Multimedia interactive display method and system for exhibition hall

Abstract

The invention relates to the technical field of exhibition hall display, and discloses a multimedia interactive display method and system for an exhibition hall. The system comprises an interactive data acquisition module, a display partition module, a resource mapping module, a content combination module, a dynamic updating module and a display management module. The interactive data acquisition module acquires behavior data and preference data of tourists in an exhibition hall; the display partitioning module partitions display contents according to themes and types on the basis of the interactive information; the resource mapping module establishes a corresponding relationship between the display partition and the multimedia resource library; the content combination module extracts key resources and calculates the resource association degree to generate a display combination; the dynamic updating module updates the mapping relation according to the tourist real-time interaction data; and the display management module determines a display sequence and a target according to the updated mapping relationship, and generates a display plan. The system can improve the suitability of display contents and tourist demands, and is suitable for display scenes of various exhibition halls.

Description

Technical Field

[0001] This invention relates to the field of exhibition hall display technology, specifically to a multimedia interactive display method and system for exhibition halls. Background Technology

[0002] In the current operation of exhibition halls, the display system, as the core carrier for conveying information and attracting visitors, is increasingly seeing its operational model adapted to visitor needs, becoming a key factor influencing the overall exhibition experience. Traditional exhibition hall display systems often employ fixed content presentation methods, with display zones typically determined at the initial stage of exhibition hall construction, making subsequent adjustments difficult based on actual visitor interaction. Under this model, the correspondence between display content and multimedia resources is relatively simplistic, often consisting of pre-set, fixed resource combinations, failing to dynamically optimize based on the behavioral characteristics and preferences of different visitor groups. From a visitor experience perspective, traditional systems cannot accurately capture visitor interaction information within the exhibition hall, such as behavioral data like visitor dwell time and frequency of interaction with specific exhibits, as well as preference data expressed through feedback devices. Due to the lack of effective utilization of this data, exhibition zones remain fixed, potentially leading to aesthetic fatigue for some visitors due to repetitive content, while content of interest to other visitors fails to be fully presented. In terms of resource utilization, traditional systems lack a flexible mapping mechanism between their multimedia resource libraries and exhibition areas; once the correspondence between resources and areas is established, it is difficult to modify. This prevents a large number of high-quality multimedia resources from being allocated in a timely manner according to visitor needs, resulting in resource waste and failing to fully realize the display value of the resources. Furthermore, traditional systems lack the ability to update the mapping relationship based on real-time visitor interaction data, leaving the displayed content in a static state, unable to keep up with the dynamic changes in visitor needs, further reducing the attractiveness of the exhibition hall and visitor participation. From the perspective of exhibition hall operation efficiency, traditional systems require staff to manually adjust display zones and resource combinations, which not only consumes a lot of manpower and time, but also the adjustment effect often depends on the experience and judgment of staff, making it impossible to guarantee the accuracy and timeliness of the adjustments. As visitors' demands for the exhibition hall experience continue to increase, the limitations of traditional display systems are becoming increasingly apparent, and there is an urgent need for a new type of display system that can achieve interactive data collection, dynamic zone adjustment, and flexible resource mapping to meet the needs of modern exhibition hall operation. Summary of the Invention

[0003] The purpose of this invention is to provide a multimedia interactive display method and system for exhibition halls, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides a multimedia interactive display system for exhibition halls, the system comprising: The interactive data collection module is used to acquire interactive information of visitors within the exhibition hall, including behavioral data and preference data; The display partitioning module is used to partition the displayed content according to themes and types based on the interactive information. The resource mapping module is used to map the display partitions to the multimedia resource library, establishing a correspondence between the partitions and the resources; The content combination module is used to extract key resources from the mapping relationship, calculate the correlation between multiple resources, and generate a display combination; The dynamic update module is used to update the mapping relationship between the display area and the multimedia resource library based on the real-time interaction data of tourists; The display management module is used to determine the display order and display objectives based on the updated mapping relationship, and to generate a display plan.

[0005] Preferably, the interactive data acquisition module is implemented in the following ways: For any interactive information from a tourist, obtain the corresponding classification model for that interactive information; Use a classification model to classify interactive information to obtain at least one interactive category; Identify keywords and related terms in the interactive information under the corresponding interactive category to form an interactive sample set; By analyzing the keywords and related words in the interactive sample set, keyword areas and related word areas are obtained, which are used as part of the interactive information.

