Visual compilation system for cloud smart screen operation chain tracking

Abstract

The application discloses a cloud intelligent screen operation chain tracking visual compilation system and relates to the technical field of intelligent office systems.The system comprises an operation chain logic modeling module, a visual interactive configuration module, a compilation deployment execution module, an abnormal intelligent diagnosis module and a repair scheme closed loop module, acquires and analyzes user operation, generates a structured logic graph and converts it into a visual flowchart, automatically generates and deploys executable code, accurately diagnoses abnormalities and generates adaptive repair schemes, and forms a whole-process closed loop management; the application integrates an automatic process module, realizes whole-process automation from operation acquisition to code generation, constructs a structured graph based on a logic dependency algorithm, reduces the configuration threshold in combination with a visual interface, supports differentiated and offline deployment, shortens the development cycle and improves maintainability, meanwhile, the abnormal diagnosis and repair closed loop mechanism accurately locates the abnormal source, generates a reliable repair scheme, reduces the cost of manual troubleshooting, and improves the fault response efficiency.

Description

Technical Field

[0001] This invention relates to the field of intelligent office system technology, specifically a visual compilation system for cloud-based intelligent screen operation chain tracking. Background Technology

[0002] With the popularization of smart terminal devices and the increasing complexity of interaction scenarios, cloud-based smart screens, as a new generation of human-computer interaction carriers, are widely used in education, conferencing, and retail. Their operation chains involve multi-step and multi-condition dependent user behavior sequences. Traditional development models rely on manual sorting of logical relationships, making it difficult to dynamically adapt to differences in device performance and offline scenario requirements. At the same time, operation chain anomaly diagnosis and repair rely on expert experience and lack automated tool support, resulting in low operation and maintenance efficiency. In existing technologies, operation chain management mostly adopts static code configuration, which makes it difficult to achieve dynamic visualization of logical dependencies and differentiated deployment. Especially on low-performance devices, it is easy to experience lag or functional loss. In addition, anomaly handling relies on manual investigation of propagation paths and lacks quantitative diagnostic models, resulting in time-consuming problem localization and easy omission of key nodes. Therefore, there is an urgent need for an operation chain management system that can automate modeling, visualize configuration, intelligent diagnosis, and closed-loop repair to improve the stability and operation and maintenance efficiency of cloud-based smart screen applications.

[0003] Traditional operation chain management technologies suffer from several drawbacks: First, the identification of logical dependencies relies on manual intervention, using code comments or documentation to record operation sequences and conditions. This makes it difficult to dynamically maintain complex logical graphs, resulting in high development costs and a high susceptibility to errors. Second, compilation and deployment lack device performance awareness, generating heavy code packages that lag on low-performance smart screens, while high-performance devices suffer from functional redundancy and low resource utilization. Third, anomaly diagnosis relies on manual experience, using log analysis or node-by-node debugging to locate problems. This lack of quantitative models to assess anomaly propagation risks leads to highly subjective and inefficient repair solutions. Furthermore, traditional repair processes require manual coding and verification, making it difficult to quickly adapt to different business scenarios and further extending business downtime. These shortcomings make traditional technologies unable to meet the efficient operation and maintenance needs of smart screens in diverse scenarios. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a visual compilation system for cloud-based intelligent screen operation chain tracking. Through operation chain logic modeling module, visual interactive configuration module, compilation deployment and execution module, anomaly intelligent diagnosis module and repair solution closed-loop module, the system realizes automated management of operation chain. The system can dynamically construct logic graphs, support differentiated deployment and offline scenario adaptation, accurately locate the root cause of anomalies using an anomaly propagation entropy model, and generate the optimal repair solution through an ecosystem adaptation algorithm.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a visual compilation system for cloud-based intelligent screen operation chain tracing, the system comprising: an operation chain logic modeling module, a visual interactive configuration module, a compilation, deployment, and execution module, an anomaly intelligent diagnosis module, and a repair solution closed-loop module;

[0006] The operation chain logic modeling module collects user operation events on the cloud smart screen, analyzes the logical dependencies between operations through the operation chain logic dependency mining algorithm, constructs and maintains an operation chain logic graph with attributes, and outputs structured logical relationship data.

[0007] The visual interactive configuration module receives the logic graph data from the operation chain logic modeling module, converts it into a visual operation chain flowchart, and provides a touch interactive interface and process configuration function.

