Exhibition and display interaction logic processing method and system and storage medium

Abstract

The invention relates to the technical field of data processing, in particular to an exhibition and display interaction logic processing method and system and a storage medium. Comprising the steps of collecting a user interaction request and a historical access log to form an initial execution path set; constructing a branch model, and identifying a high-frequency condition judgment sequence; if the sequence is not matched with the operation data, optimizing to obtain an adjusted condition judgment rule; extracting real-time features of the user interaction request from the operation data to generate feature vector representation, and traversing the adjusted rule to determine an optimal execution path; comparing the optimal path with the initial path set to obtain a path optimization index and generate a resource allocation adjustment scheme; when the response delay is reduced, the condition judgment rule is updated to form a final interaction processing mechanism; follow-up interaction requests are processed accordingly, and the execution efficiency and the dynamic adaptation capacity are evaluated. The problems that an existing exhibition and display system is fixed in interaction logic and difficult to adaptively optimize are solved, and dynamic optimization and efficient response of exhibition and display interaction logic are achieved.

Description

Exhibition interaction logic processing methods, systems and storage media Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method, system, and storage medium for processing interactive display logic. Background Technology

[0002] In today's digital exhibition environment, with the widespread application of multimedia display technologies, sensor-interactive devices, and big data analytics, exhibition systems are gradually shifting from traditional static content playback to a user-centric, dynamic interactive model. Visitors in exhibition halls engage in multimodal interactions via touchscreens, voice recognition, motion capture, or mobile devices. The system needs to generate corresponding response paths based on different users' operational characteristics, access intentions, and real-time status to achieve accurate content display and optimized user experience. However, existing exhibition interaction systems largely rely on preset fixed logic or manually defined conditional judgment rules. This static configuration approach has significant limitations when facing complex and ever-changing interactive scenarios.

[0003] Furthermore, with the diversification of exhibition content and the increasing complexity of interaction modes, relying solely on manual configuration or experience-based parameter adjustments is no longer sufficient to meet the intelligent requirements of the system. Especially under high-concurrency access or multi-scene switching, maintaining the accuracy of logical judgments and system stability while ensuring response speed has become a key technical challenge in the intelligent evolution of exhibition systems. Existing technologies have failed to establish a complete closed-loop mechanism from data collection, feature extraction, rule optimization to path updates, and also lack a systematic solution that combines algorithm self-learning and dynamic resource adjustment.

[0004] To address the aforementioned problems, this invention provides a method for processing exhibition interaction logic. By introducing a multi-layered processing mechanism, including feature vector extraction, conditional judgment rule optimization, decision tree traversal, and path simulation evaluation, it achieves dynamic optimization and adaptive updating of the exhibition system's interaction logic. This method can automatically identify conditional judgment deviations based on real-time operational data, generate optimal execution paths through algorithmic optimization, and combine resource allocation adjustments and dynamic monitoring feedback to form a continuously evolving interaction processing mechanism. This enables efficient, stable, and personalized user interaction experiences in complex and ever-changing exhibition environments. Summary of the Invention

[0005] This invention provides a method, system, and storage medium for processing exhibition interaction logic, which enables dynamic optimization and adaptive improvement of user interaction logic in an exhibition system through the synergy of multi-level data extraction, rule judgment, and path optimization mechanisms.

[0006] In a first aspect, the present invention provides a method for processing exhibition interaction logic, the method comprising: step S1: acquiring condition judgment rules and running data in the interaction logic by collecting user interaction requests and historical access logs in the exhibition system, and obtaining an initial execution path set; step S2: constructing a branch model using a decision tree algorithm based on the condition judgment rules in the initial execution path set, determining the trigger frequency of each condition judgment in the branch model, and identifying a high-frequency condition judgment sequence; step S3: if there is a part in the high-frequency condition judgment sequence that does not match the running data, optimizing the sequence using a genetic algorithm to obtain an adjusted condition judgment rule; step S4: processing the adjusted condition judgment rule... Then, extract real-time features of user interaction requests from the running data to obtain feature vector representations; based on the feature vector representations, use a decision tree algorithm to traverse the adjusted condition judgment rules, determine which rules are applicable to the current interaction request, and determine the preferred execution path; Step S5: by comparing the preferred execution path with the initial execution path set, obtain path optimization indicators to obtain a resource allocation adjustment scheme; if the resource allocation adjustment scheme shows a reduction in response latency, update the condition judgment rules through the rule editing interface to obtain the final interaction processing mechanism; Step S6: process subsequent user interaction requests according to the final interaction processing mechanism, determine whether the execution efficiency has improved, and determine the dynamic adaptation capability of path selection.

[0007] As a preferred technical solution of the present invention, step S1, obtaining the initial execution path set, includes: extracting condition judgment rules and running data from the interaction logic based on the collected user interaction requests and historical access logs, wherein the condition judgment rules include at least the logical expression for branch judgment and threshold conditions; the running data includes at least system load indicators and execution time statistics; generating the initial execution path set by performing correlation analysis on the extracted condition judgment rules and running data, wherein the initial execution path set includes at least multiple execution paths and their corresponding condition sequences; determining whether the initial execution path set covers all scenarios in the historical access logs, and if the initial execution path set does not cover all scenarios, supplementing the missing path data to obtain a complete initial execution path set.

[0008] As a preferred embodiment of the present invention, step S2, determining the high-frequency condition judgment sequence, includes: extracting condition judgment rules from the initial execution path set; modeling the extracted condition judgment rules using a decision tree algorithm to generate a branch model; traversing the path data in the initial execution path set through the branch model to calculate the trigger count of each condition judgment; determining the trigger frequency of each condition judgment based on the calculated trigger count, wherein the trigger frequency is obtained by the ratio of the trigger count to the total number of paths; selecting condition judgments with trigger frequencies higher than a preset threshold from the branch model to form a high-frequency condition judgment sequence; removing redundant judgment conditions from the high-frequency condition judgment sequence to obtain a simplified high-frequency condition judgment sequence; verifying its coverage in the initial execution path set based on the simplified high-frequency condition judgment sequence to determine the final high-frequency condition judgment sequence.

[0009] As a preferred technical solution of the present invention, step S3, obtaining the adjusted condition judgment rule includes: identifying the part that does not match the running data from the high-frequency condition judgment sequence, wherein the mismatched part includes at least the sequence segment where the condition judgment result deviates significantly from the actual running data; if the mismatched part is identified, the mismatched part is optimized by a genetic algorithm to obtain the winning sequence, and the mismatched part in the high-frequency condition judgment sequence is replaced by the optimized winning sequence to obtain the adjusted condition judgment rule.

[0010] In a preferred embodiment of the present invention, step S4, obtaining the feature vector representation, includes: extracting rule parameters from the adjusted condition judgment rules; filtering real-time data related to user interaction requests from the running data based on the extracted rule parameters; generating standardized features by normalizing the filtered real-time data; calculating vector components of user interaction requests using a feature extraction algorithm based on the standardized features, wherein the vector components include at least numerical features and categorical features; combining the calculated vector components into a feature vector representation; if there are missing components in the feature vector representation, supplementing relevant real-time features from the running data to obtain a complete feature vector representation.