[0006] Preferably, the implementation methods for obtaining the keyword area and related word area of ​​the interactive sample set also include: The keywords and related terms in the interaction sample set are merged according to the interaction category to obtain multiple merged results; Extract keyword pairs from the merged results and compare them with the keyword dictionary to obtain the keyword region; Extract the association strength of related words in the merged results, and divide the merged results according to the association strength to obtain the related word region.

[0007] Preferably, the implementation of the display partition module further includes: The display zones are evaluated, and the frequency of visit and dwell time of visitors in the display zones are analyzed. The display zones are then fitted according to the frequency of visit and dwell time to construct a mapping relationship between the display zones and visitors' interests.

[0008] Preferably, the resource mapping module is implemented in the following ways: The system retrieves the multimedia resources and historical display records corresponding to the display partition, generating multiple unlabeled resource identification results. Determine whether multiple unlabeled resource identification results are the target resource identification results. If so, treat the target resource identification results as the multimedia resource library of the display partition.

[0009] Preferably, the implementation method for establishing the mapping relationship between the display partition and the multimedia resource library includes: using the information representation of the keyword area and the related word area existing in the display partition, the description information and categories of the multimedia resource library, to construct a mapping relationship between the display partition and the multimedia resource library.

[0010] Preferably, the content combination module is implemented in the following ways: The display partitions and multimedia resource library are clustered according to content type, resource format, and display function. The largest cluster center after the clustering analysis is set as the key resource. Extract keywords from key resources, calculate the similarity between keywords, and set common sequences related to the similarity between keywords; By utilizing the common sequences related to the similarity between keywords, resources existing in the common sequences are extracted, and the matching degree between each resource is set; The matching degree between various resources is determined by setting the display combination according to the popularity distribution of each resource.

[0011] Preferably, the dynamic update module is implemented in the following ways: Extract the popularity distribution of each resource from the display combination; set the target path of the display combination according to the time period corresponding to the popularity distribution of each resource; The target paths of each resource in the display combination are fitted to obtain the fitted target paths. The probability value of the fitted target paths in each time period is set as the probability of the display combination appearing. The probability of the displayed combination is compared with the references in the multimedia resource library, the difference value is identified, and the resources in the multimedia resource library are classified according to the difference value to complete the update of the multimedia resource library.

[0012] Preferably, the display management module is implemented as follows: Based on the updated mapping relationship between the display partitions and the multimedia resource library, display targets are extracted from the interactive information, and the display targets are sorted according to their probability of occurrence to obtain the display order; The presentation plan is obtained by combining the presentation objectives and presentation order in a structured manner.

[0013] Preferably, the present invention also includes a multimedia interactive display method for exhibition halls, comprising all the modules and method flows of the aforementioned multimedia interactive display system for exhibition halls.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This multimedia interactive display system for exhibition halls, through its interactive data collection module, comprehensively acquires visitor interaction information within the hall, including behavioral and preference data. This allows the system to accurately understand visitor needs and interests, avoiding the problem of disconnected content from visitor needs caused by traditional systems that lack effective perception of visitor demands. Based on this collected interactive information, the display zoning module can categorize the display content according to themes and types, making the content classification more aligned with visitors' actual interests. Visitors can more quickly find display areas of interest, reducing wasted browsing time within the hall and improving visitor efficiency and overall experience. The resource mapping module maps display zones to a multimedia resource library, establishing a correspondence between zones and resources. This breaks the limitation of the fixed correspondence between resources and zones in traditional systems. This mapping allows resources in the multimedia resource library to be rationally allocated according to the needs of the display zones, avoiding resource waste and ensuring that each display zone is equipped with matching high-quality resources, enhancing its attractiveness. The content combination module extracts key resources from the mapping relationship and calculates the correlation between multiple resources to generate display combinations. This integrates highly related resources for display, allowing visitors to obtain more coherent and systematic information during browsing, deepening their understanding and memory of the content, rather than seeing only scattered and isolated resource displays as in traditional systems, thus improving the effectiveness of information delivery. The dynamic update module updates the mapping relationship between the display areas and the multimedia resource library based on real-time visitor interaction data, enabling the system to respond promptly to changes in visitor needs. When a visitor's interests shift, the system can update the mapping relationship to allocate corresponding multimedia resources to the relevant display areas, ensuring that the display content remains synchronized with the visitor's real-time needs. This avoids the aesthetic fatigue that can occur in traditional systems due to static and unchanging content, maintaining visitors' sense of novelty and enthusiasm for the exhibition. The exhibition management module determines the exhibition order and objectives based on the updated mapping relationships, generating an exhibition plan that makes the entire exhibition process more orderly and organized. This module optimizes the exhibition order by combining real-time updated mapping relationships, ensuring that visitors can obtain information in a logical sequence during their browsing process. It also clarifies the exhibition objectives of each section, ensuring that every exhibition segment revolves around these objectives, thus improving the relevance and effectiveness of the exhibition. The entire system operates without requiring extensive manual operation by staff, reducing labor and time costs, improving the operational efficiency of the exhibition hall, enabling it to provide higher-quality exhibition services to visitors more efficiently, enhancing its competitiveness among similar venues, and better meeting the modern exhibition hall's demand for personalized and intelligent exhibitions. Attached Figure Description

