Exhibition conference information processing method and system

Abstract

The invention discloses an exhibition and conference information processing method and system, and relates to the technical field of exhibition and conference information processing, and the method comprises the following steps: at the initial stage of system operation, according to the expected data updating demand of an exhibition and conference, setting a predefined synchronization period to periodically execute data updating operation; in each data updating process, information issuing operation in the exhibition is monitored in real time, and all related data inputs are captured to construct an information data set. The method combines the prediction capability of a machine learning model and a self-adaptive dynamic adjustment algorithm, not only can identify the activity degree of information release in real time, but also can dynamically shorten the synchronization period under the trend of high-frequency release, so as to ensure that a high-frequency update request is processed in time, avoid missing of key information and logic conflicts, and meanwhile, improve the efficiency of updating the high-frequency update request. And a predefined synchronization period is kept under a conventional release type trend, so that system resources are saved, the operation cost is reduced, and the optimal balance between resource utilization and response efficiency is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of exhibition and conference information processing, and in particular to a method and system for exhibition and conference information processing. Background Art

[0002] The exhibition and conference information processing system is a comprehensive technical platform that integrates information collection, processing, management, display and communication functions, aiming to optimize and improve the organization and operation efficiency of exhibition activities. The system usually includes conference schedule management, exhibitor and participant information management, exhibits and booth management, on-site interactive communication, conference service reservation, online registration, data analysis and report generation. By digitizing and automating the processing of various types of information in the system, it is possible to track all aspects of the exhibition in real time, ensure the smooth progress of the exhibition activities, and provide exhibitors and participants with convenient query and interaction channels, thereby improving the exhibition experience and the accuracy and efficiency of conference management. In addition, the system can also help organizers optimize decisions and improve event effectiveness and participation through big data analysis and intelligent recommendation functions.

[0003] The real-time update function is one of the core capabilities of the exhibition and conference information processing system. It means that the system can synchronize and update information in real time according to dynamic changes during the exhibition process, and quickly deliver the latest content to relevant personnel or equipment. This function ensures that all users can obtain accurate and latest data at the first time, thereby improving the efficiency, accuracy and participation experience of exhibition activities.

[0004] The prior art has the following deficiencies: In the prior art, the information processing system for exhibitions and conferences usually relies on information publishers to manually update content and uses a fixed synchronization cycle for data update. The fixed synchronization cycle achieves periodic synchronization through a preset time interval to ensure that the system runs smoothly within the controllable range of resources and avoids frequent triggering of synchronization, which causes system overload or resource waste. However, when several information publishers frequently publish information in a short period of time (this situation is particularly common in large exhibitions or multi-venue meetings, such as multiple branch venues adjusting schedules at the same time, exhibitors updating booth information, or organizers issuing emergency notices), the fixed synchronization cycle limit may cause the system to be unable to process all update requests in a timely manner, resulting in information coverage, loss, or task queue accumulation, causing the information received by participants to be incomplete or logically conflicting. For example, when the branch venue schedule is temporarily adjusted, if the organizer updates the schedule of the main venue and the branch venue at the same time, the updated information of the branch venue may be lost due to processing delays or being overwritten, resulting in participants failing to obtain accurate information and mistakenly going to the main venue, thereby causing chaos in the event order and seriously affecting the experience of participants.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a method and system for processing exhibition and conference information, which combines the predictive ability of a machine learning model and an adaptive dynamic adjustment algorithm, and can not only identify the activeness of information release in real time, but also dynamically shorten the synchronization cycle under the "high-frequency release type" trend to ensure timely processing of high-frequency update requests and avoid omissions of key information and logical conflicts. At the same time, under the "regular release type" trend, the predefined synchronization cycle is maintained to save system resources and reduce operating costs, thereby achieving an optimal balance between resource utilization and response efficiency to solve the problems in the above-mentioned background technology.

[0007] In order to achieve the above object, the present invention provides the following technical solution: a method for processing exhibition and conference information, comprising the following steps: In the initial stage of system operation, according to the expected data update requirements of exhibitions and conferences, a predefined synchronization cycle is set to periodically perform data update operations; During each data update process, the information release operation in the exhibition is monitored in real time, all relevant data inputs are captured to build information data sets, and the information data obtained in real time is sorted out; Extract features that reflect the frequency of information release, perform statistical analysis and quantitative processing on the extracted feature data during the monitoring window period, and evaluate the frequency of information release during the current monitoring window period based on the quantified indicators; The quantified indicators are used as input variables into the pre-trained machine learning model to predict the development trend of information release; Combined with the output results of the machine learning model, the development trend of information release is divided into "high-frequency release type" and "regular release type"; For the "high-frequency release" trend, the synchronization cycle is adjusted through an adaptive dynamic adjustment algorithm based on the output results of the machine learning model, and the synchronization interval is shortened to meet the frequent update needs; for the "regular release" trend, the predefined synchronization cycle is continued to save resources.

[0008] Preferably, the predefined synchronization period is preset based on the established resource configuration and historical data analysis to ensure that the system operates stably within the controllable range of resources and meets the data update requirements in conventional scenarios.