[0011] As a preferred technical solution of the present invention, step S4, determining the preferred execution path, includes: obtaining input variables from the feature vector representation; constructing a traversal tree for the adjusted condition judgment rules using a decision tree algorithm; judging the conditions of each branch based on the input variables from the root node through the traversal tree to obtain matching branch paths; determining which adjusted condition judgment rules are applicable to the current interaction request based on the matching branch paths, wherein the applicable rules are determined by the endpoint of the branch path; selecting rule sequences with high execution efficiency from the applicable rules to form candidate execution paths; calculating the expected response time of each path by simulating the candidate execution paths; determining the preferred execution path based on the candidate execution path with the lowest expected response time; and confirming the preferred execution path if the preferred execution path completely matches the feature vector representation of the current interaction request.

[0012] In a preferred embodiment of the present invention, step S5, obtaining a resource allocation adjustment scheme, includes: extracting path parameters from the preferred execution path; calculating and obtaining a path optimization index based on the path difference degree by comparing the preferred execution path with the corresponding paths in the initial execution path set; if the path optimization index shows improvement, generating a resource allocation adjustment scheme based on the optimization index; simulating the application effect of the generated resource allocation adjustment scheme in the exhibition system; determining whether the resource allocation adjustment scheme reduces response latency based on the simulation results; if the response latency is reduced, using the adjustment scheme as the basis for updating; and optimizing the initial execution path set according to the adjustment scheme to obtain an optimized path set.

[0013] As a preferred embodiment of the present invention, step S6, determining the dynamic adaptability of path selection, includes: loading updated condition judgment rules from the final interaction processing mechanism; processing subsequent user interaction requests according to the loaded rules to generate an execution path; obtaining execution data by recording the response time and resource utilization rate of processing subsequent user interaction requests; determining whether the execution efficiency has improved based on a comparison of the execution data and historical data, wherein the improvement is determined by indicators of reduced response time and optimized resource utilization rate; if the execution efficiency has improved, analyzing the dynamic adaptability of path selection, wherein the dynamic adaptability is obtained by the matching accuracy of the path for different requests; based on the analysis results, if the dynamic adaptability is higher than a preset threshold, confirming the effectiveness of the final interaction processing mechanism; adjusting the evaluation model of dynamic adaptability by continuously monitoring the processing of subsequent user interaction requests; and determining the dynamic adaptability of path selection based on the adjusted evaluation model.

[0014] Secondly, the present invention also provides an exhibition interaction logic processing system for implementing the above-mentioned method. The system includes: a data acquisition unit, used to acquire condition judgment rules and running data in the interaction logic by collecting user interaction requests and historical access logs in the exhibition system, and obtain an initial execution path set; a model construction unit, used to construct a branch model using a decision tree algorithm based on the condition judgment rules in the initial execution path set, determine the trigger frequency of each condition judgment in the branch model, and determine a high-frequency condition judgment sequence; a sequence optimization unit, used to optimize the sequence using a genetic algorithm if there are parts in the high-frequency condition judgment sequence that do not match the running data, and obtain adjusted condition judgment rules; and a path determination unit, used to determine the path for adjusting the condition judgment rules. The adjusted conditional judgment rules extract real-time features of user interaction requests from the runtime data to obtain feature vector representations. Based on these feature vector representations, a decision tree algorithm is used to traverse the adjusted conditional judgment rules, determine which rules apply to the current interaction request, and identify the preferred execution path. The interaction mechanism acquisition unit compares the preferred execution path with the initial execution path set to obtain path optimization indicators and a resource allocation adjustment scheme. If the resource allocation adjustment scheme shows a reduction in response latency, the conditional judgment rules are updated through the rule editing interface to obtain the final interaction processing mechanism. The execution efficiency evaluation unit processes subsequent user interaction requests based on the final interaction processing mechanism, determines whether execution efficiency has improved, and identifies the dynamic adaptability of path selection.

[0015] Thirdly, the present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0016] The beneficial effects of this invention are as follows: Firstly, by extracting real-time features of user interaction requests from operational data and constructing a feature vector representation, this invention accurately reflects the real-time interaction status. Secondly, based on the feature vector, a decision tree algorithm is used to traverse and adjust the conditional judgment rules. At branch nodes, the optimal path is dynamically matched according to the input variables, thereby selecting rule sequences with high execution efficiency. Multiple simulations are performed on candidate paths to obtain statistical results of expected response times. The path with the lowest response time is selected as the preferred execution path, achieving rapid response of the interaction logic and optimization of path selection. Thirdly, by comparing the preferred execution path with the initial execution path set, the path difference is calculated, and path optimization indicators are generated. Efficiency improvement rate and resource consumption reduction rate are used as evaluation criteria. Through a resource allocation adjustment scheme, computing resources are dynamically adjusted. The allocation ratio of storage resources ensures that the system can maintain efficient operation under different loads. Subsequently, the adjustment scheme is applied to a virtual environment for simulation verification, and the effectiveness of the scheme is judged. After confirming the optimization effect, the execution path set is updated, thereby achieving continuous iteration at the mechanism level. The updated condition judgment rules are loaded from the interaction processing mechanism, the response time and resource utilization rate of subsequent interaction requests are recorded, and the data is compared with historical data to determine whether the execution efficiency has improved. Combined with a real-time monitoring mechanism and an adaptive evaluation model, the system can dynamically adjust the model parameters based on operational feedback and achieve self-evolution through the collaborative optimization of genetic algorithms and decision tree algorithms. Through the cooperation of the above technical solutions, the efficiency, robustness, and adaptability of the interaction logic processing are achieved, significantly improving the user interaction experience and system resource utilization. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 is a flowchart of the display interaction logic processing method in the embodiment; Figure 2 is a structural diagram of the display interaction logic processing system in the embodiment; Figures 3-5 are effect diagrams of the interaction processing mechanism in the embodiment. Detailed Implementation

[0019] This invention provides a method, system, and storage medium for display interaction logic processing. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0020] For ease of understanding, the specific process of this embodiment of the invention is described below. As shown in Figure 1, an exhibition interaction logic processing method in this embodiment of the invention includes: Step S1: By collecting user interaction requests and historical access logs in the exhibition system, condition judgment rules and running data in the interaction logic are obtained to obtain an initial execution path set; specifically, it includes: extracting condition judgment rules and running data in the interaction logic based on the collected user interaction requests and historical access logs, wherein the condition judgment rules at least include logical expressions for branch judgments and threshold conditions; the running data at least includes system load indicators and execution time statistics; generating an initial execution path set by performing correlation analysis on the extracted condition judgment rules and running data, wherein the initial execution path set at least includes multiple execution paths and their corresponding condition sequences; determining whether the initial execution path set covers all scenarios in the historical access logs, and if the initial execution path set does not cover all scenarios, supplementing the missing path data to obtain a complete initial execution path set.

[0021] Specifically, user interaction requests are collected from the exhibition system. These user interaction requests include at least user-input query commands and operation sequences, such as a user inputting a search command and performing multiple exhibition area navigation operations. Historical access logs are also collected from the exhibition system. These historical access logs include at least records of past user access paths and corresponding response time data. For example, user access path records from the past week can be queried through the log database, such as the jump sequence from the entry page to the detailed exhibit page, and the corresponding response time data, such as an average delay of 2.5 seconds for each jump. Combining user interaction requests and historical logs, conditional judgment rules in the interaction logic are extracted. These conditional judgment rules include logical expressions for branch judgments and threshold conditions. Specifically, executable logic is formed by parsing user query commands and operation sequences. The system uses logical expressions, such as "if operation sequence length > 3 then execution path A", and combines them with threshold conditions such as response time to form a judgment standard for actual interaction requirements. It also extracts runtime data based on the aforementioned interaction requests and historical logs. This runtime data includes information such as system load indicators and execution time statistics. For example, by statistically analyzing the system's load (such as CPU utilization) and the execution time of each path at different time periods, it is possible to understand the current resource usage and potential performance bottlenecks. The runtime data is then correlated with the conditional judgment rules to generate an initial set of execution paths. This set of execution paths contains multiple possible execution paths and their corresponding conditional sequences. Through correlation analysis, a simpler response path can be selected under high load conditions, ensuring that the system maintains high response speed and stability even under heavy pressure.