[0015] Figure 1 This is a sequence diagram of the multimedia interactive display system for exhibition halls described in this invention; Figure 2 This is a diagram illustrating the working principle of the interactive data acquisition module. Figure 3 A diagram illustrating the working principle of the partitioning module's implementation is provided. Figure 4 This is a diagram illustrating the working principle of the resource mapping module implementation. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1 This invention provides a multimedia interactive display method and system for exhibition halls, the system comprising: The interactive data acquisition module is responsible for acquiring visitor interaction information within the exhibition hall, including behavioral and preference data. The display zoning module categorizes the display content by theme and type based on the interaction information. The resource mapping module maps the display zones to a multimedia resource library, establishing a correspondence between zones and resources. The content combination module extracts key resources from the mapping relationship, calculates the correlation between multiple resources, and generates display combinations. The dynamic update module updates the mapping relationship between the display zones and the multimedia resource library based on real-time visitor interaction data. The display management module determines the display order and objectives based on the updated mapping relationship and generates a display plan. The entire system achieves dynamic adjustment and personalized display of content through modular collaboration.

[0018] Example 1: See Figure 2 When a visitor interacts with an interactive touchscreen displaying dinosaur fossils in the exhibition hall, their behavioral data (such as clicking on the "Tyrannosaurus Rex" 3D model to rotate and view it, or spending a considerable amount of time in front of the "Cretaceous Ecosystem" video) and preference data (such as subsequently entering "Jurassic large carnivorous dinosaurs" into the search box) are recorded by the system, forming a raw, multimodal interactive information record. For such an interactive information, the system first retrieves its corresponding classification model. These classification models are machine learning models pre-trained based on a large amount of exhibition-themed data, and may include text classification models for identifying topics of interest, and behavioral analysis models for determining behavioral intent. The system automatically selects and calls one or more appropriate models based on the data type of the interactive information (such as text input, touch behavior sequence, and dwell time). In this example, the text classification model processes the search term "Jurassic large carnivorous dinosaurs," while the behavioral analysis model analyzes the click and dwell behavior sequences. After processing the current interactive information, these classification models output one or more interactive category labels. Text models may output topic categories such as "paleontology," "dinosaurs," and "predators"; behavioral models may output behavioral intention categories such as "deep exploration" and "content consumption." These categories provide a framework and direction for subsequent detailed analysis.

[0019] The system performs in-depth textual and semantic analysis on each identified interaction category. For text content, natural language processing techniques are used to identify core keywords and closely related conjunctions. For behavioral data, it is transformed into analyzable metadata; for example, a clicked "Tyrannosaurus Rex" model object has its associated tag set, and a watched video has its title and keyword list. All these words and tags extracted from the raw information together constitute an initial interaction sample set for that interaction event. The sample set of a single interaction is sparse and noisy. Therefore, the implementation method includes key merging and analysis steps. The system merges the sample set generated by the current interaction with historical interaction sample sets generated by the same visitor or other visitors with similar category tags. This process is not a simple superposition, but rather a merging and aggregation according to interaction category. For example, keywords and conjunctions generated by all interactions tagged with "dinosaur" are merged together. After merging, a larger vocabulary set that better reflects the group's interest patterns is formed.

[0020] The system extracts high-frequency, stable word pairs from this merged set and compares them with a pre-built keyword dictionary covering the knowledge domains of the exhibition hall. Word pairs that highly match dictionary entries are identified as "keyword regions" with clear directional implications, representing the explicit, core concepts of visitor interests. The system analyzes the co-occurrence relationships and statistical association strength between words in the word set. For example, "Tyrannosaurus Rex" and "Rex" may always appear together, indicating a very high association strength; "Cretaceous" and "extinction events" may also show a strong statistical association. Based on preset association strength thresholds, the system divides these closely related word groups into different "associative word regions." Each associated word region reveals some potential, related thematic clusters or conceptual networks within visitor interests.