[0009] Preferably, the specific steps of sorting out the information data acquired in real time are as follows: Preliminarily classify the information data obtained in real time by source, type or subject to ensure that different information is managed and processed independently; Remove invalid data or redundant information to ensure data accuracy and consistency; Convert the classified data into a unified structured format for subsequent analysis and processing; Add labels to the data according to the combing requirements to provide marking information for further analysis; The sorted data is stored in a database or cache according to classification and priority to ensure orderly management of data and provide fast access capabilities for subsequent processing.

[0010] Preferably, features reflecting the frequency of information release are extracted, wherein the extracted features include the degree of temporal overlap of multiple information updates and the diversity changes of information release content; statistical analysis and quantitative processing are performed on the degree of temporal overlap of the extracted multiple information updates and the diversity changes of information release content during the monitoring window period, and a time density reference value and a content change reference value are generated respectively; the frequency of information release during the current monitoring window period is evaluated by the time density reference value and the content change reference value.

[0011] Preferably, the time density reference value and content change reference value, which are quantified after the temporal overlap of multiple information updates and the diversity changes of information release content, are used as input variables and input into a pre-trained machine learning model. The release activity index is generated by the machine learning model, and the development trend of information release is predicted based on the release activity index.

[0012] Preferably, during the monitoring window period, the specific steps of statistically analyzing and quantifying the time overlap of the extracted multiple information updates to generate a time density reference value are as follows: Divide the monitoring window into N equal time units, record the number of information update events that occur in each time unit, and construct a time update sequence. The expression of the time update sequence is: ,in represents the number of information updates within the i-th time unit, and N represents the number of time units; The overlapping impact factor of each time unit in the time update sequence is calculated to reflect the influence of the information update in the time unit on the adjacent units. The overlapping impact factor is defined as: ,in, is the overlapping influence factor of the i-th time unit, comprehensively considering the information update contribution of this time unit and its adjacent time units, K is the influence range of time overlap, indicating the number of adjacent units that contribute to the current time unit, is an exponential decay factor, k represents the distance between the adjacent time unit and the current time unit, and Respectively represent and The number of information updates within a time unit; The time density reference value within the monitoring window period is calculated based on the overlapping impact factors of all time units. The calculation expression is: ,in, is a time density reference value, which is used to quantify the frequency of information release during the monitoring window period. is the weight of the time unit, is the weight normalization factor.

[0013] Preferably, during the monitoring window period, the specific steps of statistically analyzing and quantifying the diversity changes of the extracted information release content to generate a content change reference value are as follows: The information data within the monitoring window period is classified and counted, and the classification diversity and classification entropy are calculated according to the content proportion of each type of information to measure the complexity of the information category distribution. The calculation expression of classification diversity entropy is: ,in, is the total number of information categories within the monitoring window period, is the proportion of the p-th type of information, and the calculation formula is: ,in, is the number of information items of category p, is the total number of messages during the monitoring window period, is the classification diversity entropy, which reflects the distribution complexity of the categories; The fluctuation of information quantity at different time points during the monitoring window period is analyzed, and the content change rate is calculated. The set of release time points of each type of information during the window period is set as , then the calculation expression of content change rate is: ,in, is the set of information release time points of category p, is the number of time points in category p, and They are respectively The timestamps of the yth and yth messages, is the content change rate, reflecting the time compactness of information release; The classification diversity entropy and content change rate are combined to generate the final content change reference value, which is defined as: ,in, It is a reference value for content change, used to evaluate the diversity and frequency of information release. , is the adjustment parameter that controls the weight of the content change rate, and , .

[0014] Preferably, during the monitoring window period, the release activity index generated when predicting the development trend of exhibition and conference information release by the trained machine learning model is compared and analyzed with the pre-set reference threshold of the release activity index, and the development trend of information release is divided. The specific division steps are as follows: If the release activity index is greater than the release activity index reference threshold, the development trend of information release is classified as "high frequency release type"; If the publishing activity index is less than or equal to the publishing activity index reference threshold, the development trend of information release is classified as "regular release type".

[0015] Preferably, for the "high-frequency release type" trend, the synchronization period is adjusted through an adaptive dynamic adjustment algorithm according to the output results of the machine learning model. The specific steps are as follows: According to the publishing activity index and the publishing activity index reference threshold, the adjustment factor of the current synchronization period is calculated. The calculation expression of the adjustment factor is: ,in, It is the release activity index during the current monitoring window, reflecting the frequency of information release. It is generated by a machine learning model. It is the reference threshold of the release activity index, which is used to judge whether the development trend of information release is "high frequency" or "regular". is the adjustment factor of the synchronization cycle, which determines the degree of shortening of the synchronization cycle. It is a nonlinear adjustment parameter, and the more sensitive the adjustment factor is to high-frequency release; According to the predefined synchronization period and adjustment factor, the current target synchronization period is calculated. The calculation expression is: ,in, is a predefined synchronization period, indicating a fixed synchronization interval in the initial state. Indicates the current target synchronization period, which is the length of the synchronization period after dynamic adjustment. The minimum allowed value of the synchronization period, used to prevent resource exhaustion due to over-adjustment.

