas according to the actual environmental changes in different exhibition areas, providing personalized interactive display experience for different user groups. On this basis, the behavioral influence of users in the exhibition hall is further considered. For example, the interactive choice of user A for certain display interactive operations may attract more attention from user B, and this attractiveness varies within different groups. For example, a research group is more interested in some technical points of the exhibits. Users may be affected in the process of observing the browsing of the previous user, so that they will give priority to some operating habits of the previous user in the process of browsing the exhibits. The above-mentioned target interactive display strategy is more about realizing some high-heat options within some groups from a global perspective, and pays less attention to some relatively unpopular interactive operations. The short-term interactive behavior impact may cause some relatively low-heat display interactive operations to have a large change in local heat. In this embodiment, real-time display interaction behavior data of different display devices are further obtained, and the collaborative characteristics of some operation behaviors between two users browsing exhibits adjacent to each other are analyzed according to the user alternation situation, so as to further extract the behavioral collaborative influence characteristics of each user group, that is, each browsing behavior preference mode, so as to generate corresponding interactive collaborative optimization strategies. For example, if it is found that a relatively unpopular display interaction operation has a short-term surge in popularity within a certain group in a short period of time, the interactive display collaborative optimization of the display device is performed based on the real-time situation, and the recommendation strategy in the short term is dynamically adjusted to further enhance the user's browsing interaction experience.

[0032] Optionally, in step S2 of the above solution, the process of determining the time period distribution patterns of the multiple browsing behavior preference patterns includes the following: The global display time period of the exhibition hall is divided into time periods to determine multiple local display time periods, and multiple groups of user distribution data on multiple browsing behavior preference patterns are extracted for each local display time period. The user distribution data includes the user quantity characteristics of each browsing behavior preference pattern in the local display time period, and the browsing time period preference characteristics of the multiple browsing behavior preference patterns are obtained.

[0033] Specifically, in order to more accurately analyze the behavioral characteristics of users in different time periods, the global display period of the exhibition hall, such as the period from 8 am to 10 pm, is further divided into time periods to determine multiple local display periods. For example, the display period of a day is divided into multiple local display periods at intervals of two hours. It is worth noting that this is only an exemplary solution, and those skilled in the art can also make reasonable divisions based on factors such as the opening hours of the exhibition hall and fluctuations in customer flow.

[0034] For each local display time period, multiple groups of user distribution data on multiple browsing behavior preference patterns are extracted. For example, in a historical week, the proportion of the number of users under each browsing behavior preference pattern in a certain local display time period of the exhibition hall every day is used as the browsing time period preference features corresponding to the multiple browsing behavior preference patterns, indicating the overall distribution of the number of users under each mode in different time periods.

[0035] Analyze multiple groups of user distribution data in each local display time period, determine the traffic characteristic parameters of each group of user distribution data in the local display time period, fuse multiple user quantity characteristics of each browsing behavior preference pattern in the local display time period according to the traffic characteristic parameters, and obtain the behavior preference distribution data of each local display time period; generate a time period distribution pattern for multiple browsing behavior preference patterns based on the behavior preference distribution data of multiple local display time periods.

[0036] Specifically, in order to more accurately analyze user behavior in different time periods, the impact of human traffic on user behavior is further considered. The human traffic characteristic parameters of different local display time periods can be determined based on the user distribution data of each local display time period. It can be understood as the parameters obtained after normalization based on the total number of people in the exhibition hall in different time periods. The larger the human traffic characteristic parameter of a group of user distribution data in the local display time period, the more users there are in the exhibition hall during this period. Therefore, multiple groups of user distribution data in each local display time period are fused according to the human traffic characteristic parameters, so that the user proportion data of different browsing behavior preference modes in the scene with more total people has greater reference significance. After fusing multiple groups of user distribution data in the local display time period, the behavior preference distribution data of each local display time period can be obtained, which is used to indicate the proportion of different types of users in this period.

[0037] Finally, based on the behavior preference distribution data of multiple local display time periods, a time period distribution pattern of multiple browsing behavior preference patterns is constructed to reflect the specific user distribution of each browsing behavior preference pattern in the exhibition hall in different time periods. It can be used to provide data reference for adjusting the interactive content displayed by display devices in different time periods, so as to adjust the display strategy according to the needs of different user groups.