[0008] The compilation, deployment, and execution module receives configuration data from the visual interactive configuration module, parses and verifies the data to generate an executable code package, completes deployment and testing through the cloud smart screen device management system, and stores deployment history records.

[0009] The abnormal intelligent diagnosis module: Based on the operation monitoring data and the operation chain logic graph with attributes collected by the cloud intelligent screen agent, it identifies operation chain abnormalities through the abnormal propagation entropy diagnosis model, locates abnormal nodes, and outputs diagnostic results.

[0010] The closed-loop module of the repair solution receives abnormal diagnostic information, uses the repair solution ecosystem adaptation algorithm in combination with the control library and logic template to generate a visual repair solution, supports user adjustment and verification, and completes the compilation, deployment and verification of the repair code.

[0011] Furthermore, the calculation formula for the operation chain logic dependency mining algorithm in the operation chain logic modeling module is as follows: ,in, Indicates operation With operation The logical dependency strength, ranging from 0 to 1, with larger values ​​indicating stronger dependencies. It is an operation Successfully triggered The historical cumulative number of times, from the system log. Within 3 seconds after execution Execution records and statistics show that It is the first When triggered arrive The time interval is taken from the difference in timestamps in the logs. This is the maximum reasonable interval preset by the system, determined by the 95th percentile value of historical normal data. It is a static similarity based on operation type and data flow direction, stored in the system control library. To fix the minimum value at 0.001 and avoid a denominator of 0, the calculation first accumulates the product of the normalized time percentage of each trigger and the static similarity, then divides by 0. The result is obtained, which is generated by the operation chain logic modeling module and passed to the visualization module for display.

[0012] Furthermore, the specific steps for constructing and maintaining the attributed operation chain logic graph in the operation chain logic modeling module are as follows: collect and store basic information and related data of user operations; clean and organize the original operation events to generate standardized operation sequences containing deduplication, temporal sorting, and association tags; based on the operation sequences, mine three types of logical dependencies: temporal, data, and conditional; transform the operations and dependencies into nodes and edges in the graph database to complete the graph initialization; periodically collect new data, add nodes, adjust relationships, and clean up outdated data to dynamically maintain the graph; and output structured logical relationship data containing node lists, edge lists, and version information through a standardized interface.

[0013] Furthermore, the specific steps for generating a runnable code package in the compilation, deployment, and execution module are as follows: receiving configuration data from the visualization interaction configuration module through a preset data interface and temporarily storing it; calling the parsing engine to convert it into an intermediate representation format recognizable by the system; verifying the legality of the configuration data based on the structured logical relationship data of the operation chain logic modeling module, and generating a verification report if problems exist; matching the corresponding template from the code template library based on the verified intermediate representation format, associating the cloud smart screen adaptation control component code, filling in specific parameters to generate code snippets, and optimizing them; finally, integrating the code snippets, resource files, and configuration description files into a runnable code package with a unique identifier according to the specifications, and storing it in the system code repository.

[0014] Furthermore, the compilation, deployment, and execution module supports differentiated deployment based on the performance of cloud-based smart screen devices. When device hardware parameters are detected, it automatically optimizes and generates a lightweight code package to ensure stable operation of the operation chain on low-performance devices, while retaining the complete functional code package for high-performance devices.

[0015] Furthermore, the compilation, deployment, and execution module pre-caches compiled code packages for commonly used operation chains for offline scenarios of the smart screen. When the smart screen is disconnected from the server network, the cached code packages are automatically activated to complete the deployment test, and the cached packages are periodically updated with the server to ensure that the operation chain logic of offline deployment is consistent with the latest dependencies.

[0016] Furthermore, the calculation formula for the anomaly propagation entropy diagnostic model in the anomaly intelligent diagnostic module is as follows: ,in This is the anomaly propagation entropy; the larger the value, the higher the risk of anomaly. An anomaly is determined when the value is greater than or equal to a preset threshold of 0.6. It represents the total number of nodes in the operation chain, taken from the logical graph; It is an operation The probability of triggering a downstream anomaly, ranging from 0 to 1. It is an operation Current logical dependency values ​​are collected in real time through the cloud smart screen agent; yes The normal baseline value is the average of normal data over the past 30 days, and it is automatically updated daily. These are weighting coefficients, which can be adjusted according to the scenario. (Formula part 1) Reflecting the uncertainty of abnormal propagation, when When this item is 0, it conforms to the logic that deterministic anomalies do not require the calculation of uncertainty; Part Two This reflects the deviation from the normal baseline; the larger the deviation, the higher the value. An anomaly propagation entropy threshold is preset. Calculated The operation chain is determined to be abnormal. It is an index variable used to traverse nodes in the operation chain.