[0016] An exhibition and conference information processing system includes a synchronization cycle initialization module, a real-time data monitoring and combing module, an information release feature extraction and evaluation module, a machine learning prediction module, an information release trend classification module, and a synchronization cycle dynamic control module; The synchronization cycle initialization module sets a predefined synchronization cycle to periodically perform data update operations according to the expected data update requirements of the exhibition and conference at the initial stage of system operation; Real-time data monitoring and sorting module: During each data update process, it monitors the information release operation in the exhibition in real time, captures all relevant data inputs to build information data sets, and sorts out the information data obtained in real time; The information release feature extraction and evaluation module extracts features that reflect the frequency of information release, performs statistical analysis and quantitative processing on the extracted feature data during the monitoring window period, and evaluates the frequency of information release during the current monitoring window period based on the quantified indicators; The machine learning prediction module uses the quantified indicators as input variables and inputs them into the pre-trained machine learning model to predict the development trend of information release; The information release trend classification module, combined with the output results of the machine learning model, divides the development trend of information release into "high-frequency release type" and "regular release type"; The synchronization cycle dynamic control module adjusts the synchronization cycle through an adaptive dynamic adjustment algorithm based on the output results of the machine learning model for the "high-frequency release" trend, shortens the synchronization interval to meet the needs of frequent updates; while for the "regular release" trend, it continues to maintain the predefined synchronization cycle to save resources.

[0017] In the above technical solution, the technical effects and advantages provided by the present invention are: The present invention combines the predictive ability of the machine learning model and the adaptive dynamic adjustment algorithm, which can not only identify the activeness of information release in real time, but also dynamically shorten the synchronization cycle under the "high-frequency release" trend to ensure timely processing of high-frequency update requests and avoid omissions of key information and logical conflicts. At the same time, under the "regular release" trend, it maintains the predefined synchronization cycle, saves system resources, and reduces operating costs, thereby achieving an optimal balance between resource utilization and response efficiency, significantly improving the real-time, accuracy and flexibility of exhibition and conference information processing, providing efficient and reliable technical support for complex and changeable exhibition environments, improving the user experience of participants, and ensuring the smooth progress of activities. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0019] Figure 1The present invention is a method flow chart of a method for processing exhibition and conference information.

[0020] Figure 2 The present invention is a module schematic diagram of an exhibition and conference information processing system. DETAILED DESCRIPTION

[0021] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0022] The present invention provides Figure 1 A method for processing exhibition and conference information is shown, comprising the following steps: In the initial stage of system operation, according to the expected data update requirements of exhibitions and conferences, a predefined synchronization cycle is set to periodically perform data update operations; The predefined synchronization cycle is preset by the system based on the established resource configuration and historical data analysis to ensure that the system can operate stably within the controllable range of resources and meet the data update requirements in common scenarios.

[0023] When the system is running, the synchronization cycle is a fixed time interval pre-set after in-depth analysis of resource usage and historical data to control the frequency of data updates. The system will determine a reasonable synchronization cycle based on the established hardware resources (such as computing power, storage capacity, bandwidth, etc.) and the data accumulated during historical operation (such as past information release frequency, data traffic, and peak update period). This synchronization cycle is set to ensure that the system can operate efficiently in most cases, without causing resource exhaustion or system overload due to excessive update frequency, while also maintaining the timeliness of data synchronization to meet the needs of information release and update in daily exhibitions and meetings. This method provides a stable and efficient operation mode by balancing resource consumption and update requirements, laying the foundation for subsequent dynamic adjustment and intelligent optimization.

[0024] During each data update process, the information release operation in the exhibition is monitored in real time, all relevant data inputs are captured to build information data sets, and the information data obtained in real time is sorted out; During each data update process, the information release operation in the exhibition is monitored in real time to capture all relevant data inputs, such as schedule adjustments of branch venues, exhibitors updating booth information, emergency notices, etc. These data are collected and stored in real time through sensors or system logs to build information data sets, providing a basis for subsequent data combing and feature extraction, ensuring the comprehensiveness and real-time nature of the data.

[0025] The specific steps for sorting out the information data obtained in real time are as follows: The information data obtained in real time will be preliminarily classified by source, type or subject. For example, the data can be divided into categories such as sub-venue schedule adjustments, exhibitor information updates, emergency notices, etc., to ensure that different information can be managed and processed independently.

[0026] Remove invalid data or redundant information, such as duplicate content, data with irregular formats, or noisy data, to ensure data accuracy and consistency.

[0027] Convert classified data into a unified structured format, such as parsing unstructured text information into standard fields (such as time, location, and publication content) for subsequent analysis and processing.

[0028] Add labels to the data according to the combing needs, such as release frequency, urgency, relevance, etc., to provide tag information for further analysis.

[0029] The sorted data is stored in a database or cache according to classification and priority to ensure orderly management of data and provide fast access capabilities for subsequent processing.

[0030] Information data acquired in real time needs to go through a systematic combing process to ensure the accuracy and availability of the data. In actual operation, the data must first be classified and data from different sources must be managed in categories to avoid confusion; then, data quality must be improved by removing duplicate or invalid content through cleaning; then, valid data must be structured, the data format must be standardized, and key fields must be extracted; then, the data must be labeled, such as indicating the degree of urgency or update frequency, to ensure the pertinence of the analysis; finally, the cleaned and labeled data must be stored in the database to provide orderly input for subsequent analysis, quantification, and intelligent prediction. This combing process ensures that the data acquired in real time is efficient and orderly.