[0038] Optionally, in step S3 of the above scheme, the preferred interactive operation features under each browsing behavior preference mode are extracted, and the first interactive display strategy for multiple display devices in the exhibition hall is determined according to the time period distribution patterns corresponding to the multiple browsing behavior preference modes, including the following contents: Step S31, extracting browsing interaction data of multiple users in each browsing behavior preference mode from historical display interaction behavior data, determining the user's interactive operation preference data based on the user's browsing interaction data, and generating preferred interactive operation features in the browsing behavior preference mode based on the interactive operation preference data of multiple users in the browsing behavior preference mode, including a group preference list for multiple display interaction operations.

[0039] Specifically, the browsing interaction data includes data related to the specific interaction mode of the user on the display device, that is, the specific operation record of the user on the display device, such as the user's click selection and duration of different interaction options. According to the duration of different interaction options, that is, the time information of staying in the content provided by the interaction option, the user's preference information on different interaction operations can be preliminarily determined. For example, after the duration of different interaction options is normalized, it is used as the preference score data of the interaction operation, thereby determining the user's individual preference scores corresponding to multiple display interaction operations respectively, and generating a group preference score for each display interaction operation by fusing the interaction operation preference data of multiple users, thereby generating the preferred interaction operation features under the browsing behavior preference mode according to the interaction operation preference data of multiple users under the browsing behavior preference mode. Among them, the preferred interaction operation features include a group preference list for multiple display interaction operations.

[0040] In this process, the average of multiple individual preference scores of different display interaction operations can be calculated as the group preference score of the display interaction operation, and the multiple display interaction operations can be sorted according to the group preference score of the display interaction operation, so as to generate the preferred interaction operation characteristics under the browsing behavior preference mode, that is, to obtain a group preference list for multiple display interaction operations, which is used to indicate the relative importance or popularity of various display interaction operations for this user group under the browsing behavior preference mode.

[0041] Step S32: analyzing the preferred interactive operation characteristics under each browsing behavior preference mode according to the time period distribution pattern of multiple browsing behavior preference modes, determining the allocation parameters of each browsing behavior preference mode in the local display time period according to the behavior preference distribution data of the local display time period, and determining the allocation strategy for the preset number of interactive operations according to the allocation parameters.

[0042] Specifically, after obtaining the preferred interactive operation characteristics under each browsing behavior preference mode, the preferred interactive operation characteristics are analyzed in combination with the time period distribution patterns of multiple browsing behavior preference modes. The preferred interactive operation characteristics reflect the preferences of user groups for different display interactive operations under the browsing behavior preference mode, and the time period distribution pattern reflects the distribution of user groups under different time periods. Combining the two can further determine the allocation parameters of each browsing behavior preference mode in the local display time period.

[0043] In this process, the interactive interface size of the display device is limited. It is difficult to introduce the details of the exhibits simply through a few interactive options. Users are required to perform in-depth interactive operations based on the prompts to browse more detailed content. Considering that different user groups have different browsing behavior preference patterns, their preferred interactive options in the actual browsing process of exhibits vary. Therefore, if the most popular interactive options in each group are displayed on the main interface of the interactive operation based on the distribution of user groups in different time periods, it can help users browse their favorite content more quickly.

[0044] Therefore, by analyzing the distribution of the number of user groups in different time periods, the pre-set number of interactive operations to be allocated is actually allocated. For example, the display device can provide choices corresponding to the display interactive operations of six large icons. After determining the allocation ratio of different user groups in a certain time period, that is, the allocation parameters of each browsing behavior preference mode, the preset number of interactive operations can be actually allocated according to the allocation parameters, thereby obtaining an allocation strategy for the preset number of interactive operations, specifically including the allocated number of interactive operations for each browsing behavior preference mode. During the allocation process, in scenarios involving situations where rounding cannot be performed, priority may be given to groups with a higher proportion of users. In this way, the number of interactive operation allocations corresponding to different browsing behavior preference patterns is determined, and then based on the preferred interactive operation characteristics and the number of interactive operation allocations under the browsing behavior preference patterns, local display interactive operation recommendation schemes for multiple browsing behavior preference patterns under local display time periods are determined. Specifically, based on the number of interactive operation allocations, a corresponding number of multiple display interactive operations are selected from the preferred interactive operation characteristics, that is, a corresponding number of multiple display interactive operations with the highest popularity in the group, and these display interactive operations are used as recommended options for the browsing behavior preference pattern. In this way, local display interactive operation recommendation schemes for multiple browsing behavior preference patterns can be generated, and finally, the first interactive display strategy under the global display time period is generated based on the local display interactive operation recommendation schemes under multiple local display time periods.