[0022] Next, it is determined whether the initial execution path set can cover all scenarios in the historical access logs. If the initial path set fails to cover certain scenarios, the missing path data is supplemented based on the missing scenarios in the historical logs, thus obtaining a complete initial execution path set. The supplementation process usually involves simulating the missing path scenarios to generate corresponding path condition sequences, ensuring the comprehensiveness and coverage of the path set. This allows the system to adapt to more interaction scenarios, thereby improving its ability to handle complex interaction requests. The above technical solution, by collecting and analyzing user interaction requests and historical access data in real time, provides a foundation for gradually optimizing interaction paths by combining system load and execution time data. By supplementing missing scenarios and identifying high-frequency paths, it provides an efficient and flexible interaction processing mechanism for the exhibition system.

[0023] Step S2: Based on the condition judgment rules in the initial execution path set, a branch model is constructed using a decision tree algorithm. The trigger frequency of each condition judgment in the branch model is determined to identify the high-frequency condition judgment sequence. Specifically, this includes: extracting condition judgment rules from the initial execution path set; modeling the extracted condition judgment rules using a decision tree algorithm to generate a branch model; traversing the path data in the initial execution path set through the branch model to calculate the trigger count of each condition judgment; determining the trigger frequency of each condition judgment based on the calculated trigger count, where the trigger frequency is obtained as the ratio of the trigger count to the total number of paths; selecting condition judgments with trigger frequencies higher than a preset threshold from the branch model to form a high-frequency condition judgment sequence; removing redundant judgment conditions from the high-frequency condition judgment sequence to obtain a simplified high-frequency condition judgment sequence; verifying the coverage of the simplified high-frequency condition judgment sequence in the initial execution path set to determine the final high-frequency condition judgment sequence.

[0024] Specifically, in this embodiment, the key step of the interactive logic processing method is to construct a branch model using a decision tree algorithm based on the condition judgment rules in the initial execution path set, and then determine the trigger frequency of condition judgments through this model, thereby identifying the high-frequency condition judgment sequence. Specifically, condition judgment rules are extracted from the initial execution path set. These rules include at least logical operators and variable dependencies, such as logical operators like AND and OR, and dependencies between user input variables and system state variables. To extract these rules from log data, the log data in the initial execution path set is parsed, the condition expressions of each interactive path are read, and the relevant logical operators and variable dependencies are extracted to form a complete rule list.

[0025] The extracted conditional judgment rules are modeled using a decision tree algorithm to generate a branch model. This branch model contains a tree structure, where each branch node corresponds to a conditional judgment and each leaf node corresponds to an execution result. The construction process of the decision tree is based on the conditional judgment rules. The decision tree recursively splits the data and selects the optimal split point. To this end, the decision tree uses indicators such as information gain or Gini index to calculate the merits of each split point, thereby deciding how to split the data at each node. Information gain is used to select the optimal split point by measuring the reduction of entropy after each split, ensuring that the generated tree structure can effectively capture the patterns of user interaction behavior and model complex interaction paths in the display system. Through the generated branch model, all path data in the initial execution path set are traversed, and the trigger count of each conditional judgment is calculated. Specifically, each path data is matched starting from the root node of the tree structure, and the number of times each branch node is visited is recorded. The number of visits is used to represent the trigger frequency of each conditional judgment, which is obtained by calculating the ratio of the trigger count of each conditional judgment to the total number of paths, thus providing a basis for subsequent path optimization.

[0026] After calculating the trigger frequency of each conditional judgment, conditional judgments with trigger frequencies higher than a preset frequency threshold are selected to form a high-frequency conditional judgment sequence. If redundant conditions appear in the high-frequency conditional judgment sequence, the sequence is further simplified by merging similar conditions. The redundancy detection process includes comparing the logical operators and variable dependencies in the conditional judgments. If two conditions have the same operator and their dependent variables have a high degree of overlap, these two conditions are considered redundant and are then merged into a single conditional expression. This not only reduces redundant conditions but also significantly reduces the complexity of subsequent optimization calculations and improves the system's response efficiency. Based on the simplified high-frequency conditional judgment sequence, its coverage in the initial execution path set is verified. If the coverage is high... If the simplified sequence can cover most paths, then the above sequence will serve as the final high-frequency condition judgment sequence to support subsequent path selection and execution, thereby ensuring the efficiency of path optimization and making the generated interactive paths more in line with actual user needs. The above technical solution models and optimizes the condition judgment rules through a decision tree algorithm, analyzes the trigger frequency of path data using the branch model generated by the algorithm, filters high-frequency condition judgment sequences, and reduces the complexity of condition sequences through redundancy merging, thereby achieving efficient interactive logic processing in the exhibition system. This not only effectively improves the execution efficiency of interactive paths but also allows for dynamic optimization according to different scenarios, ensuring that the system maintains flexibility and efficiency in a changing user interaction environment.

[0027] Step S3: If there is a part in the high-frequency condition judgment sequence that does not match the running data, then the sequence is optimized by a genetic algorithm to obtain the adjusted condition judgment rule; specifically, this includes: identifying the part in the high-frequency condition judgment sequence that does not match the running data, wherein the mismatched part includes at least the sequence segment where the condition judgment result deviates significantly from the actual running data; if a mismatched part is identified, the mismatched part is optimized by a genetic algorithm to obtain the winning sequence, and the mismatched part in the high-frequency condition judgment sequence is replaced by the optimized winning sequence to obtain the adjusted condition judgment rule.

[0028] Specifically, in this embodiment, since user behavior in actual interactive systems often does not completely follow the patterns predicted by historical data, but is influenced by multiple dynamic factors, such as system load, number of concurrent users, and hardware status, by comparing the high-frequency condition judgment sequence with the actual running data, it can be found whether the system accurately reflects the user's real needs, avoiding performance bottlenecks caused by overly idealistic assumptions. When there are parts in the high-frequency condition judgment sequence that do not match the running data, for example, if the high-frequency condition judgment rule assumes that a certain condition will trigger a fast path, but the execution time in the actual running data is much higher than expected, this deviation is a typical manifestation of mismatch. Identifying this deviation helps to improve and optimize the condition judgment rule, making it more in line with the actual scenario. By aggregating multiple consecutive large deviation sequence segments into a mismatched part, the core part of the problem in the entire interaction path can be identified more effectively, helping the system to better locate which judgment logics need optimization, rather than just local adjustments, thereby improving the overall interaction efficiency. The sequence is optimized using a genetic algorithm to obtain the adjusted condition judgment rule. Specifically, firstly... The system identifies segments in high-frequency conditional judgment sequences that do not match the running data. These mismatched segments include at least those with significant deviations between the conditional judgment results and the actual running data. To address this, each conditional judgment result is compared with its corresponding running data, the deviation value is calculated, and mismatched segments are marked according to a preset threshold. This process accurately identifies rule segments in the interaction logic that do not conform to actual user behavior, providing a foundation for subsequent optimization and improving system response efficiency. After identifying the mismatched segments, the genetic algorithm population is initialized. This population includes multiple mutated versions of candidate conditional judgment sequences, randomly generated based on the initial high-frequency conditional judgment sequences. Through crossover operations, segments from these candidate sequences are combined to generate new sequences. This crossover operation, by exchanging segments from different sequences, helps explore better rule combinations to handle complex interaction scenarios. Furthermore, the genetic algorithm randomly adjusts the newly generated sequences through mutation operations, such as modifying thresholds or logical operators in certain conditional judgments, to ensure population diversity and avoid getting trapped in local optima.