[0021] The raw, unstructured visitor interaction information is transformed into structured data representations containing "keyword areas" and "association word areas." These enriched areas accurately depict the visitor's current and potential areas of interest and knowledge exploration paths. This structured data provides high-quality, computable input for downstream display zoning and resource mapping modules, enabling the system to truly understand visitor intent and laying a solid data foundation for subsequent personalized content combinations and dynamic displays. The entire process is automated and continuously executed, ensuring real-time and accurate analysis and representation of every interaction from every visitor.

[0022] Example 2: See Figure 3 The system demonstrates the operational mechanism of the exhibition zoning module. This module is responsible for zoning the exhibition content based on visitor interaction information and further analyzing visitor behavior data for these zones to construct a mapping relationship between zones and visitor interests. The entire implementation process is based on continuous monitoring and analysis of zone access frequency and dwell time, aiming to dynamically adapt the exhibition zoning settings to the actual interest patterns of visitors. During system initialization, the physical or logical space of the exhibition hall has been divided into multiple exhibition zones, each organized around a core theme or content type. For example, a large natural history museum might have main zones such as a "Paleontological Evolution Hall," a "Modern Ecological Diversity Corridor," a "Geological and Mineral Exhibition Area," and a "Human-Environment Interaction Hall." Each zone contains several exhibits, such as fossil specimens, ecological landscaping, interactive screens, and artifact displays. The division of these zones is not only based on content themes but also considers the rationality of spatial layout and the smoothness of visitor flow.

[0023] The interactive data acquisition module continuously collects raw behavioral data of visitors within each zone. This data is acquired through a network of sensors deployed throughout the exhibition hall, including but not limited to: infrared or camera visitor counters located at zone entrances and in front of key exhibits to count the number of visitors entering the zone and lingering in front of specific exhibits; touch sensors attached to interactive exhibits to record the number of operations and their duration; and anonymous movement trajectory data obtained from location beacons worn by visitors or via Wi-Fi probes to calculate the overall dwell time of visitors within each zone. All sensor data is indexed by timestamps and zone IDs to form a complete visitor behavior log. These data streams are transmitted in real time to the data processing center, where, after data cleaning and preprocessing, they are aggregated and stored according to zone identifiers.

[0024] For calculating "visit frequency," the system employs a sliding time window mechanism, counting the number of unique visitors to each section at fixed time intervals (e.g., hourly, daily, or weekly). The system uses a cardinality estimation algorithm based on HyperLogLog to accurately count unique visitors while maintaining computational efficiency. For example, the system might find that the "Paleontology Evolution Hall" had 500 unique visitors during a Saturday afternoon time window, while the "Geology and Minerals Exhibition Area" had only 150 visitors during the same period. The system not only records the total number of visitors but also analyzes patterns over time; for instance, the hall's visitor numbers are lower on weekday afternoons but significantly increase on weekends when families flock in. The system builds a time-series model of visit frequency for each section to predict future visit trends.

[0025] For calculating "stay time," a precise analysis method based on trajectory data is employed. The system analyzes visitor location data sequences and uses stay point detection algorithms (such as density-based clustering algorithms) to identify the actual stay areas and times of visitors within each zone. When calculating the average duration from initial entry to final departure from a zone, abnormally short stays (such as simply passing through) are excluded. For example, the analysis found that the average stay time for visitors in the "Modern Ecobiological Diversity Corridor" was 25 minutes, while in the "Human-Environment Interaction Pavilion," due to the greater number of interactive exhibits, the average stay time reached 40 minutes. The system also analyzes the distribution characteristics of stay time, draws a histogram of stay time distribution, and identifies major stay time intervals and outliers.

[0026] The core operation of the partitioning module lies in the "fitting" process, which performs multi-dimensional comprehensive calculations on the two numerical indicators mentioned above: visit frequency and dwell time. The system employs a multi-indicator comprehensive evaluation method based on entropy weighting to calculate a single "interest heat value" for each partition. This method determines the weight of each indicator by calculating its entropy value, avoiding the bias of subjective weighting. During the calculation process, the system assigns higher weight to dwell time because longer dwell times often reflect deeper interest more than frequent short visits. For example, the "Geology and Minerals Exhibition Area" may not have a high visit frequency, but a few mineral enthusiasts who enter may stay for extremely long periods, resulting in a potentially high calculated interest heat value. The system also considers the coefficients of variation of visit frequency and dwell time to assess the stability and reliability of the data.