[0045] Optionally, in step S4 of the above solution, content environment association features are extracted and a content environment analysis model is constructed, and a second interactive display strategy for multiple display devices in the exhibition hall is generated through the content environment analysis model, including the following contents: Multiple groups of display environment features for each display device are extracted from the historical display environment data. Based on the time information of the display environment features, the exhibit type heat data of each group of display environment features is extracted from the historical display interaction behavior data. Each group of display environment features is associated with the corresponding exhibit type heat data to construct a sample data set.

[0046] Specifically, the generation method of the second interactive display strategy focuses on the correlation analysis between the display environment characteristics of the display equipment in the exhibition hall and the popularity of the exhibit types. The purpose is to optimize the match between the exhibit content and the environment through in-depth analysis of the exhibit display effects and user interaction patterns in different display environments, thereby further improving the exhibition hall's display effects and user experience. In this process, multiple groups of display environment characteristics of each display device are first extracted from the historical display environment data, where each group of display environment characteristics can be specifically about the overall environmental conditions in a certain period of time, such as the average brightness and average noise decibels of the display area in a certain period of time. Specifically, the environmental level of the display area can be monitored by light sensors, microphones and other devices.

[0047] The division of time periods can be combined with the aforementioned multiple local display time periods, or considering that information such as lighting or noise in the environment does not change significantly in a shorter period of time, a longer duration method can be selected for division, which is not specifically limited here. For the exhibit type popularity data of display environment characteristics, it can specifically be the number of views of each type of exhibit in the corresponding period of time, or the data obtained by comprehensively measuring the interactive frequency and participation of the exhibits. Those skilled in the art can choose a suitable method to measure the popularity of each type of exhibit in different time periods, thereby obtaining the exhibit type heat data for each group of display environment characteristics. Finally, a sample data set for training the content environment analysis model is constructed by combining multiple groups of display environment characteristics and corresponding exhibit type heat data.

[0048] In the process of training the content environment analysis model through the sample data set, the multiple groups of display environment features in the sample data set will be used as the input of the content environment analysis model, and the heat data of the exhibit types corresponding to the multiple groups of display environment features in the sample data set will be used as the training target of the content environment analysis model to guide the model to learn the association pattern between different environmental features and exhibit heat. By training the model with the sample data set, the content environment analysis model will be able to learn the heat performance of different exhibit types in a specific display environment. During the training process, the model analyzes the changing rules of exhibit heat in different environments and establishes a mapping relationship between exhibit display effects and environmental factors. Among them, the content environment analysis model is specifically constructed by selecting a multi-layer perceptron model to extract the deep correlation information between environmental features and exhibit heat and perform recognition modeling.

[0049] After the content environment analysis model is trained, when the target display environment data of multiple display devices in the exhibition hall is collected, such as data containing the current environmental feature information of each display device such as lighting and noise, the target display environment data is input into the trained content environment analysis model. The content environment analysis model generates the heat distribution data of the exhibit types of multiple display devices according to the input environmental features, which is used to indicate the expected heat of displaying different types of exhibits in a specific display environment, and then generates the second interactive display strategy for multiple display devices according to the target heat distribution data, which is used to indicate the types of exhibits recommended for display by different display devices. For example, in an environment where the lighting of a certain display device is weak and the space is relatively small, displaying a certain audio electronic product may obtain a higher display heat.