[0029] The fitness of the mutated sequences is evaluated based on the operational data. Fitness is measured by calculating the matching degree between the mutated sequences and the operational data. Specifically, the mutated sequences are applied to historical operational data to simulate execution results, and the matching degree is calculated based on the overlap between the simulation results and the actual operational data. For example, the matching degree is calculated using the overlap rate formula, where the overlap rate refers to the ratio of the number of correct judgments in the simulation to the total number of judgments. A higher matching degree indicates that the mutated sequences can better adapt to the actual scenario, thereby improving the system's execution efficiency. After evaluating the fitness, winning sequences are selected from the mutated sequences with high fitness, and these winning sequences are further improved through an iterative optimization process until convergence is reached. Specifically, the sequences in the population are sorted, and sequences with high fitness are selected as parents. New offspring populations are generated through crossover and mutation operations. The above process is repeated until the improvement in fitness is less than a preset threshold, indicating that the algorithm has converged and the generated winning sequences are close to the optimal solution.

[0030] After optimization using a genetic algorithm, the winning sequence is used to replace the parts of the original high-frequency condition judgment sequence that do not match the running data, thus obtaining the adjusted condition judgment rules. It is then determined whether the adjusted condition judgment rules improve the matching degree with the running data. If the adjusted rules improve the matching degree, these rules are confirmed as the basis for subsequent processing and applied to the actual interaction logic, thereby further optimizing the system's interaction efficiency. This technical solution not only adjusts mismatched high-frequency condition judgment sequences but also improves the adaptability of rules and the system's response efficiency through continuous iterative optimization. By accurately identifying mismatched parts and performing targeted optimization, it can achieve efficient path selection in dynamic interaction scenarios, thereby improving user experience and system processing capabilities.

[0031] Step S4: For the adjusted conditional judgment rules, extract real-time features of user interaction requests from the running data to obtain feature vector representations; based on the feature vector representations, use a decision tree algorithm to traverse the adjusted conditional judgment rules, determine which rules are applicable to the current interaction request, and determine the preferred execution path; wherein, in step S4, obtaining the feature vector representation includes: extracting rule parameters from the adjusted conditional judgment rules; for the extracted rule parameters, filtering real-time data related to user interaction requests from the running data; generating standardized features by normalizing the filtered real-time data; calculating vector components of user interaction requests using a feature extraction algorithm based on the standardized features, wherein the vector components include at least numerical features and categorical features; combining the calculated vector components into a feature vector representation; if there are missing components in the feature vector representation, supplementing relevant real-time features from the running data to obtain a complete feature vector representation.

[0032] Specifically, to improve the dynamic adaptability and response efficiency of the exhibition system in complex interactive scenarios, rule parameters are extracted from the adjusted conditional judgment rules. These rule parameters include at least condition variables and judgment thresholds. The rule parameters embody the logical basis of the exhibition system's interaction judgments. For example, condition variables describe interactive behavior characteristics such as click frequency, dwell time, and access path, while judgment thresholds define the boundary standards for the corresponding conditions to be met. A parameter list can be formed by parsing the rule structure. This list fully describes the variable space of the rule judgment logic, providing a basis for subsequent data matching and filtering. Based on the extracted parameter list, real-time data related to the current user interaction request is filtered from the runtime data. This real-time data includes current... The input parameters and system status indicators of the previous request are used to characterize the real-time state of user interaction behavior, such as interaction type, input content, and request path. The system status indicators reflect the system's operational characteristics such as load, response latency, and cache status. Through the above filtering process, it can be ensured that the extracted data is consistent with the logical context of the current interaction request. In order to avoid the impact of the difference in the scale between different types of data on subsequent analysis, the real-time data obtained by filtering is normalized to generate standardized features, which maps data of different numerical ranges to a unified scale, so that the features of each dimension are comparable in subsequent calculations. The processed standardized features form a data matrix in which numerical features and categorical features coexist in a structured manner.

[0033] Based on standardized features, a feature extraction algorithm is used to calculate vector components of user interaction requests. These vector components include at least numerical and categorical features. Specifically, the standardized feature set is first subjected to dimensionality reduction and feature compression to reduce redundant information and enhance the expressive power of the main feature components. For example, principal component analysis can be used to perform matrix transformations on numerical features, thereby extracting the top principal components with high feature variance contribution rates to represent the main trends in the original high-dimensional data. For categorical features, feature encoding is performed using embedding vectors to establish a correspondence between them and numerical features in a continuous vector space. Numerical features reflect the quantitative trends of user behavior, while categorical features reflect the semantic differences in interaction methods. Together, they constitute a multi-dimensional vector component set describing the user interaction request. After obtaining the vector group... After processing, these components are combined to form a feature vector representation. This feature vector is expressed as a multi-dimensional array, with each dimension corresponding to a feature type. The overall structure reflects the comprehensive feature distribution of the user interaction request in the current running state. If a missing component is detected in the feature vector during the combination process, the relevant real-time features are automatically supplemented by accessing the historical logs of the running data, thereby obtaining a complete feature vector representation. For numerical missing features, they can be filled based on the historical average of the corresponding parameters. For categorical missing features, the category with the highest frequency is selected as the substitute. This supplementation process not only maintains the consistency of feature dimensions but also ensures the integrity and timeliness of the feature representation through a real-time update mechanism. When the deviation of the supplemented data exceeds a preset threshold, the data source is queried again for correction, thereby preventing abnormal data from affecting subsequent judgments.

[0034] After generating the complete feature vector representation, the compatibility between the feature vector and the adjusted conditional judgment rules is further verified. The compatibility verification involves traversing the rule branches, matching each dimension of the feature vector with the rule parameters, and evaluating whether it meets the judgment threshold of the corresponding conditional variable. If all components of the feature vector are within the threshold range, the feature vector is considered compatible with the rule and can be used for subsequent high-frequency conditional judgment sequence traversal and path optimization. If some components are incompatible, the source of the deviation is recorded and returned to the supplementary stage for data correction. This closed-loop verification mechanism ensures the consistency between the input features and the rule logic, thereby improving the accuracy and stability of the traversal and optimization stages. This technical solution not only enhances the expressive power of the interaction rules but also improves the robustness of data processing and the accuracy of execution path selection through algorithm fusion, providing a solid technical foundation for the efficient response and intelligent optimization of the exhibition system.

[0035] Further, in step S4, determining the preferred execution path includes: obtaining input variables from the feature vector representation; constructing a traversal tree for the adjusted condition judgment rules using a decision tree algorithm; judging the conditions of each branch based on the input variables starting from the root node through the traversal tree to obtain matching branch paths; determining which adjusted condition judgment rules are applicable to the current interaction request based on the matching branch paths, wherein the applicable rules are determined by the endpoint of the branch path; selecting rule sequences with high execution efficiency from the applicable rules to form candidate execution paths; calculating the expected response time of each path by simulating the candidate execution paths; determining the preferred execution path based on the candidate execution path with the lowest expected response time; and confirming the preferred execution path if the preferred execution path completely matches the feature vector representation of the current interaction request.