[0031] Extract features that reflect the frequency of information release, perform statistical analysis and quantitative processing on the extracted feature data within a monitoring window period (e.g., 1 minute or 5 minutes), and evaluate the frequency of information release within the current monitoring window period based on the quantified indicators; Features reflecting the frequency of information release are extracted, wherein the extracted features include the degree of temporal overlap of multiple information updates and the diversity changes of information release content; within the monitoring window period, the extracted degrees of temporal overlap of multiple information updates and the diversity changes of information release content are statistically analyzed and quantified; a time density reference value and a content change reference value are generated respectively; and the frequency of information release within the current monitoring window period is evaluated by the time density reference value and the content change reference value.

[0032] When the time overlap of multiple information updates is high during the release of exhibition and conference information, it indicates that the information is released relatively frequently. This is because the information overlap reflects that the system needs to process and synchronize a large number of update requests in the same time period. This situation usually occurs during peak hours, such as when multiple venues adjust their schedules at the same time, exhibitors update their booth information, or the organizer issues an emergency notice. When update tasks occur in a concentrated manner, the higher the degree of time overlap, the greater the information processing density faced by the system, and the more frequent the information release. The frequent appearance of overlapping information not only tests the processing capacity of the system, but may also place higher requirements on the timeliness and accuracy of information transmission. Therefore, by monitoring the degree of information time overlap, the high frequency of information release can be effectively evaluated.

[0033] During the monitoring window period, the specific steps of statistically analyzing and quantifying the temporal overlap of multiple information updates extracted to generate a temporal density reference value are as follows: Divide the monitoring window into N equal time units, record the number of information update events that occur in each time unit, and construct a time update sequence. The expression of the time update sequence is: ,in represents the number of information updates within the i-th time unit, and N represents the number of time units; By dividing time units and recording the number of update events, a basic time update sequence is established, laying the foundation for subsequent overlap analysis.

[0034] The overlapping impact factor of each time unit in the time update sequence is calculated to reflect the influence of the information update in the time unit on the adjacent units. The overlapping impact factor is defined as: ,in, is the overlapping influence factor of the i-th time unit, comprehensively considering the information update contribution of this time unit and its adjacent time units, K is the influence range of time overlap (in terms of the number of time units), indicating the number of adjacent units that contribute to the current time unit, is an exponential decay factor, which is used to control the influence of time units farther from the current time unit on the overlap factor to gradually decrease. k represents the distance between the adjacent time unit and the current time unit. and Respectively represent and The number of information updates within a time unit; By calculating the overlapping impact factor, the temporal overlapping information in the time update sequence is quantified, providing a basis for evaluating the temporal density.

[0035] The time density reference value within the monitoring window period is calculated based on the overlapping impact factors of all time units. The calculation expression is: ,in, is a time density reference value, which is used to quantify the frequency of information release during the monitoring window period. is the weight of the time unit, which can be defined according to specific needs (such as giving higher weight to urgent updates). is the weight normalization factor, ensuring that the numerical range of the time density reference value is reasonable; Taking into account the impact of time overlap and weight, a time density reference value is generated to quantify the intensity of information release during the monitoring window period, providing a quantitative basis for subsequent evaluation and regulation.

[0036] It can be seen from the time density reference value that during the monitoring window period, the larger the performance value of the time density reference value generated by statistical analysis and quantitative processing of the time overlap of multiple information updates extracted, the more information update activities there are during the monitoring window period, and the degree of time overlap of these updates is high, indicating that the information release is relatively intensive and frequent; on the contrary, the smaller the value of the time density reference value, the fewer information update activities during the monitoring window period, the lower the degree of time overlap, and the relatively sparse and infrequent information release. This quantitative indicator provides an accurate assessment of the frequency of information release by statistically analyzing the distribution and overlap of information updates in time.

[0037] During the release of exhibition and conference information, when the diversity of information release content changes more, it can usually be considered that the information release is relatively frequent. The diversity of information release content refers to the richness of the types, sources or topics of information released within a certain period of time. For example, if there are both schedule adjustments for branch venues and updates to exhibitor booth information and emergency notices in a short period of time, this diversity change reflects the complexity and frequency of information update needs. Higher content diversity is often accompanied by greater data traffic and more frequent update needs, so it can be used as one of the important features to measure the frequency of information release. This feature helps the system identify scenarios where information is released frequently.

[0038] During the monitoring window period, the specific steps for statistically analyzing and quantifying the diversity changes of the extracted information release content to generate content change reference values ​​are as follows: The information data within the monitoring window period is classified and counted, and the classification diversity and classification entropy are calculated according to the content proportion of each type of information to measure the complexity of the information category distribution. The calculation expression of classification diversity entropy is: ,in, is the total number of information categories within the monitoring window period, is the proportion of the p-th type of information, and the calculation formula is: ,in, is the number of information items of category p, is the total number of messages during the monitoring window period, It is the classification diversity entropy, which reflects the distribution complexity of categories. The higher the value, the more evenly distributed the categories are. By classifying diversity counts, we can preliminarily measure the complexity of information content, laying the foundation for the subsequent calculation of comprehensive changes.