[0050] Optionally, in step S5 of the above solution, the first interactive display strategy and the second interactive display strategy are merged to generate a target interactive display strategy, and interactive display optimization is performed on multiple display devices in the exhibition hall based on the target interactive display strategy, including the following contents: According to the heat distribution data of exhibit types about multiple display devices, multiple recommended exhibit types are set for each display device. For the target display device, after determining the target exhibit currently browsed by the user, according to the local display time period to which the current moment belongs, a local display interactive operation recommendation scheme for the target display device is determined, and according to the local display interactive operation recommendation scheme, multiple recommended display interactive operations for the target display device are set.

[0051] Specifically, the target interactive display strategy generated by the fusion of the first interactive display strategy and the second interactive display strategy comprehensively considers the influence of multiple factors such as user behavior patterns, time period changes and environmental characteristics on the display effect. In the process of optimizing the interactive display of multiple display devices in the exhibition hall according to the target interactive display strategy, multiple recommended exhibit types for each display device can be set according to the heat distribution data of the exhibit types of multiple display devices, that is, the best multiple types of exhibits for different display devices in the current display environment. This process is only used to indicate which types of exhibits may achieve better results, and in the actual display process, the exhibition hall can provide multiple key types of exhibit browsing options according to actual needs, or provide one of the exhibits as a representative of these types of exhibits. This embodiment does not specifically limit it. On the target display device, when it is detected that the user selects the target exhibit, the local display time period to which the current moment belongs is analyzed to determine the display interaction operation that multiple users in the local display time period will have a greater probability of selecting.

[0052] Since the user has not yet performed a specific browsing operation, it is difficult to determine the specific browsing behavior preference pattern corresponding to the user. Through the previously determined first interactive display strategy, the browsing behavior preference pattern that the user may correspond to can be analyzed and predicted in advance, so as to recommend multiple display interaction operations that are more preferred by the groups under these browsing behavior preference patterns. For example, in certain time periods, the user group may be more inclined to view product details, while in other time periods, they may be inclined to interact with the product such as rotating, enlarging, and virtual trial, etc. After determining the local display interaction operation recommendation scheme for the target display device, multiple recommended display interaction operations for the target display device are set according to the local display interaction operation recommendation scheme, that is, the interactive options that are more likely to be selected by multiple users are set on the interactive interface, so that users can quickly select their intended interactive operations and avoid being disturbed by the user in the actual browsing of exhibits due to the large number of interface functions.

[0053] Through the above method, the target interactive display strategy can be implemented into specific exhibits and interactive operations through the display device control system, and the display content can be dynamically adjusted in real time. This method of generating targeted interactive operation recommendations can ensure that each display device can provide a good display effect in each time period, while maximizing the user experience, and ultimately achieving dynamic and personalized display content optimization.

[0054] Optionally, in step S6 of the above solution, the behavior synergy influence feature of each browsing behavior preference pattern is extracted, and the interactive synergy optimization strategy of each browsing behavior preference pattern is generated, including the following contents: Multiple groups of candidate collaborative interaction data are segmented from the real-time display interaction behavior data, and time-series correlation screening is performed on the multiple groups of candidate collaborative interaction data to determine multiple groups of target collaborative interaction data. The interaction operation evolution path of each user in each group of target collaborative interaction data is extracted, and collaborative impact analysis is performed on the multiple interaction operation evolution paths of each group of target collaborative interaction data. The behavioral collaborative impact parameters corresponding to multiple display interaction operations under each group of target collaborative interaction data are calculated.

[0055] Specifically, the candidate collaborative interaction data is the actual operation data of any two adjacent users in the display interaction behavior data of the display device. In order to improve the reference degree of synergy, technicians in this field may further limit the number of users involved in the candidate collaborative interaction data, such as data corresponding to three or five users who continuously browse exhibits based on the same display device. This is not specifically limited in this embodiment. Taking into account the timeliness of behavioral impact, multiple groups of candidate collaborative interaction data can be screened for temporal correlation, that is, in-depth analysis is performed on some data with adjacent intervals lower than a preset time threshold, for example, within 15 seconds after a user leaves the current display device, the next user starts a new browsing. Furthermore, technicians in this field can also make more detailed number limits on the target collaborative interaction data based on the traffic data involved in the display device, that is, analyze the scene with multiple users watching, and further limit the traffic corresponding to the target collaborative interaction data to a specific threshold, so as to extract behavioral collaborative impact features with strong correlation.