[0036] Specifically, to achieve intelligent response and dynamic path optimization for user interaction requests in the exhibition system, based on the generated feature vector representation, a decision tree algorithm is used to traverse the adjusted condition judgment rules to determine which rules are applicable to the current interaction request and thus determine the preferred execution path, thereby achieving adaptive logic processing for different interaction scenarios. Specifically, input variables are first obtained from the feature vector representation. These input variables include at least numerical and categorical features extracted from real-time features. Numerical features are used to characterize the quantitative attributes of user interaction behavior, such as request frequency, dwell time, and system response time, while categorical features are used to express the semantic information of the interaction request, such as interaction type, exhibit category, or access terminal type. By parsing the feature vector, it is decomposed into multiple structured variables, so that each type of feature can be mapped to the input dimension of subsequent rule judgments, ensuring a one-to-one correspondence between data and rule conditions. A decision tree structure is constructed based on the adjusted condition judgment rules. These condition judgment rules are high-frequency condition judgment sequences optimized by a genetic algorithm, thus possessing representativeness and stability for user interaction logic. To achieve efficient rule traversal, a decision tree algorithm is used to construct the traversal tree, where... Each condition judgment serves as the basis for splitting nodes. By calculating splitting indices such as Gini impurity or information gain for condition nodes, the optimal splitting method is determined, ensuring that each branch corresponds to a judgment result. This forms a logical path from the root node to the leaf node in the tree structure. The construction logic of the traversal tree is based on condition variables as the core of judgment, unfolding complex interaction logic layer by layer to achieve branch coverage for different interaction modes. To avoid overfitting the model to historical data, a limit parameter is set on the tree depth during construction, thus ensuring judgment accuracy while taking into account generalization performance, so that the generated traversal tree can better adapt to the diverse interaction scenarios in the exhibition system. During the traversal, starting from the root node, the conditions of each branch are judged layer by layer based on the input variables. The relationship between the value of the input variable and the condition threshold is compared, and the branch path that meets the condition is selected to continue recursively downward until the leaf node is reached. The selection of each branch in the above process reflects the matching relationship between the specific value in the feature vector and the rule condition, forming a data-driven logical reasoning chain. After reaching the leaf node, the set of rule branches corresponding to the current input variable can be determined, and the condition judgment rule associated with the leaf node is identified as the applicable rule for the current interaction request.

[0037] For the identified applicable rules, a further selection process is used to filter out rule sequences with high execution efficiency to form candidate execution paths. During this selection process, the execution efficiency metrics for each rule, such as average processing time or response stability, are first calculated based on historical running data, and all applicable rules are then ranked accordingly. Subsequently, several highly efficient rules are selected based on the ranking results to form a sequence, creating multiple candidate execution paths. Since different rules may have logical dependencies or conflicts, compatibility verification is performed simultaneously during the selection phase. Consistency checks are conducted on the condition variables that depend on each other in the rule sequences to ensure that the generated candidate execution paths are logically continuous and consistent. After obtaining the candidate execution paths, they are simulated to calculate the execution efficiency of each path. To improve the accuracy of the evaluation, a Monte Carlo simulation method is introduced. This method involves randomly sampling the input variables multiple times, repeatedly running the rule sequence under different initial conditions, recording the time interval from request input to response output, and averaging the results of multiple simulations to obtain the expected response time for each path. This statistical simulation method can reflect the comprehensive performance of the path under multidimensional variable perturbations, thereby avoiding bias in a single scenario. If system parameters, such as network bandwidth and resource utilization, change during the simulation, the simulation parameters can be dynamically adjusted to maintain the authenticity of the results. By comparing the expected response times of all candidate paths, the path with the lowest response time is identified and determined as the preferred execution path.

[0038] Furthermore, the system performs matching verification between the preferred execution path and the feature vector representation of the current interaction request. It compares the corresponding values ​​of the condition variables of each rule in the path with the corresponding components of the feature vector to determine whether all judgment thresholds are fully met. If all match, the preferred execution path is confirmed as the final applicable path and used as input to the resource allocation module for subsequent dynamic adjustment of computing, storage, and bandwidth resources. This ensures the consistency between the path selection result and the actual interaction characteristics, enabling the system to achieve optimal resource scheduling at the execution level based on specific interaction requests. The above technical solution forms a dynamic mapping relationship between the running data and the judgment rules. Different types of input variables correspond to different decision paths in the traversal tree after processing, thereby realizing intelligent judgment of the interaction logic of the exhibition system and adaptive determination of the preferred path. This not only improves the decision accuracy and response speed of the system under complex interaction conditions, but also enhances the stability of path selection and resource utilization efficiency through simulation optimization mechanism. Overall, it realizes efficient and intelligent interaction logic processing of the exhibition system.

[0039] Step S5: By comparing the preferred execution path with the initial execution path set, obtain path optimization indicators to get a resource allocation adjustment plan; if the resource allocation adjustment plan shows a reduction in response latency, update the condition judgment rules through the rule editing interface to obtain the final interactive processing mechanism; specifically, this includes: extracting path parameters from the preferred execution path, calculating and obtaining path optimization indicators based on the path difference degree by comparing the preferred execution path with the corresponding paths in the initial execution path set; if the path optimization indicators show improvement, generate a resource allocation adjustment plan based on the optimization indicators, and simulate its application effect in the exhibition system for the generated resource allocation adjustment plan; determine whether the resource allocation adjustment plan reduces response latency based on the simulation results, and if the response latency is reduced, use the adjustment plan as the basis for updating; optimize the initial execution path set according to the adjustment plan to obtain an optimized path set.

[0040] Specifically, in order to achieve dynamic optimization and adaptive resource scheduling of the exhibition system in a complex interactive environment, the path optimization index is obtained by comparing the optimized execution path with the initial execution path set, and a resource allocation adjustment scheme is generated accordingly, thereby achieving a comprehensive improvement in system execution efficiency and resource utilization. The above process takes path difference analysis as the core, and through quantitative node change and sequence adjustment measurement, the path optimization effect is accurately evaluated, and the optimization results are further mapped to the reallocation of computing and storage resources to form a closed-loop optimization mechanism.