[0017] Furthermore, after identifying anomalies in the operation chain, the anomaly intelligent diagnosis module automatically generates a visualization map of the anomaly propagation path. The propagation trajectory of the anomaly from the initial node to the affected node is marked with red dynamic lines and overlaid on the operation chain flowchart on the cloud smart screen to help users intuitively locate the scope of the anomaly spread. The propagation path map supports gesture zooming to view details.

[0018] Furthermore, the determination of abnormal nodes in the abnormal intelligent diagnosis module is based on the abnormal propagation entropy diagnosis model. The abnormal propagation entropy value of each operation node is calculated by combining the extracted abnormal features. During the calculation process, the abnormal features of the node itself and the degree of influence on downstream nodes are considered. The nodes are sorted from high to low entropy value, and the node with the highest entropy value is marked as the suspected abnormal root source node. The suspected abnormal root source node is subjected to multi-dimensional cross-validation, including logical verification, historical verification, and correlation verification. After cross-validation, the abnormal root source node is determined.

[0019] Furthermore, the calculation formula for the ecological adaptation algorithm of the repair scheme in the closed-loop module of the repair scheme is as follows: ,in This refers to the compatibility of the repair solution, with a value ranging from 0 to 1. A higher value indicates better compatibility. It's cosine similarity, calculated using the system's built-in vector library. It is a 10-dimensional feature vector of the candidate repair scheme. This is the feature vector of the normal operating chain, generated based on normal data from the past 30 days. It is the feature vector of the exception operation chain, generated based on the current exception data. This is the corrected predicted anomaly entropy, calculated using the anomaly propagation entropy model described above. The fixed risk coefficient is 0.3. During calculation, it is first encoded using one-hot encoding. , , Convert the vectors to the same dimension, calculate the cosine similarity, and finally obtain the fit using the formula. Calculate the ecological fit of each candidate solution. ,according to The candidate solutions are sorted from high to low based on their values, and the repair solution with the highest fit is selected.

[0020] Compared with existing technologies, this cloud-based intelligent screen operation chain tracing visualization compilation system has the following advantages:

[0021] I. This invention achieves full-process automation from user operation collection to runnable code generation by integrating operation chain logic modeling, visual interactive configuration, and compilation, deployment, and execution modules. The system constructs an attribute-based operation chain graph through a logical dependency mining algorithm, transforming complex operation relationships into structured data. Combined with a visual interactive interface, it lowers the configuration threshold. The compilation and deployment module supports differentiated code generation and can automatically optimize code packages based on device performance. It is also compatible with caching deployment mechanisms in offline scenarios, ensuring stable operation of low-performance devices. This design not only shortens the development cycle but also provides data support for subsequent anomaly diagnosis and repair through dynamic graph maintenance and historical record storage functions, improving the systematicness and maintainability of operation chain management.

[0022] Second, this invention achieves precise location and automated repair of operational chain anomalies through the collaboration of an intelligent anomaly diagnosis module and a closed-loop repair solution module. The anomaly propagation entropy diagnosis model, combined with operational monitoring data and logic graphs, can quantify the impact of anomaly nodes on downstream operations and generate a visualized propagation path graph, helping users quickly locate the root cause of the problem. The repair solution ecosystem adaptation algorithm evaluates candidate solutions through cosine similarity and predicted anomaly entropy, prioritizing the repair strategy that highly matches the normal logic and has the lowest risk. It supports manual adjustment and verification by users. This closed-loop mechanism not only reduces the cost of manual investigation but also ensures the reliability and compatibility of the repair solution through ecosystem adaptation ranking, improving the system's response speed and resolution quality for complex operational chain failures.

[0023] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0025] Figure 1 A complete execution flowchart of the visual compilation system for tracing the operation chain of the cloud-based smart screen;

[0026] Figure 2 This is a diagram illustrating the interactive framework of the visualization compilation system module for tracing the operation chain of the cloud-based smart screen. Detailed Implementation

[0027] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0028] Example 1:

[0029] Deployment and anomaly repair scenarios for the cloud-based intelligent screen operation chain in enterprise conference rooms.