[0056] After screening out multiple groups of target collaborative interaction data, the interactive operation evolution path of each user in the target collaborative interaction data is extracted, such as multiple display interactive operations selected by the user in the process of browsing exhibits, and the timestamps corresponding to different display interactive operations. Then, the collaborative impact analysis of adjacent users is performed through the user's interactive operation evolution path, specifically the behavioral collaborative impact parameter for each display interactive operation. Exemplarily, for each display interactive operation, the difference between the timestamps is analyzed, and the timestamps of different display interactive operations are different. The order between the display operations is further considered, and the sequence weight is introduced to correct the difference between the timestamps. For example, the first operation of user B is affected by the third-to-last display interactive operation of user A, and user A spends a long time on the last two operations, resulting in a large difference in the timestamps of operation x between user A and user B, resulting in low synergy, while the large sequence difference still shows strong synergy, so the corrected weight is determined by sequence, so that the behavioral collaborative impact parameter finally calculated can reduce the negative impact caused by the duration of a single operation.

[0057] Browsing behavior preference pattern matching is performed on each interactive operation evolution path, pattern matching results are generated, and an operation synergy influence set under each browsing behavior preference pattern is constructed, including multiple behavior synergy influence parameters corresponding to each display interactive operation, multiple interactive synergy influence operations of the browsing behavior preference pattern are determined according to the operation synergy influence set, and an interactive synergy optimization strategy for the browsing behavior preference pattern is generated according to the multiple interactive synergy influence operations.

[0058] Specifically, after analyzing and obtaining the behavioral synergy influence parameters corresponding to multiple display interaction operations under different target collaborative interaction data, group identification analysis is performed on users in the collaborative interaction data of different targets, that is, matching users with different browsing behavior preference patterns according to the evolution path of their interactive operations. Specifically, multiple display interaction operations involved in the evolution path of the interactive operation are matched with popular operations in the browsing behavior preference pattern, and finally the browsing behavior preference pattern with the highest matching degree is selected as the browsing behavior preference pattern to which the user belongs. After obtaining the matching result of each user in this way, the candidate collaborative interaction data under each browsing behavior preference pattern is screened out, that is, the candidate collaborative interaction data whose multiple interactive operation evolution paths belong to the same group are divided into corresponding groups, so as to construct an operation synergy influence set under each browsing behavior preference pattern according to the behavioral synergy influence parameters corresponding to the multiple display interaction operations in the evolution path of the interactive operation, which includes the quantitative synergy characteristics of different display interaction operations in the short term. Based on the average coordination level of the display interaction operations, such as the mean of the behavioral synergy influence parameters, multiple display interaction operations with high coordination in the browsing behavior preference pattern can be determined and recorded as interactive synergy influence operations, and based on the determined multiple interactions The synergistic influence operation re-optimizes the order of display interaction operations used for recommendation by this group in the short term. The high synergy in the short term indicates that some display interaction operations are also in line with the preferences of a large number of users in this group to a certain extent. During the analysis of historical data, some display interaction operations did not comprehensively consider the synergistic influence of the user group. This may cause some display interaction operations with high recommendation value to be ignored due to the analysis based solely on popularity, resulting in the actual display effect being affected. Through the above-mentioned user behavior synergistic influence analysis, the interactive display strategies of different display devices can be dynamically adjusted according to real-time data. By capturing real-time collaborative behaviors, the adaptability of the display optimization strategy to local scenarios is greatly improved, thereby achieving better interactive display effects.