[0027] Based on the calculated interest popularity values ​​for all partitions, the system performs partition-level judgments. It uses the K-means clustering algorithm to divide all partitions into different interest levels, such as "high interest partitions," "medium interest partitions," and "low interest partitions." This process constructs a mapping relationship from physical partitions to abstract visitor interests. The system maintains a dynamically updated mapping table, using a Redis database to store each partition ID and its current corresponding interest level label, with reasonable expiration times set to ensure automatic data updates. This mapping relationship is not static; the system continuously monitors new interactive data streams and uses streaming computing frameworks (such as Apache Flink) to process sensor data in real time. The system periodically (e.g., every two hours) recalculates the access frequency, dwell time, and interest popularity values ​​for all partitions. A CUSUM (cumulative sum) control chart algorithm is used to detect changes in popularity values; once the popularity value of a partition exceeds a preset threshold, the system automatically updates the interest level in the mapping table. For example, a temporary "Special Exhibition: Wonders of the Deep Sea" may have a high frequency of visits and long dwell time in the early stages of its launch, and be mapped as a "high interest zone"; a few weeks later, as the novelty wears off, its data will drop, and the system will remap it as a "medium interest zone".

[0028] This module also supports finer-grained analysis, employing spatial clustering algorithms (such as DBSCAN) to identify differences in popularity among different exhibits within a given section. For example, in the "Paleontological Evolution Hall," the system might use local sensors to detect dense crowds and long dwell times in the dinosaur skeleton exhibit, while the flow of visitors to the early trilobite fossil exhibit is faster. This micro-data within each section is visualized through heatmaps, providing data support for optimizing the internal layout of each section. The system also establishes association rules between exhibits, analyzes the probability of visitors moving between different exhibits, and optimizes exhibit layout and guided tour routes.

[0029] This dynamically constructed mapping relationship becomes a key input to the exhibition hall's content management system. It enables the system to identify the most attractive content themes for visitors, providing a basis for resource allocation, visitor flow recommendations, and subsequent content combinations. Based on this mapping relationship, the system generates zone popularity reports, including real-time popularity rankings, popularity trends, and predicted future popularity directions for each zone. These reports are provided to other system modules via a RESTful API. The entire implementation embodies a closed-loop process from data collection to semantic mapping, allowing static display zones to dynamically reflect and adapt to the changing interests of the flowing visitor group.

[0030] Example 3: See Figure 4 The system establishes and maintains a dynamic, semantic correspondence between display partitions and the multimedia resource repository. This is achieved through a multi-step reasoning and matching process, beginning with extensive access to and initial identification of relevant resources for each partition. During system initialization, the resource mapping module receives the processing results from the display partition module, namely the semantic representation of each display partition, which includes keyword and related word areas generated by the interactive data acquisition module. For example, for the "Paleontological Evolution Hall" partition, its keyword area might contain "dinosaurs," "fossils," and "extinction," while its related word area might contain "excavation," "strata," and "climate change." This semantic information is stored in vectorized form, using pre-trained models such as Word2Vec or BERT to convert text into high-dimensional vector representations. Simultaneously, the system accesses a centralized multimedia resource repository built on a distributed file system, storing all available display resources. Each resource comes with rich metadata and a history of its display. All metadata is stored in Elasticsearch to support efficient full-text search and complex queries.

[0031] The first step of the module is to retrieve multimedia resources related to the current target display partition and their historical records. This retrieval is based on preliminary semantic filtering. The system uses the partition's keyword region and resource metadata for a coarse matching based on vector similarity. Cosine similarity is used to calculate the similarity between the partition vector and the resource description vector, and a low threshold is set to generate an initial set of resource candidates. Each resource in this set is considered an "unlabeled resource identification result." The next crucial step is the judgment step. The system needs to filter out the most relevant and highest-quality resources from this batch of unlabeled resource identification results. This judgment process uses a multi-factor comprehensive evaluation model, constructing an evaluation function to calculate the comprehensive matching degree between each candidate resource and the target partition. This evaluation function considers three main dimensions, and its calculation formula is as follows: ; in: Representing resources Overall matching score with the target partition. The text semantic similarity score is represented by the semantic association between the partition keywords and the resource descriptions calculated using a fine-tuned BERT model. The time freshness score is calculated based on the most recent usage time of the resource and uses an exponential decay function. ,in This represents the time difference between the current time and the most recently used time. This is the attenuation coefficient. The popularity score is calculated by combining indicators such as historical usage frequency, user dwell time, number of interactions, and user ratings, using a weighted average after min-max standardization. , , For the weighting coefficients, satisfying It can be dynamically adjusted through the management interface.