[0030] A company's conference room cloud-based smart screen needs to support document presentations, video conferencing, and device linkage operations during the meeting process. Staff need to quickly configure the operation chain through the system and ensure the smooth running of the meeting.

[0031] Operation Chain Logic Modeling Module: This module proactively collects various operational events performed by staff on the cloud-based smart screen, such as "starting video conferencing software," "uploading presentation documents," and "switching display modes." It analyzes these operations using an operation chain logic dependency mining algorithm to accurately identify temporal, data, or conditional dependencies between operations. The calculation formula for the operation chain logic dependency mining algorithm is as follows: ,in, Indicates operation With operation The logical dependency strength, ranging from 0 to 1, with larger values ​​indicating stronger dependencies. It is an operation Successfully triggered The historical cumulative number of times, from the system log. Within 3 seconds after execution Execution records and statistics show that It is the first When triggered arrive The time interval is taken from the difference in timestamps in the logs. This is the maximum reasonable interval preset by the system, determined by the 95th percentile value of historical normal data. It is a static similarity based on operation type and data flow direction, stored in the system control library. To fix the minimum value at 0.001 and avoid a denominator of 0, the calculation first accumulates the product of the normalized time percentage of each trigger and the static similarity, then divides by 0. The result is generated by the operation chain logic modeling module and passed to the visualization module for display. For example, "uploading a demo document" usually triggers "switching display mode," and there is a clear temporal dependency between the two. The attributed operation chain logic graph built based on these dependencies records the basic attributes (such as operation type and triggering conditions) and relationships of each operation. The output structured logical relationship data provides accurate logical basis for subsequent visualization configuration and anomaly diagnosis, ensuring that subsequent steps have an accurate understanding of the operation chain logic. Figure 1 As shown.

[0032] The visual interactive configuration module receives the logic graph data output by the operation chain logic modeling module and transforms the abstract logical relationships into an intuitive visual operation chain flowchart. Each operation corresponds to a node, and dependencies are clearly connected by lines. The provided touch interface allows staff to directly click on nodes to adjust operation parameters (such as setting the list of participants in a video conference). The process configuration function supports adding or deleting operation nodes (such as temporarily adding a "screen sharing" operation). This visual configuration method lowers the operational threshold, allowing staff to quickly complete the operation chain configuration that meets the meeting requirements without needing to master professional coding knowledge, thus greatly improving configuration efficiency.

[0033] The compilation, deployment, and execution module receives configuration data from the visual interactive configuration module. It first parses and verifies the data to prevent subsequent operational anomalies due to incorrect configuration. After successful verification, it generates a runnable code package, which is then deployed and tested through the cloud-based smart screen device management system. Because the hardware performance of the conference room's cloud-based smart screen is relatively balanced, the system generates a code package adapted to its performance, ensuring the completeness of the operation chain without excessively consuming hardware resources. Simultaneously, the stored deployment history allows staff to easily trace the content of each deployment, and if subsequent problems arise, the possible causes in the deployment process can be quickly identified.

[0034] Anomaly Intelligent Diagnosis Module: During the meeting, if the cloud-based smart screen suddenly experiences a "video conference screen freezing" issue, the anomaly intelligent diagnosis module will immediately activate. Based on real-time operational monitoring data (such as network transmission rate and CPU utilization) collected by the cloud-based smart screen's agent and an attributed operation chain logic graph, it accurately identifies operation chain anomalies through an anomaly propagation entropy diagnostic model. The calculation formula for the anomaly propagation entropy diagnostic model is as follows: ,in This is the anomaly propagation entropy; the larger the value, the higher the risk of anomaly. An anomaly is determined when the value is greater than or equal to a preset threshold of 0.6. It represents the total number of nodes in the operation chain, taken from the logical graph; It is an operation The probability of triggering a downstream anomaly, ranging from 0 to 1. It is an operation Current logical dependency values ​​are collected in real time through the cloud smart screen agent; yes The normal baseline value is the average of normal data over the past 30 days, and it is automatically updated daily. These are weighting coefficients, which can be adjusted according to the scenario. (Formula part 1) Reflecting the uncertainty of abnormal propagation, when When this item is 0, it conforms to the logic that deterministic anomalies do not require the calculation of uncertainty; Part Two This reflects the deviation from the normal baseline; the larger the deviation, the higher the value. An anomaly propagation entropy threshold is preset. Calculated The operation chain is determined to be abnormal. It is an index variable used to traverse nodes in the operation chain. This model can comprehensively analyze the propagation of anomalies in the operation chain, quickly locate the abnormal node "network data reception", and generate detailed diagnostic results. At the same time, it automatically generates a visualization map of the anomaly propagation path, marking the anomaly propagation trajectory with red dynamic lines, allowing staff to intuitively understand the scope of the anomaly's impact and providing a clear direction for subsequent repairs.