[0059] Figure 2 The structure diagram of a data interactive display system for a digital exhibition hall provided by one of the implementation processes of the present invention is shown. A data interactive display system for a digital exhibition hall provided by an embodiment of the present invention is specifically used to implement the data interactive display method for a digital exhibition hall described above, and the system includes: The browsing behavior preference analysis module 11 is used to collect historical display interaction behavior data of multiple display devices in the exhibition hall, extract browsing behavior preference features corresponding to multiple users from the historical display interaction behavior data, perform browsing behavior preference analysis on multiple users according to the browsing behavior preference features, and determine multiple browsing behavior preference patterns; The time period distribution analysis module 12 is used to determine the browsing behavior label corresponding to each user according to the browsing behavior preference pattern to which the user belongs, extract the browsing time period preference features of multiple browsing behavior preference patterns according to the browsing behavior label, perform browsing time period distribution analysis on the multiple browsing behavior preference patterns according to the browsing time period preference features, and determine the time period distribution pattern of the multiple browsing behavior preference patterns; Among them, the time period distribution analysis module includes a browsing time period preference analysis unit and a browsing time period distribution analysis unit. The browsing time period preference analysis unit is used to determine the browsing behavior label corresponding to each user according to the browsing behavior preference pattern to which the user belongs, and to extract browsing time period preference features about multiple browsing behavior preference patterns according to the browsing behavior labels; the browsing time period distribution analysis unit is used to perform browsing time period distribution analysis on multiple browsing behavior preference patterns according to the browsing time period preference features, and to determine the time period distribution patterns about multiple browsing behavior preference patterns.

[0060] The first interactive display optimization module 13 is used to perform time period behavior association analysis on the historical display interactive behavior data according to the time period distribution patterns corresponding to the multiple browsing behavior preference patterns, extract the preferred interactive operation features under each browsing behavior preference pattern, and determine the first interactive display strategy for multiple display devices in the exhibition hall according to the time period distribution patterns corresponding to the multiple browsing behavior preference patterns; The second interactive display optimization module 14 is used to collect historical display environment data of multiple display devices, perform content environment association analysis on historical display interaction behavior data based on the historical display environment data, extract content environment association features and construct a content environment analysis model, and generate a second interactive display strategy for multiple display devices in the exhibition hall through the content environment analysis model; An interactive display management module 15 is used to merge the first interactive display strategy and the second interactive display strategy to generate a target interactive display strategy, and optimize the interactive display of multiple display devices in the exhibition hall based on the target interactive display strategy; The interactive collaborative optimization module 16 is used to obtain the real-time display interactive behavior data of each display device, perform user behavior collaborative impact analysis on the real-time display interactive behavior data, extract the behavioral collaborative impact characteristics of each browsing behavior preference pattern, generate an interactive collaborative optimization strategy for each browsing behavior preference pattern, and perform interactive display collaborative optimization on multiple display devices based on the interactive collaborative optimization strategy.

[0061] The above is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A data interactive display method for a digital exhibition hall, characterized in that: include: Collect historical display interaction behavior data of multiple display devices in the exhibition hall, extract browsing behavior preference features corresponding to multiple users from the historical display interaction behavior data, perform browsing behavior preference analysis on multiple users according to the browsing behavior preference features, and determine multiple browsing behavior preference patterns; Determine a browsing behavior tag corresponding to each user according to the browsing behavior preference pattern to which the user belongs, extract browsing time period preference features of multiple browsing behavior preference patterns according to the browsing behavior tags, perform browsing time period distribution analysis on the multiple browsing behavior preference patterns according to the browsing time period preference features, and determine a time period distribution pattern for the multiple browsing behavior preference patterns; Performing a time period behavior association analysis on the historical display interaction behavior data according to the time period distribution patterns corresponding to the multiple browsing behavior preference patterns, extracting the preferred interactive operation features under each browsing behavior preference pattern, and determining the first interactive display strategy for multiple display devices in the exhibition hall according to the time period distribution patterns corresponding to the multiple browsing behavior preference patterns; Collect historical display environment data of multiple display devices, perform content environment association analysis on historical display interaction behavior data based on the historical display environment data, extract content environment association features and construct a content environment analysis model, and generate a second interactive display strategy for multiple display devices in the exhibition hall through the content environment analysis model; The first interactive display strategy and the second interactive display strategy are integrated to generate a target interactive display strategy, and interactive display optimization is performed on multiple display devices in the exhibition hall based on the target interactive display strategy; Acquire the real-time display interaction behavior data of each display device, perform user behavior collaborative impact analysis on the real-time display interaction behavior data, extract the behavioral collaborative impact characteristics of each browsing behavior preference pattern, generate an interactive collaborative optimization strategy for each browsing behavior preference pattern, and perform interactive display collaborative optimization on multiple display devices based on the interactive collaborative optimization strategy.