[0032] The system calculates the comprehensive matching score for all candidate resources and uses an adaptive threshold setting method to dynamically set the threshold based on the overall quality distribution of the current resource library. Any resource whose matching score exceeds this threshold is considered a target resource identification result. The system also considers resource diversity to avoid selecting too many similar resources. By calculating the similarity matrix between resources, it ensures that the final selected resource set is both relevant and diverse. The final step in establishing the mapping relationship is to persistently associate these target resources with the display partitions. This is achieved through nearest neighbor search based on vector space. The vectorized representations of the keyword and related word regions of the partition are used as query vectors, and the vectorized representations of the target resource's description information and category labels are used as database vectors. The proximity of the two in the vector space is efficiently calculated using an approximate nearest neighbor search library. The system records the matching strength of each resource with the partition; this strength value is the normalized result of the comprehensive matching score. This mapping relationship is usually stored in the form of a graph database, establishing a "partition-resource" relationship graph, where the edge weights represent the matching strength.

[0033] Example 4: Displaying a partition and its mapped multimedia resource library. Taking a display partition named "Mysteries of Marine Ecology" as an example, its mapped resource library may contain the following resources: a 4K high-definition documentary video about coral reefs (Resource ID: V_001), an interactive marine food chain simulation software (Resource ID: S_002), a collection of high-definition photos of rare deep-sea creatures (Resource ID: P_003), an expert interpretation article on ocean acidification (Resource ID: T_004), and an immersive experience program that allows visitors to virtually drive a submarine to explore underwater mountains (Resource ID: I_005). The module first performs cluster analysis, reading the metadata of all resources and automatically grouping them according to three preset dimensions: content type (e.g., video, software, image, text, immersive experience), resource format (e.g., MP4, EXE, JPEG, PDF, VR), and display function (e.g., information delivery, interactive operation, visual appreciation, knowledge deepening, experience simulation). The analysis process uses an unsupervised machine learning algorithm to calculate the distance of all resources in these dimensional feature spaces, grouping resources with similar features into the same cluster. For example, the algorithm might cluster V_001 (video, MP4, information delivery) and P_003 (image set, JPEG, visual appreciation) into one category based on the "visual media" feature; while clustering S_002 (software, EXE, interactive operation) and I_005 (program, VR, experience simulation) into another category based on the "interactive experience" feature; and T_004 (text, PDF, knowledge deepening) might temporarily form its own group.

[0034] After clustering, the system calculates the centroid of each cluster and designates the resource corresponding to the centroid of the cluster with the most resource members or the most representative cluster as the key resource for this analysis. Assuming the "Interactive Experience" cluster contains S_002 and I_005, and its centroid features are closest to S_002, then resource S_002 (marine food chain simulation software) is established as a key resource in the current partition resource library. The module extracts core keywords from the metadata of the key resource S_002, such as "food chain," "energy flow," "predator," and "plankton." The system also extracts keywords from all other resources, for example, "coral reef," "symbiosis," and "biodiversity" from V_001, and "submarine," "exploration," and "topography" from I_005. Next, the system calculates the semantic similarity between the keywords of the key resource and the keywords of each other resource. This calculation is based on a word vector model, evaluating the proximity of words in the semantic space. Through analysis, the system may find that "food chain" has a high semantic association with "symbiosis" (from V_001) and also has a certain conceptual relevance with "exploration" (from I_005).

[0035] Based on these pairwise similarity relationships, the system aims to find a public sequence that maximally covers these highly relevant keywords. This sequence is essentially a content theme or narrative thread, such as "life interactions and energy exploration in the ocean." This public sequence acts as a filter, allowing the system to extract resources from the resource library whose keywords have the highest match with this sequence. Next, the system calculates the pairwise match between these extracted resources, evaluating their logical coherence and thematic consistency. For example, V_001 (coral reef ecosystem) and S_002 (food chain simulation) have a high match, while their match with I_005 (submarine exploration) might be moderate. The module needs to sort and combine these resources into an attractive presentation. The system incorporates historical popularity distribution data for each resource, such as average user engagement, dwell time, or number of interactions over the past week. The presentation arrangement prioritizes arranging highly matched and popular resources adjacently to create a smooth user experience; it also considers strategically placing highly popular but moderately matched resources within the sequence.

[0036] Table 1: Content Combination Analysis Process.