[0039] The fluctuation of information quantity at different time points during the monitoring window period is analyzed, and the content change rate is calculated. The set of release time points of each type of information during the window period is set as ( ), then the calculation expression of content change rate is: ,in, is the set of information release time points of category p, is the number of time points in category p, and They are respectively The timestamps of the yth and yth messages, The content change rate reflects the time compactness of information release. The higher the value, the higher the release frequency. This step quantifies the compactness of information release in the time dimension and provides dynamic time characteristics for the content change reference value.

[0040] The classification diversity entropy and content change rate are combined to generate the final content change reference value, which is defined as: ,in, It is a reference value for content change, used to evaluate the diversity and frequency of information release. , is the adjustment parameter that controls the weight of the content change rate, and , ; The final content change reference value is generated by combining the diversity of information categories and time compactness, which fully reflects the dynamic changes and complexity of information release and provides a quantitative basis for subsequent decision-making.

[0041] It can be seen from the content change reference value that during the monitoring window period, the larger the performance value of the content change reference value generated by statistical analysis and quantitative processing of the diversity changes of the extracted information release content, the higher the diversity and change rate of the information release content during this period, reflecting that the frequency of information release is higher. This is because the content change reference value combines the classification diversity entropy (reflecting the complexity of the information category) and the content change rate (reflecting the time compactness of information release). The increase of both means that during the monitoring window period, not only the types of information are rich and diverse, but also the release frequency is significantly increased. Conversely, the smaller the performance value of the content change reference value, the more single the information release category is and the lower the release frequency is, reflecting that the information is not frequently released during the monitoring window period. Therefore, the content change reference value can be used as an effective indicator to measure the frequency of information release, and provide a scientific basis for the system to adjust the synchronization cycle and resource allocation.

[0042] The quantified indicators are used as input variables into the pre-trained machine learning model to predict the development trend of information release; The time density reference value and content change reference value, which are quantified from the temporal overlap of multiple information updates and the diversity changes in information release content, are used as input variables and input into a pre-trained machine learning model. The release activity index is generated through the machine learning model, and the development trend of information release is predicted based on the release activity index.

[0043] A trained machine learning model is a model that has a certain predictive ability after being trained with known historical data and labels. Specifically, in this scenario, the machine learning model uses the information release data and its corresponding release activity labels in the past monitoring window period to learn, and gradually grasp the correlation between the information release time density and content change characteristics and release activity. This model can receive new input data (such as time density reference value and content change reference value), and make quantitative predictions on the activity of information release based on its internal rules, thereby generating a release activity index.

[0044] During the training phase, the model uses multidimensional data from the historical monitoring window as a training set, including time density reference values ​​(reflecting the degree of overlap between multiple information updates in time) and content change reference values ​​(reflecting the dynamic changes in the diversity of information content), and uses known release activity labels (such as high activity, medium activity, and low activity) as the target variable for supervised learning. The model repeatedly adjusts parameters to minimize the error between the predicted value and the actual label, thereby extracting the complex relationship between data features and release activity. After sufficient training, the model can accurately predict the release activity under new input data, becoming an intelligent prediction tool that adapts to the current scenario.

[0045] In this scenario, the trained machine learning model acts as the core prediction engine, using input variables (time density reference value and content change reference value) to generate a publishing activity index. The time density reference value reflects the overlap and compactness of information updates in the time dimension, while the content change reference value quantifies the diversity and complexity of information release content. Together, these two describe the key characteristics of information release frequency and content dynamics, and are important determinants of release activity. By analyzing and processing these input features, the machine learning model outputs a high-precision publishing activity index, which fully reflects the overall dynamics of information release during the current monitoring window.

[0046] Based on the release activity index, the system can further predict the development trend of information release, such as whether it will enter a high-frequency release stage or whether it will stabilize. This prediction capability provides powerful intelligent support for the exhibition and conference information management system, enabling it to dynamically adjust the synchronization cycle and resource allocation to avoid information coverage, loss or delay. At the same time, the introduction of this model also improves the system's adaptive ability, enabling it to make more accurate decisions in the face of complex and changing exhibition environments, and provide guarantees for the timeliness of information updates and the efficiency of system operations.

[0047] The machine learning model is not specifically limited here, and can achieve the time density reference value and content change reference value Conduct comprehensive analysis to generate a publishing activity index In order to realize the technical solution of the present invention, the present invention provides a specific implementation method; Release Activity Index The resulting calculation formula is: , where , are the time density reference values and content change reference value The preset scaling factor of , Both are greater than 0.

[0048] The preset proportionality factor here (such as in the formula and ) refers to the calculation of the release activity index for different features (such as time density reference value and content change reference value ). These preset proportional coefficients are used to balance the importance and influence of each feature in the comprehensive analysis to reflect the actual situation of information release dynamics. The significance of the preset proportional coefficients is to flexibly adapt to the actual needs of different scenarios by adjusting the weights of different features, thereby generating a more accurate and representative release activity index. This design not only improves the interpretability of the model, but also enhances its applicability and robustness.

[0049] It can be seen from the release activity index that during the monitoring window period, the greater the performance value of the time density reference value generated by statistical analysis and quantitative processing of the temporal overlap degree of multiple extracted information updates, the greater the performance value of the content change reference value generated by statistical analysis and quantitative processing of the diversity changes of the extracted information release content, that is, the greater the performance value of the release activity index generated when the development trend of exhibition and conference information release is predicted by the trained machine learning model, the more frequent the release of exhibition and conference information during the monitoring window period, and vice versa.