[0041] In the implementation process, path parameters are first extracted from the preferred execution path. These path parameters include at least a condition sequence and execution nodes. The condition sequence refers to the logical branch order in the interaction logic, and the execution nodes correspond to specific processing modules. The preferred execution path, as the rule sequence with the best performance after simulation verification, contains condition judgment points and corresponding execution action nodes. By structurally decomposing the path, the logical judgment process is restored to an ordered arrangement of condition sequences, and the functional modules corresponding to each execution node are extracted, such as data analysis modules, display control modules, or cache update modules, thereby establishing a mapping relationship between logical conditions and physical execution nodes. After obtaining the path parameters, the preferred execution path is matched and compared with the corresponding paths in the initial execution path set to calculate the path difference. The calculation of the difference adopts a quantitative combination of node changes and sequence adjustments. Specifically, the two paths are first aligned at the node level to identify the same execution nodes and condition sequence parts. Based on the matching results, the node changes are calculated, i.e., the number of new nodes and the number of execution nodes are counted. The number of execution nodes deleted or modified is assigned a quantitative score to represent the adjustment range of the node structure. The difference in the execution order of the condition sequence is measured by using the edit distance algorithm to calculate the sequence adjustment distance. The edit distance algorithm reflects the degree of difference between two sequences by minimizing the number of insertion, deletion and replacement operations, thus accurately reflecting the impact of rule rearrangement on the execution logic. The sequence adjustment influence score is obtained based on the sequence adjustment distance. The node change score and the sequence adjustment influence score are both proportional to the corresponding adjustment range. Finally, the node change score and the corresponding sequence adjustment influence score are weighted and summed to obtain the path difference index, which reflects the overall change of the path. The higher the difference, the greater the optimization range of the preferred path in terms of structure compared with the original path. It usually corresponds to a higher execution efficiency potential. For example, when processing user touch interaction requests in an exhibition system, the initial path may include the sequence condition ABC, while the preferred path is adjusted to ACB. The node change is 0, but the sequence adjustment distance is 2, and the difference is 2.

[0042] After obtaining the path difference degree, it is mapped to path optimization metrics to evaluate the optimization effectiveness. These metrics include at least the efficiency improvement rate and the resource consumption reduction rate. The efficiency improvement rate is calculated by establishing a proportional relationship between the path difference degree and the initial path length, reflecting the degree of improvement in system response speed or task processing time after optimization. The resource consumption reduction rate is calculated as a function of the difference degree and the resource utilization ratio, reflecting the improvement in system resource usage during the optimization process. The combined optimization metric system can comprehensively evaluate the contribution of path adjustment to both execution performance and resource utilization. When the path optimization metrics show improvement, a resource allocation adjustment plan is generated based on these metrics. This plan includes at least the reallocation of computing and storage resources. The generation process determines the adjustment ratio based on the ratio of efficiency improvement rate to resource reduction rate. Specifically, it first identifies high-load nodes that undertake the main logical judgment and data processing in the preferred path, and transfers some computing resources from inefficient nodes in the initial path to these efficient nodes to improve processing parallelism and responsiveness. At the same time, it optimizes the storage structure according to the resource reduction rate, reduces the storage occupation of redundant data or low-frequency call modules in the initial path, and reallocates the freed storage space to relevant nodes in the preferred path for caching feature vectors or intermediate results, thereby reducing the latency caused by repeated calculations. For multi-user concurrent scenarios, a priority queue mechanism is further introduced to allocate higher resource scheduling priority to high-frequency paths according to optimization indicators, thereby maintaining the stability of overall performance under high load conditions.

[0043] After generating the resource allocation adjustment scheme, it is simulated in a virtual environment to verify its actual effect. During the simulation, the input is the same interactive request sequence as the real operation, and the processing time and resource usage are recorded. By comparing the changes in system response latency before and after the simulation, it is determined whether the adjustment scheme has achieved the optimization goal. If the simulation results show a significant reduction in response latency, the adjustment scheme is confirmed to be effective and is used as the basis for subsequent updates in the system resource management module. The above verification process ensures the feasibility and effectiveness of resource adjustment and avoids system load fluctuations caused by blind optimization. Based on the verified resource allocation adjustment scheme, the initial execution path set is optimized to form an optimized path set for subsequent mechanism updates. The optimization process applies the resource weights in the adjustment scheme to the initial path set, updates the resource configuration values ​​of each execution node in each path, and recalculates the path weights and priority ranking. Inefficient paths are eliminated in the re-ranking, while high-frequency and efficient paths are given priority and resource allocation. The optimized path set not only remains logically compatible with the original system, but also achieves adaptive adjustment of resource allocation at the execution level, thus providing an updated optimization baseline for subsequent interactive processing. During continuous operation, this optimization set will be continuously updated with new interactive data and simulation results, forming the dynamic evolution capability of the exhibition system's interactive mechanism.

[0044] The above technical solution establishes a closed-loop feedback system from path optimization to resource reconstruction through processing steps such as path parameter extraction, difference quantification, optimization index calculation, resource adjustment generation, simulation verification, and path set update. This not only realizes the intelligent evolution of the interactive logic of the exhibition system, but also establishes a dynamic balance between resource utilization and response efficiency, thereby significantly improving the overall interactive performance and adaptability of the system in diverse exhibition scenarios.

[0045] Step S6: Process subsequent user interaction requests according to the final interaction processing mechanism, determine whether the execution efficiency has improved, and determine the dynamic adaptation capability of path selection; specifically, this includes: loading updated condition judgment rules from the final interaction processing mechanism, processing subsequent user interaction requests according to the loaded rules, and generating execution paths; obtaining execution data by recording the response time and resource utilization rate of processing subsequent user interaction requests; determining whether the execution efficiency has improved by comparing the execution data with historical data, wherein the improvement is determined by indicators of reduced response time and optimized resource utilization rate; if the execution efficiency has improved, analyzing the dynamic adaptation capability of path selection, wherein the dynamic adaptation capability is obtained by the matching accuracy of the path for different requests; based on the analysis results, if the dynamic adaptation capability is higher than a preset threshold, confirming the effectiveness of the final interaction processing mechanism; adjusting the evaluation model of dynamic adaptation capability by continuously monitoring the processing of subsequent user interaction requests; and determining the dynamic adaptation capability of path selection based on the adjusted evaluation model.

[0046] Specifically, in order to enable the system to learn and adapt to user interaction requests in complex exhibition scenarios, subsequent user interaction requests are processed based on the final interaction processing mechanism. Through the evaluation of execution efficiency and the analysis of the dynamic adaptation capability of path selection, a closed-loop feedback process from operation monitoring to mechanism optimization is constructed to ensure the continuous evolution and stability of the system.

[0047] In the specific implementation process, the updated condition judgment rules are first loaded from the final interaction processing mechanism. These updated rules include at least sequences and parameters optimized by a genetic algorithm. The optimized rule sequence is stored in the mechanism's internal rule base. Once loaded into the system's runtime module, it can be directly used for logical judgment and path decision-making in subsequent user interaction requests, ensuring optimal adaptability of the rules in the latest runtime data environment. After the rules are loaded, subsequent user interaction requests are processed according to the updated rules. The user interaction request is first parsed into a feature vector representation, which serves as the input variable for the decision tree algorithm, thereby generating the execution path that best matches the current interaction request. During the path execution phase, the response time and resource utilization of subsequent user interaction requests are recorded in real time to generate an execution data set. This execution data includes time-series information from request reception to path execution completion. The system monitors the system to collect data on CPU, memory, bandwidth, and other resource usage metrics. This data is automatically collected and logged to form a traceable execution dataset, providing a foundation for subsequent efficiency analysis. The execution data is then compared with historical access logs to determine if execution efficiency has improved. The comparison quantifies performance improvement by reducing response time and optimizing resource utilization. Specifically, the average response time for the current stage is calculated and compared with the average response time for historical stages. If the current value is significantly lower than the historical value, the response speed is considered improved. Simultaneously, the trend of resource utilization is evaluated, and the degree of resource optimization is determined by calculating the decrease in CPU and memory usage ratios. A comprehensive judgment on the improvement in execution efficiency is formed by combining the results of these two types of metrics, which measures the overall performance improvement effect of the optimized final interaction processing mechanism.