[0035] Repair Solution Closed-Loop Module: Upon receiving anomaly diagnostic information, the module utilizes the repair solution ecosystem adaptation algorithm, combined with the system's control library (such as network adaptation controls) and logic templates (such as network anomaly handling templates), to generate multiple visual repair solutions (such as "Switch to Backup Network" and "Optimize Data Reception Cache"). The calculation formula for the repair solution ecosystem adaptation algorithm is as follows: ,in This refers to the compatibility of the repair solution, with a value ranging from 0 to 1. A higher value indicates better compatibility. It's cosine similarity, calculated using the system's built-in vector library. It is a 10-dimensional feature vector of the candidate repair scheme. This is the feature vector of the normal operating chain, generated based on normal data from the past 30 days. It is the feature vector of the exception operation chain, generated based on the current exception data. This is the corrected predicted anomaly entropy, calculated using the anomaly propagation entropy model described above. The fixed risk coefficient is 0.3. During calculation, it is first encoded using one-hot encoding. , , Convert the vectors to the same dimension, calculate the cosine similarity, and finally obtain the fit using the formula. Calculate the ecological fit of each candidate solution. ,according to The algorithm sorts the candidate solutions from high to low values ​​and selects the repair solution with the highest adaptability. From the perspective of ecological adaptability, the algorithm can select the solution most suitable for the current cloud smart screen and meeting scenario. Staff can adjust the solution parameters and verify them according to the actual situation. After confirmation, the system automatically completes the compilation, deployment and verification of the repair code. The whole process forms a closed loop to ensure that the operation chain after repair can run stably and quickly restore the normal operation of the meeting.

[0036] In summary, in the context of enterprise conference rooms, this system constructs logical relationships through an operation chain logic modeling module, reduces configuration difficulty through a visual interactive configuration module, ensures adaptability and traceability through a compilation, deployment, and execution module, and, in the event of anomalies, uses an intelligent anomaly diagnosis module to quickly locate problems and a closed-loop repair solution module to efficiently resolve issues. The collaboration of these modules not only improves the efficiency of operation chain configuration and deployment but also enables timely handling of anomalies, ensuring the stable operation of the cloud-based intelligent screen operation chain during meetings.

[0037] Example 2: Example of a digital product promotion scenario in a corporate showroom.

[0038] A company's showroom has built a digital brand promotion system using a cloud-based smart screen. The system needs to automate the entire brand promotion process: "visitor touches the screen to wake up → automatically plays the company promotional video → displays 3D product models → pushes electronic brochures." This scenario places high demands on the smoothness of the operation chain, the adaptability of the display effects, and the speed of emergency response. The following is a description of the specific applications and functions of each module in the system:

[0039] Operation chain logic modeling module:

[0040] Real-time data collection of user operation events on the exhibition hall's smart screen (such as visitors touching to wake the screen, swiping to switch product models), and analysis of the logical relationships between operations using an operation chain logical dependency mining algorithm. The calculation formula for the operation chain logical dependency mining algorithm is as follows: For example, "touch wake-up" is the prerequisite for all subsequent operations, and "play promotional video" needs to be started after "wake-up complete". There is a clear time dependency between the two. "Display 3D model" and "push electronic brochure" can be executed in parallel, but they depend on the condition of "promotional video playing to 50% progress" to be triggered.

[0041] Based on these dependencies, the module will build and maintain an operation chain logic graph with attributes, record the device association information (such as the rendering engine corresponding to the 3D model, the file format of the brochure) and triggering rules for each operation, and output structured logical data.

[0042] Visual interactive configuration module:

[0043] After receiving the operation chain logic diagram data, it is transformed into an intuitive and visual operation chain flowchart. Each brand promotion link is presented in the form of nodes, and the dependencies between nodes are marked with lines (such as the "wake-up" node pointing to the "play promotional video" node). Exhibition hall staff can directly configure parameters through the touch interactive interface: for example, adjusting the playback duration of the promotional video (from the default 3 minutes to 5 minutes), setting the rotation display angle of the 3D model, and selecting the push timing of the electronic brochure (such as automatically pushing it after the audience stays for 10 seconds).