[0050] Combined with the output results of the machine learning model, the development trend of information release is divided into "high-frequency release type" and "regular release type"; During the monitoring window period, the release activity index generated by the trained machine learning model when predicting the development trend of exhibition and conference information release is compared and analyzed with the pre-set reference threshold of the release activity index to divide the development trend of information release. The specific division steps are as follows: If the release activity index is greater than the release activity index reference threshold, the development trend of information release is classified as "high frequency release type"; High-frequency release refers to the release frequency of exhibition and conference information significantly higher than normal during the monitoring window period, which is characterized by extremely short information release intervals, diverse content types and rapid changes. This type of release usually occurs at the peak of exhibition activities, such as when multiple venues adjust their schedules at the same time, exhibitors centrally update their booth information or the organizer issues a series of emergency notices.

[0051] If the release activity index is less than or equal to the release activity index reference threshold, the development trend of information release is classified as "regular release type"; Regular release refers to the frequency of release of exhibition and conference information and the degree of change in content within the normal range during the monitoring window period, which is characterized by a long interval between information releases, relatively single content types and a low rate of change. This type usually occurs during the non-peak period of exhibition activities, such as when the conference schedule runs steadily, exhibitor information rarely changes, or the content released by the organizer is relatively sparse.

[0052] For the "high-frequency release" trend, the synchronization cycle is adjusted through an adaptive dynamic adjustment algorithm based on the output results of the machine learning model, shortening the synchronization interval to meet the needs of frequent updates; for the "regular release" trend, the predefined synchronization cycle is maintained to save resources; In response to the "high-frequency release" trend, the synchronization period is adjusted through an adaptive dynamic adjustment algorithm based on the output results of the machine learning model. The specific steps are as follows: According to the publishing activity index and the publishing activity index reference threshold, the adjustment factor of the current synchronization period is calculated. The calculation expression of the adjustment factor is: ,in, It is the release activity index during the current monitoring window, reflecting the frequency of information release. It is generated by a machine learning model. It is the reference threshold of the release activity index, which is used to judge whether the development trend of information release is "high frequency" or "regular". is the adjustment factor of the synchronization cycle, which determines the degree of shortening of the synchronization cycle. It is a nonlinear adjustment parameter. The larger the value, the more sensitive the adjustment factor is to high-frequency releases. By adjusting the factor Dynamically associate publishing activity with the synchronization cycle to provide a basis for optimizing and adjusting the synchronization cycle.

[0053] According to the predefined synchronization period and adjustment factor, the current target synchronization period is calculated. The calculation expression is: ,in, is a predefined synchronization period, which indicates a fixed synchronization interval in the initial state of the system. Indicates the current target synchronization period, which is the length of the synchronization period after dynamic adjustment. The minimum allowed value of the synchronization period, used to prevent excessive adjustment from causing resource exhaustion; Adjust the synchronization cycle according to the activity index of high-frequency publishing, so that the system can respond to frequent information update requests more quickly, while avoiding system resource problems caused by too short a synchronization cycle.

[0054] In response to the "high-frequency publishing" trend, the synchronization cycle is shortened through an adaptive dynamic adjustment algorithm to quickly respond to frequent information update needs, ensuring that the system can synchronize data in a high-activity publishing environment in a timely manner to avoid information delays, overwriting or loss due to long synchronization intervals. This step dynamically adjusts the synchronization cycle length so that the system can more efficiently process and distribute high-frequency update content during the monitoring window period, thereby improving the real-time and accuracy of information updates and providing users with a better user experience. In addition, adaptive adjustment avoids the problem of insufficient resource allocation in high-frequency scenarios with fixed synchronization cycles, effectively balancing system performance and resource utilization efficiency.

[0055] For the "regular release" trend, continuing to maintain the predefined synchronization cycle means that when the frequency and complexity of information release are within the normal range, the system does not need to adjust the synchronization interval, but updates the data according to the fixed time period initially set. The core purpose of this strategy is to reduce the frequent occupation of system resources by maintaining a longer synchronization cycle when the demand for information updates is relatively stable, avoid unnecessary computing overhead and bandwidth consumption, and ensure the reliability and consistency of data updates. By maintaining the predefined synchronization cycle, the system can achieve efficient use of resources while meeting current update needs, and reserve more computing and processing capabilities to cope with future high-frequency release scenarios.

[0056] When solving the problem proposed by the above scheme, a significant beneficial effect is that the real-time of information synchronization and the adaptability of the system are significantly improved, thereby greatly improving the experience of participants and the efficiency of exhibition management. Specifically, the method ensures the stable update of basic information by setting a predefined synchronization cycle in the initial stage of system operation; on this basis, the information release operation is monitored in real time, all relevant data are captured and sorted, key features are extracted, and these features are statistically analyzed and quantified during the monitoring window period to generate content change reference values. These quantitative indicators are used as input variables, and trend prediction is performed through a pre-trained machine learning model, so that the system can intelligently identify the "high-frequency release type" and "regular release type" trends of information release. When the "high-frequency release type" trend is detected, the system quickly shortens the synchronization interval through an adaptive dynamic adjustment algorithm to ensure that all update requests can be processed in a timely and efficient manner during the peak period of frequent information release, avoiding information coverage, loss or logical conflicts, ensuring that participants obtain accurate and real-time information, and preventing confusion and experience degradation caused by information errors. At the same time, for the "regular release type" trend, the system maintains a predefined synchronization cycle to optimize resource usage and avoid unnecessary resource waste. This intelligent and dynamic synchronization mechanism not only improves the real-time and accuracy of information processing, but also ensures that the system can run stably and efficiently under various information release frequencies through flexible resource management, thereby comprehensively improving the performance and user satisfaction of the exhibition and conference information processing system.