[0037] Example 5: The dynamic update module begins with a deep analysis of the currently generated display combinations. This module extracts the popularity distribution data embedded in each resource from the output of the content combination module. Popularity distribution is a multi-dimensional metric that includes not only the historical access frequency of resources but also time-based patterns, such as the intensity of a resource's popularity during a specific time of day (e.g., afternoon), a specific date of the week (e.g., weekend), or a specific seasonal exhibition. The system uses time series analysis to decompose the popularity data into trend components, seasonal components, and residual components, thereby more accurately capturing the regularity of popularity changes. This popularity distribution data is correlated with precise time periods, outlining an expected "target path" for each display combination. This path depicts the expected trajectory of the combination being called and displayed by the system in various future time periods under ideal conditions.

[0038] The next step is to integrate and refine these expected paths. The system does not view the target path of a single display combination in isolation, but rather comprehensively fits the target paths of all currently active display combinations. The fitting process employs a Kalman filter-based data fusion algorithm, which effectively handles noise and uncertainty between multiple predicted paths, extracting the most probable future display trend from a large number of individual predictions. Through fitting, a comprehensive target path representing the system's global intent is obtained. For this fitted path, the system uses Monte Carlo simulation to assign a probability value to it for each future time period (such as the next one-hour window). This probability value is called the "probability of occurrence of a display combination," which quantifies the likelihood that a specific display combination will be actually triggered within that time period.

[0039] The dynamic update module initiates a comparison and verification process, meticulously comparing the calculated occurrence probabilities of each display combination with the actual citation data recorded in the multimedia resource library. Actual citation data includes objective indicators such as the recent actual number of times the resource has been accessed, its actual position in the display sequence, and the completion rate of being fully viewed by visitors. The comparison process employs statistical hypothesis testing methods, calculating the standardized residuals between observed and expected values ​​to generate a quantified "difference value." This difference value reveals the degree of deviation between the system's predictions and actual conditions, and its statistical significance. For example, a display combination themed around deep-sea exploration might be predicted by the system to have a high probability of occurrence during the evening, but actual citation data shows it is rarely triggered during that time, resulting in a positive difference value. Conversely, a combination that is not fully valued by the system may be actively used in reality, resulting in a negative difference value.

[0040] These discrepancies become key signals driving the updates of the multimedia resource library. The system reclassifies and identifies resources in the library based on the magnitude and sign of the discrepancies. The system establishes a three-layer classification system based on machine learning: the first layer categorizes resources as "overvalued resources" and "undervalued resources" based on the sign of the discrepancy; the second layer performs fine-grained grading based on the magnitude of the discrepancy; and the third layer dynamically adjusts the classification based on the duration and trend of the discrepancies. Resources with consistently large positive discrepancies indicate that the system has overestimated their attractiveness, and their priority may be lowered, or they may be assigned a lower weight in subsequent mappings. Resources with large negative discrepancies indicate that their actual popularity exceeds the system's expectations; these resources are marked as "emerging hotspots," their priority is increased, and they may be more frequently associated with relevant sections in subsequent mappings. Furthermore, the system establishes a resource lifecycle management mechanism, triggering the introduction of new resources or archiving of outdated resources based on the pattern of discrepancies, thereby completing the update of the multimedia resource library's composition and evaluation system.

[0041] The display management module begins operation after the dynamic update module completes its work. It receives the updated mapping relationship between display zones and the multimedia resource library, a dynamic view reflecting the latest visitor interests and resource attractiveness. This module first identifies and extracts the core "display objectives" that require immediate response from the real-time inflow of interactive information, using natural language processing technology and behavioral pattern recognition algorithms. These objectives may originate directly from visitors' proactive queries, or from potential points of interest inferred from behavioral data analysis, or even include display demands predicted based on spatiotemporal context.

[0042] Based on the occurrence probability data provided by the dynamically updated module, the module performs multi-objective optimization ranking on all identified display targets. The ranking algorithm considers not only occurrence probability but also the diversity, novelty, and educational value of the display targets. Targets with higher occurrence probabilities are considered to better align with current overall visitor interest trends and are therefore given higher priority, thus securing a more prominent position in the display order. Simultaneously, the system employs an exploration-exploitation strategy, reserving a certain percentage of opportunities for display targets with lower occurrence probabilities but potential value or novelty, to maintain the diversity and exploratory nature of the display content.

[0043] The module encapsulates the sorted sequence of display targets and their display order in a structured data format, generating a clear "display plan" that can be directly parsed and executed by the display terminal devices. This plan, encoded in JSON-LD format, not only specifies in detail which content resources should be presented sequentially at a specific time, for a specific audience, or in a specific context, but also includes meta-information such as suggestions for transition effects between resources, display duration guidelines, and alternative solutions. After the plan is generated, it is distributed to each display terminal via a message queue. The terminal devices make appropriate adaptive adjustments based on their local context (such as the current number of viewers, device status, etc.) to achieve refined guidance for the visitor experience.