[0044] In addition, the module supports adding temporary operation nodes (such as adding a "Book an Appointment" button after "Push Brochure"), and the process adjustment can be completed without writing code.

[0045] Compile, deploy, and execute the module:

[0046] After receiving the visualization configuration data, the module first parses and verifies its legality to avoid display interruptions due to configuration errors. Once the verification is successful, a runnable code package is generated and deployed through the exhibition hall equipment management system. Considering that there may be multiple cloud-based smart screens in the exhibition hall (some are high-performance devices supporting 4K rendering, and some are ordinary high-definition devices), the module will automatically detect the device performance: generate a complete functional code package for high-performance devices (preserving the lighting effects of 3D models and the 4K resolution of promotional videos); generate a lightweight code package for ordinary devices (simplifying model details and reducing video resolution) to ensure that all devices can run smoothly. At the same time, in case of unstable network conditions in the exhibition hall, the module will pre-cache the code packages of commonly used brand promotion processes and automatically activate the cached packages when the network is interrupted to avoid the termination of brand promotion processes due to network outages.

[0047] Anomaly Intelligent Diagnosis Module:

[0048] If the Cloud Smart Screen encounters an anomaly during display (such as 3D model loading failure or inability to push brochures), the module will immediately initiate a diagnostic process: based on the real-time operational data collected by the Cloud Smart Screen (such as memory usage and file loading progress) and the operation chain logic diagram, the anomaly propagation entropy diagnostic model is used to analyze the scope of the anomaly. The calculation formula for the anomaly propagation entropy diagnostic model is as follows: Calculated The operation chain is determined to be abnormal. It is an index variable used to traverse nodes in the operation chain. For example, after the "3D model loading" node is found to be abnormal, the algorithm will evaluate its impact on downstream operations (such as whether it causes the "pushing brochure" synchronization to be interrupted) and locate the root cause of the abnormality (such as model file corruption or insufficient device memory). At the same time, the module will generate a visual abnormality propagation path map: the trajectory of the abnormality spreading from the "model loading" node to other nodes is marked with red lines and displayed on the cloud smart screen backend. Staff can intuitively see the scope of the fault (such as only the model display is affected, while the promotional video is still playing normally).

[0049] Repair solution closed-loop module:

[0050] After receiving the anomaly diagnosis information, the system combines the brand promotion scene control library (such as lightweight 3D model templates and backup brochure files) and the repair solution ecosystem adaptation algorithm to generate multiple repair solutions. For example, for "model loading failure," it recommends "replacing with a low-precision model" or "skipping model display and directly pushing the brochure." The algorithm will prioritize the solution that best suits the current scene. The calculation formula for the repair solution ecosystem adaptation algorithm is as follows: Calculate the ecological suitability of each candidate solution. ,according to The system sorts candidate solutions from highest to lowest value and selects the repair solution with the highest suitability. (For example, during peak exhibition periods, the system prioritizes "skipping the model" to ensure process continuity, and during customer reception, the system prioritizes "replacing the low-precision model" to preserve the integrity of the display.) Staff can select a solution and make minor adjustments on the visual interface (such as setting the display angle of the low-precision model). After confirmation, the system automatically completes the compilation, deployment, and verification of the repair code. For example, after deploying the low-precision model, the 3D display function is restored, and the brand promotion process continues to run.

[0051] In summary, in the context of digital brand promotion in corporate showrooms, the system achieves seamless collaboration across its modules: the operation chain logic modeling module ensures rigorous process logic, the visual configuration module lowers the customization threshold, the compilation and deployment module adapts to different devices and network environments, the anomaly diagnosis module quickly locates problems, and the closed-loop repair module efficiently resolves faults. Overall, it realizes "zero-code customization," "full device compatibility," and "second-level fault response" in the brand promotion process, enhancing the professionalism and stability of corporate brand presentation.