[0057] The present invention provides Figure 2 The exhibition and conference information processing system shown includes a synchronization cycle initialization module, a real-time data monitoring and combing module, an information release feature extraction and evaluation module, a machine learning prediction module, an information release trend classification module and a synchronization cycle dynamic control module; The synchronization cycle initialization module sets a predefined synchronization cycle to periodically perform data update operations according to the expected data update requirements of the exhibition and conference at the initial stage of system operation; Real-time data monitoring and sorting module: During each data update process, it monitors the information release operation in the exhibition in real time, captures all relevant data inputs to build information data sets, and sorts out the information data obtained in real time; The information release feature extraction and evaluation module extracts features that reflect the frequency of information release, performs statistical analysis and quantitative processing on the extracted feature data during the monitoring window period, and evaluates the frequency of information release during the current monitoring window period based on the quantified indicators; The machine learning prediction module uses the quantified indicators as input variables and inputs them into the pre-trained machine learning model to predict the development trend of information release; The information release trend classification module, combined with the output results of the machine learning model, divides the development trend of information release into "high-frequency release type" and "regular release type"; The synchronization cycle dynamic control module adjusts the synchronization cycle through an adaptive dynamic adjustment algorithm based on the output results of the machine learning model for the "high-frequency release" trend, shortening the synchronization interval to meet the needs of frequent updates; while for the "regular release" trend, the predefined synchronization cycle is maintained to save resources; A method for processing exhibition and conference information provided in an embodiment of the present invention is implemented by the above-mentioned exhibition and conference information processing system. The specific method and process of a conference and conference information processing system are detailed in the embodiment of the above-mentioned method for processing exhibition and conference information, which will not be repeated here.

[0058] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0059] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0060] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for processing exhibition and conference information, characterized in that: The following steps are involved: In the initial stage of system operation, according to the expected data update requirements of exhibitions and conferences, a predefined synchronization cycle is set to periodically perform data update operations; During each data update process, the information release operation in the exhibition is monitored in real time, all relevant data inputs are captured to build information data sets, and the information data obtained in real time is sorted out; Extract features that reflect the frequency of information release, perform statistical analysis and quantitative processing on the extracted feature data during the monitoring window period, and evaluate the frequency of information release during the current monitoring window period based on the quantified indicators; The quantified indicators are used as input variables into the pre-trained machine learning model to predict the development trend of information release; Combined with the output results of the machine learning model, the development trend of information release is divided into "high-frequency release type" and "regular release type"; For the "high-frequency release" trend, the synchronization cycle is adjusted through an adaptive dynamic adjustment algorithm based on the output results of the machine learning model, and the synchronization interval is shortened to meet the needs of frequent updates; while for the "regular release" trend, the predefined synchronization cycle is continued to save resources.

2. A method for processing exhibition and conference information according to claim 1, characterized in that: The predefined synchronization cycle is preset based on the established resource configuration and historical data analysis to ensure that the system runs stably within the controllable range of resources and meets the data update requirements in common scenarios.

3. A method for processing exhibition and conference information according to claim 1, characterized in that: The specific steps for sorting out the information data obtained in real time are as follows: Preliminarily classify the information data obtained in real time by source, type or subject to ensure that different information is managed and processed independently; Remove invalid data or redundant information to ensure data accuracy and consistency; Convert the classified data into a unified structured format for subsequent analysis and processing; Add labels to the data according to the combing requirements to provide marking information for further analysis; The sorted data is stored in a database or cache according to classification and priority to ensure orderly management of data and provide fast access capabilities for subsequent processing.

4. A method for processing exhibition and conference information according to claim 1, characterized in that: Features reflecting the frequency of information release are extracted, wherein the extracted features include the degree of temporal overlap of multiple information updates and the diversity changes of information release content; within the monitoring window period, the extracted degrees of temporal overlap of multiple information updates and the diversity changes of information release content are statistically analyzed and quantified; a time density reference value and a content change reference value are generated respectively; and the frequency of information release within the current monitoring window period is evaluated by the time density reference value and the content change reference value.

5. A method for processing exhibition and conference information according to claim 4, characterized in that: The time density reference value and content change reference value, which are quantified from the temporal overlap of multiple information updates and the diversity changes in information release content, are used as input variables and input into a pre-trained machine learning model. The release activity index is generated through the machine learning model, and the development trend of information release is predicted based on the release activity index.