2. The data interactive display method of a digital exhibition hall according to claim 1 is characterized in that: For determining the time period distribution pattern of multiple browsing behavior preference patterns, it also includes: The browsing behavior preference feature includes a plurality of browsing behavior characteristic parameters. After constructing a browsing behavior preference characteristic vector of each user through the plurality of browsing behavior characteristic parameters, clustering processing is performed on the plurality of users according to the browsing behavior preference characteristic vectors to determine a plurality of browsing behavior preference patterns. The global display period of the exhibition hall is divided into time periods to determine multiple local display periods, and multiple groups of user distribution data about multiple browsing behavior preference patterns are extracted for each local display period, wherein the user distribution data includes the user quantity characteristics of each browsing behavior preference pattern in the local display period, and the browsing period preference characteristics about the multiple browsing behavior preference patterns are obtained; Analyze multiple groups of user distribution data in each local display time period, determine the traffic characteristic parameters of each group of user distribution data in the local display time period, fuse multiple user quantity characteristics of each browsing behavior preference pattern in the local display time period according to the traffic characteristic parameters, and obtain the behavior preference distribution data of each local display time period; generate a time period distribution pattern for multiple browsing behavior preference patterns based on the behavior preference distribution data of multiple local display time periods.

3. The data interactive display method of a digital exhibition hall according to claim 2 is characterized in that: Extracting the preferred interactive operation features under each browsing behavior preference mode, and determining the first interactive display strategy for multiple display devices in the exhibition hall according to the time period distribution patterns corresponding to the multiple browsing behavior preference modes, including: Extracting browsing interaction data of multiple users in each browsing behavior preference mode from the historical display interaction behavior data, determining the user's interactive operation preference data according to the user's browsing interaction data, and generating the preferred interactive operation features in the browsing behavior preference mode according to the interactive operation preference data of multiple users in the browsing behavior preference mode, including a group preference list for multiple display interactive operations; The browsing interaction data includes individual preference scores corresponding to multiple display interaction operations, and the group preference score of each display interaction operation is generated by fusing the interaction operation preference data of multiple users, and the multiple display interaction operations are sorted according to the group preference scores to obtain a group preference list; According to the time period distribution pattern of multiple browsing behavior preference patterns, the preferred interactive operation characteristics under each browsing behavior preference pattern are analyzed, and the allocation parameters of each browsing behavior preference pattern in the local display time period are determined according to the behavior preference distribution data of the local display time period. According to the allocation parameters, an allocation strategy for a preset number of interactive operations is determined, including the allocated number of interactive operations for each browsing behavior preference pattern. According to the preferred interactive operation characteristics and the allocated number of interactive operations under the browsing behavior preference patterns, local display interactive operation recommendation schemes for multiple browsing behavior preference patterns in the local display time period are determined, and according to the local display interactive operation recommendation schemes under the multiple local display time periods, a first interactive display strategy under the global display time period is generated.

4. The data interactive display method of a digital exhibition hall according to claim 3 is characterized in that: The content environment correlation features are extracted and a content environment analysis model is constructed. The second interactive display strategy for multiple display devices in the exhibition hall is generated through the content environment analysis model, including: Extract multiple groups of display environment features of each display device from the historical display environment data, extract the exhibit type heat data of each group of display environment features from the historical display interaction behavior data according to the time information of the display environment features, associate each group of display environment features with the corresponding exhibit type heat data to construct a sample data set, use the multiple groups of display environment features in the sample data set as the input of the content environment analysis model, use the exhibit type heat data corresponding to the multiple groups of display environment features in the sample data set as the training target of the content environment analysis model, and obtain the content environment analysis model through training with the sample data set; After acquiring target display environment data about multiple display devices, the target display environment data is input into the trained content environment analysis model to generate exhibit type heat distribution data about the multiple display devices, and a second interactive display strategy for the multiple display devices is generated according to the target heat distribution data.