[0044] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0045] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multimedia interactive display system for exhibition halls, characterized in that, include: The interactive data collection module is used to acquire interactive information of visitors within the exhibition hall, including behavioral data and preference data; The display partitioning module is used to partition the displayed content according to themes and types based on the interactive information. The resource mapping module is used to map the display partitions to the multimedia resource library, establishing a correspondence between the partitions and the resources; The content combination module is used to extract key resources from the mapping relationship, calculate the correlation between multiple resources, and generate a display combination; The dynamic update module is used to update the mapping relationship between the display area and the multimedia resource library based on the real-time interaction data of tourists; The display management module is used to determine the display order and display objectives based on the updated mapping relationship, and to generate a display plan.

2. The multimedia interactive display system for exhibition halls according to claim 1, characterized in that, The interactive data acquisition module is implemented in the following ways: For any interactive information from a tourist, obtain the corresponding classification model for that interactive information; Use a classification model to classify interactive information to obtain at least one interactive category; Identify keywords and related terms in the interactive information under the corresponding interactive category to form an interactive sample set; By analyzing the keywords and related words in the interactive sample set, keyword areas and related word areas are obtained, which are used as part of the interactive information.

3. The multimedia interactive display system for exhibition halls according to claim 2, characterized in that, Other methods for obtaining the keyword and related word regions of the interactive sample set include: The keywords and related terms in the interaction sample set are merged according to the interaction category to obtain multiple merged results; Extract keyword pairs from the merged results and compare them with the keyword dictionary to obtain the keyword region; Extract the association strength of related words in the merged results, and divide the merged results according to the association strength to obtain the related word region.

4. The multimedia interactive display system for exhibition halls according to claim 1, characterized in that, The implementation of the display partition module also includes: The display zones are evaluated, and the frequency of visit and dwell time of visitors in the display zones are analyzed. The display zones are then fitted according to the frequency of visit and dwell time to construct a mapping relationship between the display zones and visitors' interests.

5. The multimedia interactive display system for exhibition halls according to claim 1, characterized in that, The resource mapping module is implemented in the following ways: The system retrieves the multimedia resources and historical display records corresponding to the display partition, generating multiple unlabeled resource identification results. Determine whether multiple unlabeled resource identification results are the target resource identification results. If so, treat the target resource identification results as the multimedia resource library of the display partition.

6. The multimedia interactive display system for exhibition halls according to claim 3, characterized in that, The methods for establishing a mapping relationship between display zones and multimedia resource libraries include: using the information representation of keyword areas and related word areas in the display zones, and the descriptive information and categories of the multimedia resource libraries, to construct a mapping relationship between the display zones and the multimedia resource libraries.

7. The multimedia interactive display system for exhibition halls according to claim 1, characterized in that, The content combination module is implemented in the following ways: The display partitions and multimedia resource library are clustered according to content type, resource format, and display function. The largest cluster center after the clustering analysis is set as the key resource. Extract keywords from key resources, calculate the similarity between keywords, and set common sequences related to the similarity between keywords; By utilizing the common sequences related to the similarity between keywords, resources existing in the common sequences are extracted, and the matching degree between each resource is set; The matching degree between various resources is determined by setting the display combination according to the popularity distribution of each resource.

8. The multimedia interactive display system for exhibition halls according to claim 7, characterized in that, The implementation methods of the dynamic update module include: Extract the popularity distribution of each resource from the display combination; set the target path of the display combination according to the time period corresponding to the popularity distribution of each resource; The target paths of each resource in the display combination are fitted to obtain the fitted target paths. The probability value of the fitted target paths in each time period is set as the probability of the display combination appearing. The probability of the displayed combination is compared with the references in the multimedia resource library, the difference value is identified, and the resources in the multimedia resource library are classified according to the difference value to complete the update of the multimedia resource library.

9. The multimedia interactive display system for exhibition halls according to claim 1, characterized in that, The display management module is implemented as follows: Based on the updated mapping relationship between the display partitions and the multimedia resource library, display targets are extracted from the interactive information, and the display targets are sorted according to their probability of occurrence to obtain the display order; The presentation plan is obtained by combining the presentation objectives and presentation order in a structured manner.

10. A multimedia interactive display method for exhibition halls, characterized in that, It includes all modules and method flows of the multimedia interactive display system for exhibition halls as described in any one of claims 1 to 9.

Images