[0048] After confirming the improved execution efficiency, the dynamic adaptability of path selection was further analyzed. Dynamic adaptability reflects the stability and flexibility of the system's path matching accuracy under different types of user requests. Its calculation process is achieved by measuring the matching accuracy of paths for multiple request types. Specifically, the actual optimal paths for different request types, such as query, update, and display requests, are compared with the execution paths generated by the mechanism. The proportion of matching nodes is statistically analyzed to form a path matching accuracy index. To enhance the focus of the analysis, the aforementioned decision tree algorithm is used to further refine the evaluation of high-frequency request categories. Splitting features are calculated using information gain to verify the real-time response capability of the path to mutated requests. For complex requests containing multi-condition judgment logic, sequences optimized by a genetic algorithm are introduced for evaluation. By comparing the changes in the matching rate of sequences before and after optimization, the adaptability improvement of the optimization algorithm in dynamic scenarios is identified. This not only reveals the accuracy of path selection but also verifies the effectiveness of the optimization sequence. The robustness of the model in complex interactive environments is assessed. If the analysis results show that the dynamic adaptation capability is higher than the preset threshold, the effectiveness of the final interaction processing mechanism is confirmed. This determination is achieved through threshold comparison. When the matching accuracy exceeds the preset threshold, the current mechanism is recorded as an effective version and the current optimization loop ends, entering the stable operation phase. In order to maintain long-term adaptive capability, the processing of subsequent user interaction requests is continuously monitored, and the evaluation of adaptation capability is dynamically adjusted. The monitoring module collects new execution data in real time and updates the model parameters according to the fluctuation of response time and the change of accuracy. When a downward trend in matching accuracy is detected, the mutation mechanism of the genetic algorithm is automatically introduced to optimize the evaluation model, so that the model structure continues to evolve to adapt to new interaction behavior patterns. The training data of the above evaluation model includes historical running data, feature vector representations extracted from the corresponding user's historical interaction requests, historical path execution data, and corresponding dynamic adaptation capability indicators.

[0049] During model adjustment, parameters are retrained based on feedback from the running data. For example, if the log shows that a specific high-frequency condition sequence no longer matches the running data, the mismatch is input into the iteration of the genetic algorithm to update the fitness function, thereby readjusting the model's weight allocation so that it can more accurately predict path matching results in subsequent judgments. The above feedback mechanism enables the evaluation model to maintain high adaptability and prediction accuracy when facing variable interaction logic.

[0050] Finally, the subsequently collected operational data, feature vector representations, and path execution data are input into the adjusted evaluation model to determine the dynamic adaptability of path selection. This capability value is then used as feedback input to the resource allocation adjustment module to update path optimization indicators and conditional judgment rule parameters. This feedback input forms a closed-loop logical chain of data, rules, paths, and performance within the system: new interactive data drives the evaluation model update, and the model's output adaptability indicators, in turn, influence rule optimization and resource reallocation, thereby continuously improving the system's execution efficiency and stability. In practical application scenarios of exhibition systems, this closed-loop mechanism ensures that the system self-evolves during long-term operation, achieving continuous reduction in response time, continuous optimization of resource consumption, and dynamic adaptation of interactive paths, thus significantly improving the intelligence level and user interaction experience of the exhibition system.

[0051] For example, the initial execution path set showed a mismatch between the high-frequency condition judgment sequence and the running data. After optimization using a genetic algorithm, subsequent requests were processed by loading rules, as shown in Figures 3-5. The response time decreased from 2 seconds to 1.5 seconds, and the resource utilization rate was optimized from 75% to 65%. After confirming the efficiency improvement, the dynamic adaptation capability analysis showed a matching accuracy rate of 88%, which was higher than the threshold of 80%, confirming the effectiveness. Under continuous monitoring, the evaluation model parameters were adjusted, raising the accuracy rate to 93%. This was used as feedback input to optimize the resource allocation scheme, thus demonstrating the role of the dynamic adaptation capability of path selection in the optimization of interaction logic.

[0052] This invention also provides an exhibition interaction logic processing system for implementing the above-mentioned method. As shown in Figure 2, the system includes: a data acquisition unit, used to acquire user interaction requests and historical access logs in the exhibition system, obtain condition judgment rules and running data in the interaction logic, and obtain an initial execution path set; a model construction unit, used to construct a branch model using a decision tree algorithm based on the condition judgment rules in the initial execution path set, determine the trigger frequency of each condition judgment in the branch model, and determine the high-frequency condition judgment sequence; a sequence optimization unit, used to optimize the sequence using a genetic algorithm if there are parts in the high-frequency condition judgment sequence that do not match the running data, to obtain adjusted condition judgment rules; and a path determination unit, used to determine the path for adjusting the condition judgment rules. The adjusted conditional judgment rules extract real-time features of user interaction requests from the runtime data to obtain feature vector representations. Based on these feature vector representations, a decision tree algorithm is used to traverse the adjusted conditional judgment rules, determine which rules apply to the current interaction request, and identify the preferred execution path. The interaction mechanism acquisition unit compares the preferred execution path with the initial execution path set to obtain path optimization indicators and a resource allocation adjustment scheme. If the resource allocation adjustment scheme shows a reduction in response latency, the conditional judgment rules are updated through the rule editing interface to obtain the final interaction processing mechanism. The execution efficiency evaluation unit processes subsequent user interaction requests based on the final interaction processing mechanism, determines whether execution efficiency has improved, and identifies the dynamic adaptability of path selection.

[0053] The present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0054] In summary, this invention extracts real-time features of user interaction requests from runtime data to construct feature vector representations, accurately reflecting the real-time interaction status. Secondly, based on the feature vectors, a decision tree algorithm is used to traverse and adjust the conditional judgment rules. At branch nodes, the optimal path is dynamically matched according to the input variables, thereby selecting rule sequences with high execution efficiency. Multiple simulations are performed on candidate paths to obtain statistical results of expected response times. The path with the lowest response time is selected as the preferred execution path, achieving rapid response of the interaction logic and optimization of path selection. By comparing the preferred execution path with the initial set of execution paths, the path difference is calculated, and path optimization indicators are generated. Efficiency improvement rate and resource consumption reduction rate are used as evaluation criteria. Through a resource allocation adjustment scheme, computing and storage resources are dynamically adjusted. The allocation ratio is optimized to ensure the system maintains high efficiency under different loads. The adjusted scheme is then applied to a virtual environment for simulation verification, and its effectiveness is assessed. After confirming the optimization effect, the execution path set is updated, thus achieving continuous iteration at the mechanism level. Updated conditional judgment rules are loaded from the interaction processing mechanism, and the response time and resource utilization of subsequent interaction requests are recorded. These are compared with historical data to determine if execution efficiency has improved. Combined with a real-time monitoring mechanism and an adaptive evaluation model, the system can dynamically adjust model parameters based on operational feedback and achieve self-evolution through the collaborative optimization of genetic algorithms and decision tree algorithms. Through the synergy of these technical solutions, high efficiency, robustness, and adaptability in interaction logic processing are achieved, significantly improving user interaction experience and system resource utilization.