5. The data interactive display method of a digital exhibition hall according to claim 4 is characterized in that: The first interactive display strategy and the second interactive display strategy are integrated to generate a target interactive display strategy, and interactive display optimization is performed on multiple display devices in the exhibition hall based on the target interactive display strategy, including: According to the heat distribution data of exhibit types about multiple display devices, multiple recommended exhibit types are set for each display device. For the target display device, after determining the target exhibit currently browsed by the user, according to the local display time period to which the current moment belongs, a local display interactive operation recommendation scheme for the target display device is determined, and according to the local display interactive operation recommendation scheme, multiple recommended display interactive operations for the target display device are set.

6. The data interactive display method of a digital exhibition hall according to claim 5 is characterized in that: Extract the behavioral synergistic influence features of each browsing behavior preference pattern and generate an interactive synergistic optimization strategy for each browsing behavior preference pattern, including: Segment multiple groups of candidate collaborative interaction data from real-time display interaction behavior data, perform time-series correlation screening on multiple groups of candidate collaborative interaction data to determine multiple groups of target collaborative interaction data, extract the interaction operation evolution path of each user in each group of target collaborative interaction data, perform collaborative impact analysis on multiple interaction operation evolution paths of each group of target collaborative interaction data, and calculate the behavioral collaborative impact parameters corresponding to multiple display interaction operations under each group of target collaborative interaction data; Browsing behavior preference pattern matching is performed on each interactive operation evolution path, pattern matching results are generated, and an operation synergy influence set under each browsing behavior preference pattern is constructed, including multiple behavior synergy influence parameters corresponding to each display interactive operation, multiple interactive synergy influence operations of the browsing behavior preference pattern are determined according to the operation synergy influence set, and an interactive synergy optimization strategy for the browsing behavior preference pattern is generated according to the multiple interactive synergy influence operations.

7. A data interactive display system for a digital exhibition hall, characterized in that: The system is used to implement the data interactive display method of a digital exhibition hall as described in any one of claims 1 to 6, comprising: A browsing behavior preference analysis module is used to collect historical display interaction behavior data of multiple display devices in the exhibition hall, extract browsing behavior preference features corresponding to multiple users from the historical display interaction behavior data, perform browsing behavior preference analysis on multiple users based on the browsing behavior preference features, and determine multiple browsing behavior preference patterns; A time period distribution analysis module is used to determine the browsing behavior label corresponding to each user according to the browsing behavior preference pattern to which the user belongs, extract browsing time period preference features of multiple browsing behavior preference patterns according to the browsing behavior label, perform browsing time period distribution analysis on multiple browsing behavior preference patterns according to the browsing time period preference features, and determine the time period distribution pattern of multiple browsing behavior preference patterns; A first interactive display optimization module is used to perform time period behavior association analysis on historical display interactive behavior data according to time period distribution patterns corresponding to multiple browsing behavior preference patterns, extract the preferred interactive operation features under each browsing behavior preference pattern, and determine a first interactive display strategy for multiple display devices in the exhibition hall according to the time period distribution patterns corresponding to the multiple browsing behavior preference patterns; A second interactive display optimization module is used to collect historical display environment data of multiple display devices, perform content environment association analysis on historical display interaction behavior data based on the historical display environment data, extract content environment association features and construct a content environment analysis model, and generate a second interactive display strategy for multiple display devices in the exhibition hall through the content environment analysis model; An interactive display management module, used to merge the first interactive display strategy and the second interactive display strategy to generate a target interactive display strategy, and optimize the interactive display of multiple display devices in the exhibition hall based on the target interactive display strategy; The interactive collaborative optimization module is used to obtain the real-time display interactive behavior data of each display device, perform user behavior collaborative impact analysis on the real-time display interactive behavior data, extract the behavioral collaborative impact characteristics of each browsing behavior preference pattern, generate an interactive collaborative optimization strategy for each browsing behavior preference pattern, and perform interactive display collaborative optimization on multiple display devices based on the interactive collaborative optimization strategy.

Images