6. A method for processing exhibition and conference information according to claim 4, characterized in that: During the monitoring window period, the specific steps of statistically analyzing and quantifying the temporal overlap of multiple information updates extracted to generate a temporal density reference value are as follows: Divide the monitoring window into N equal time units, record the number of information update events that occur in each time unit, and construct a time update sequence. The expression of the time update sequence is: ,in represents the number of information updates within the i-th time unit, and N represents the number of time units; The overlapping impact factor of each time unit in the time update sequence is calculated to reflect the influence of the information update in the time unit on the adjacent units. The overlapping impact factor is defined as: ,in, is the overlapping influence factor of the i-th time unit, comprehensively considering the information update contribution of this time unit and its adjacent time units, K is the influence range of time overlap, indicating the number of adjacent units that contribute to the current time unit, is an exponential decay factor, k represents the distance between the adjacent time unit and the current time unit, and Respectively represent and The number of information updates within a time unit; The time density reference value within the monitoring window period is calculated based on the overlapping impact factors of all time units. The calculation expression is: ,in, is a time density reference value, which is used to quantify the frequency of information release during the monitoring window period. is the weight of the time unit, is the weight normalization factor.

7. A method for processing exhibition and conference information according to claim 4, characterized in that: During the monitoring window period, the specific steps for statistically analyzing and quantifying the diversity changes of the extracted information release content to generate content change reference values ​​are as follows: The information data within the monitoring window period is classified and counted, and the classification diversity and classification entropy are calculated according to the content proportion of each type of information to measure the complexity of the information category distribution. The calculation expression of classification diversity entropy is: ,in, is the total number of information categories within the monitoring window period, is the proportion of the p-th type of information, and the calculation formula is: ,in, is the number of information items of category p, is the total number of messages during the monitoring window period, is the classification diversity entropy, which reflects the distribution complexity of the categories; The fluctuation of information quantity at different time points during the monitoring window period is analyzed, and the content change rate is calculated. The set of release time points of each type of information during the window period is set as , then the calculation expression of content change rate is: ,in, is the set of information release time points of category p, is the number of time points in category p, and They are respectively The timestamps of the yth and yth messages, is the content change rate, reflecting the time compactness of information release; The classification diversity entropy and content change rate are combined to generate the final content change reference value, which is defined as: ,in, It is a reference value for content change, used to evaluate the diversity and frequency of information release. , is the adjustment parameter that controls the weight of the content change rate, and , .

8. A method for processing exhibition and conference information according to claim 5, characterized in that: During the monitoring window period, the release activity index generated by the trained machine learning model when predicting the development trend of exhibition and conference information release is compared and analyzed with the pre-set reference threshold of the release activity index to divide the development trend of information release. The specific division steps are as follows: If the release activity index is greater than the release activity index reference threshold, the development trend of information release is classified as "high frequency release type"; If the publishing activity index is less than or equal to the publishing activity index reference threshold, the development trend of information release is classified as "regular release type".

9. A method for processing exhibition and conference information according to claim 8, characterized in that: In response to the "high-frequency release" trend, the synchronization period is adjusted through an adaptive dynamic adjustment algorithm based on the output results of the machine learning model. The specific steps are as follows: According to the publishing activity index and the publishing activity index reference threshold, the adjustment factor of the current synchronization period is calculated. The calculation expression of the adjustment factor is: ,in, It is the release activity index during the current monitoring window, reflecting the frequency of information release. It is generated by a machine learning model. It is the reference threshold of the release activity index, which is used to judge whether the development trend of information release is "high frequency" or "regular". is the adjustment factor of the synchronization cycle, which determines the degree of shortening of the synchronization cycle. It is a nonlinear adjustment parameter, and the more sensitive the adjustment factor is to high-frequency release; According to the predefined synchronization period and adjustment factor, the current target synchronization period is calculated. The calculation expression is: ,in, is a predefined synchronization period, indicating a fixed synchronization interval in the initial state. Indicates the current target synchronization period, which is the length of the synchronization period after dynamic adjustment. The minimum allowed value of the synchronization period, used to prevent resource exhaustion due to over-adjustment.

10. A conference information processing system, used to implement the conference information processing method according to any one of claims 1 to 9, characterized in that: It includes synchronization cycle initialization module, real-time data monitoring and combing module, information release feature extraction and evaluation module, machine learning prediction module, information release trend classification module and synchronization cycle dynamic control module; The synchronization cycle initialization module sets a predefined synchronization cycle to periodically perform data update operations according to the expected data update requirements of the exhibition and conference at the initial stage of system operation; Real-time data monitoring and sorting module: During each data update process, it monitors the information release operation in the exhibition in real time, captures all relevant data inputs to build information data sets, and sorts out the information data obtained in real time; The information release feature extraction and evaluation module extracts features that reflect the frequency of information release, performs statistical analysis and quantitative processing on the extracted feature data during the monitoring window period, and evaluates the frequency of information release during the current monitoring window period based on the quantified indicators; The machine learning prediction module uses the quantified indicators as input variables and inputs them into the pre-trained machine learning model to predict the development trend of information release; The information release trend classification module, combined with the output results of the machine learning model, divides the development trend of information release into "high-frequency release type" and "regular release type"; The synchronization cycle dynamic control module adjusts the synchronization cycle through an adaptive dynamic adjustment algorithm based on the output results of the machine learning model for the "high-frequency release" trend, shortens the synchronization interval to meet the needs of frequent updates; while for the "regular release" trend, it continues to maintain the predefined synchronization cycle to save resources.

Images