[0055] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0056] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0057] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing interactive logic in exhibitions, characterized in that, The method includes: Step S1: By collecting user interaction requests and historical access logs from the exhibition system, obtaining the condition judgment rules and running data in the interaction logic, and obtaining an initial execution path set; Step S2: Based on the condition judgment rules in the initial execution path set, constructing a branch model using a decision tree algorithm, determining the trigger frequency of each condition judgment in the branch model, and identifying a high-frequency condition judgment sequence; Step S3: If there are parts in the high-frequency condition judgment sequence that do not match the running data, optimizing the sequence using a genetic algorithm to obtain adjusted condition judgment rules; Step S4: For the adjusted condition judgment rules, extracting user... The real-time characteristics of the interaction request are used to obtain a feature vector representation. Based on the feature vector representation, a decision tree algorithm is used to traverse the adjusted condition judgment rules to determine which rules are applicable to the current interaction request and to determine the preferred execution path. Step S5: By comparing the preferred execution path with the initial execution path set, path optimization indicators are obtained to obtain a resource allocation adjustment scheme. If the resource allocation adjustment scheme shows a reduction in response latency, the condition judgment rules are updated through the rule editing interface to obtain the final interaction processing mechanism. Step S6: Subsequent user interaction requests are processed according to the final interaction processing mechanism to determine whether the execution efficiency has improved and to determine the dynamic adaptation capability of path selection.

2. The method as described in claim 1, characterized in that, In step S1, obtaining the initial execution path set includes: extracting condition judgment rules and runtime data from the interaction logic based on the collected user interaction requests and historical access logs. The condition judgment rules include at least logical expressions for branch judgments and threshold conditions; the runtime data includes at least system load indicators and execution time statistics; generating the initial execution path set by performing correlation analysis on the extracted condition judgment rules and runtime data. The initial execution path set includes at least multiple execution paths and their corresponding condition sequences; determining whether the initial execution path set covers all scenarios in the historical access logs. If the initial execution path set does not cover all scenarios, the missing path data is supplemented to obtain a complete initial execution path set.

3. The method as described in claim 1, characterized in that, In step S2, determining the high-frequency condition judgment sequence includes: extracting condition judgment rules from the initial execution path set; modeling the extracted condition judgment rules using a decision tree algorithm to generate a branch model; traversing the path data in the initial execution path set through the branch model to calculate the trigger count of each condition judgment; determining the trigger frequency of each condition judgment based on the calculated trigger count, wherein the trigger frequency is obtained by the ratio of the trigger count to the total number of paths; selecting condition judgments with trigger frequencies higher than a preset threshold from the branch model to form a high-frequency condition judgment sequence; removing redundant judgment conditions from the high-frequency condition judgment sequence to obtain a simplified high-frequency condition judgment sequence; verifying its coverage in the initial execution path set based on the simplified high-frequency condition judgment sequence to determine the final high-frequency condition judgment sequence.

4. The method as described in claim 1, characterized in that, In step S3, obtaining the adjusted condition judgment rule includes: identifying the part that does not match the running data from the high-frequency condition judgment sequence, wherein the mismatched part includes at least the sequence segment where the condition judgment result deviates significantly from the actual running data; if the mismatched part is identified, the mismatched part is optimized by a genetic algorithm to obtain the winning sequence, and the mismatched part in the high-frequency condition judgment sequence is replaced by the optimized winning sequence to obtain the adjusted condition judgment rule.

5. The method as described in claim 1, characterized in that, In step S4, the feature vector representation is obtained, including: extracting rule parameters from the adjusted conditional judgment rules; filtering real-time data related to user interaction requests from the running data based on the extracted rule parameters; generating standardized features by normalizing the filtered real-time data; calculating vector components of user interaction requests using a feature extraction algorithm based on the standardized features, wherein the vector components include at least numerical features and categorical features; combining the calculated vector components into a feature vector representation; if there are missing components in the feature vector representation, supplementing relevant real-time features from the running data to obtain a complete feature vector representation.

6. The method as described in claim 5, characterized in that, In step S4, determining the preferred execution path includes: obtaining input variables from the feature vector representation; constructing a traversal tree for the adjusted condition judgment rules using a decision tree algorithm; judging the conditions of each branch based on the input variables starting from the root node through the traversal tree to obtain matching branch paths; determining which adjusted condition judgment rules are applicable to the current interaction request based on the matching branch paths, wherein the applicable rules are determined by the endpoint of the branch path; selecting rule sequences with high execution efficiency from the applicable rules to form candidate execution paths; calculating the expected response time of each path by simulating the candidate execution paths; determining the preferred execution path based on the candidate execution path with the lowest expected response time; and confirming the preferred execution path if the preferred execution path completely matches the feature vector representation of the current interaction request.

7. The method as described in claim 1, characterized in that, In step S5, the resource allocation adjustment scheme is obtained, including: extracting path parameters from the preferred execution path; calculating and obtaining path optimization indicators based on the path difference degree by comparing the preferred execution path with the corresponding paths in the initial execution path set; if the path optimization indicators show improvement, generating a resource allocation adjustment scheme based on the optimization indicators; simulating the application effect of the generated resource allocation adjustment scheme in the exhibition system; judging whether the resource allocation adjustment scheme reduces response latency based on the simulation results; if the response latency is reduced, using the adjustment scheme as the basis for updating; and optimizing the initial execution path set according to the adjustment scheme to obtain the optimized path set.

8. The method as described in claim 1, characterized in that, In step S6, determining the dynamic adaptation capability of path selection includes: loading updated condition judgment rules from the final interaction processing mechanism; processing subsequent user interaction requests according to the loaded rules to generate an execution path; obtaining execution data by recording the response time and resource utilization rate of processing subsequent user interaction requests; determining whether the execution efficiency has improved by comparing the execution data with historical data, wherein the improvement is determined by indicators such as reduced response time and optimized resource utilization rate; if the execution efficiency has improved, analyzing the dynamic adaptation capability of path selection, wherein the dynamic adaptation capability is obtained by the matching accuracy of the path for different requests; based on the analysis results, if the dynamic adaptation capability is higher than a preset threshold, confirming the effectiveness of the final interaction processing mechanism; adjusting the evaluation model of dynamic adaptation capability by continuously monitoring the processing of subsequent user interaction requests; and determining the dynamic adaptation capability of path selection based on the adjusted evaluation model.

9. An exhibition interaction logic processing system, used to implement the method as described in any one of claims 1-8, characterized in that, The system includes: a data acquisition unit, used to acquire condition judgment rules and runtime data in the interaction logic by collecting user interaction requests and historical access logs in the exhibition system, and obtain an initial execution path set; a model construction unit, used to construct a branch model using a decision tree algorithm based on the condition judgment rules in the initial execution path set, determine the trigger frequency of each condition judgment in the branch model, and determine the high-frequency condition judgment sequence; a sequence optimization unit, used to optimize the sequence using a genetic algorithm if there are parts in the high-frequency condition judgment sequence that do not match the runtime data, and obtain adjusted condition judgment rules; and a path determination unit, used to extract from the runtime data the adjusted condition judgment rules. The system extracts real-time features of user interaction requests to obtain feature vector representations. Based on these feature vector representations, a decision tree algorithm is used to traverse the adjusted conditional judgment rules, determine which rules apply to the current interaction request, and identify the preferred execution path. An interaction mechanism acquisition unit compares the preferred execution path with the initial execution path set to obtain path optimization indicators and a resource allocation adjustment scheme. If the resource allocation adjustment scheme shows a reduction in response latency, the conditional judgment rules are updated through the rule editing interface to obtain the final interaction processing mechanism. An execution efficiency evaluation unit processes subsequent user interaction requests based on the final interaction processing mechanism, determines whether execution efficiency has improved, and identifies the dynamic adaptability of path selection.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the method as described in any one of claims 1-8.

Images