[0052] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A visual compilation system for tracing the operation chain of a cloud-based smart screen, characterized in that, The system includes: an operation chain logic modeling module, a visual interactive configuration module, a compilation, deployment and execution module, an intelligent anomaly diagnosis module, and a closed-loop repair solution module; The operation chain logic modeling module collects user operation events from the cloud smart screen and analyzes the logical dependencies between operations using an operation chain logic dependency mining algorithm. The calculation formula for the operation chain logic dependency mining algorithm is as follows: ,in This represents the logical dependency strength between operation i and operation j, with a value ranging from 0 to 1. A larger value indicates a stronger dependency. It is the cumulative number of times operation i has successfully triggered operation j. It is the first The actual time interval between operation i and operation j at the time of the next trigger. It is the maximum reasonable time interval preset by the system, obtained through historical data statistics. It is a static similarity based on operation type and data flow direction, stored in the system control library, and preset based on operation type and data flow direction, with a value range of 0~1. It is a minimum value, fixed at 0.001, to avoid the denominator being 0. It constructs and maintains an operation chain logic graph with attributes and outputs structured logical relationship data. The visual interactive configuration module receives the logic graph data from the operation chain logic modeling module, converts it into a visual operation chain flowchart, and provides a touch interactive interface and process configuration function. The compilation, deployment, and execution module receives configuration data from the visual interactive configuration module, parses and verifies the data to generate an executable code package, completes deployment and testing through the cloud smart screen device management system, and stores deployment history records. The anomaly intelligent diagnosis module, based on the operation monitoring data and attributed operation chain logic graph collected by the cloud intelligent screen agent, identifies operation chain anomalies through an anomaly propagation entropy diagnosis model. The calculation formula for the anomaly propagation entropy diagnosis model is as follows: ,in This is the anomaly propagation entropy; the higher the value, the higher the risk of anomalies. It is the total number of nodes in the operation chain. It represents the probability that operation k will trigger a subsequent exception, based on historical exception data statistics, and its value ranges from 0 to 1. It is the current logical dependency value of operation k. It is the normal logical dependency benchmark value of operation k, calculated as the average value through normal historical data. These are weighting coefficients, set according to the business scenario, with a pre-defined anomaly propagation entropy threshold. Calculated The operation chain is determined to be abnormal. k is the index variable used to traverse the nodes in the operation chain, locate the abnormal node and output the diagnosis result. The closed-loop module of the repair scheme receives anomaly diagnosis information and generates a visual repair scheme by combining the repair scheme ecosystem adaptation algorithm with the control library and logic template. The calculation formula for the repair scheme ecosystem adaptation algorithm is as follows: ,in This refers to the compatibility of the repair solution, with a value ranging from 0 to 1. A higher value indicates better compatibility. It's cosine similarity, calculated using the system's built-in vector library. These are the feature vectors of the candidate repair schemes. These are feature vectors of the normal operating chain, generated based on historical normal data. It is the feature vector of the exception operation chain, generated based on the current exception data. This is the corrected predicted anomaly entropy, with a value ranging from 0 to 1. It's a risk factor, fixed at 0.3, based on... The candidate solutions are sorted from high to low based on their values, and the solution with the highest adaptability is selected. Users can adjust and verify the solution, and the repair code is compiled, deployed, and verified.

2. The visual compilation system for cloud-based intelligent screen operation chain tracing according to claim 1, characterized in that, The specific steps for constructing and maintaining an attributed operation chain logic graph in the operation chain logic modeling module are as follows: Collect and store basic information and associated data of user operations; clean and organize the original operation events to generate standardized operation sequences containing deduplication, temporal sorting, and association tags; based on the operation sequences, mine three types of logical dependencies: temporal, data, and conditional; transform operations and dependencies into nodes and edges in the graph database to complete graph initialization; periodically collect new data, add nodes, adjust relationships, and clean up outdated data to dynamically maintain the graph; and output structured logical relationship data containing node lists, edge lists, and version information through a standardized interface.

3. The visual compilation system for cloud-based intelligent screen operation chain tracing according to claim 1, characterized in that, The specific steps for generating a runnable code package in the compilation, deployment, and execution module are as follows: receiving configuration data from the visualization interaction configuration module through a preset data interface and temporarily storing it; calling the parsing engine to convert it into an intermediate representation format recognizable by the system; verifying the legality of the configuration data based on the structured logical relationship data of the operation chain logic modeling module, and generating a verification report if problems exist; matching the corresponding template from the code template library based on the verified intermediate representation format, associating the cloud smart screen adaptation control component code, filling in specific parameters to generate code snippets, and optimizing them; finally, integrating the code snippets, resource files, and configuration description files into a runnable code package with a unique identifier according to the specifications, and storing it in the system code repository.

4. The visual compilation system for cloud-based intelligent screen operation chain tracing according to claim 3, characterized in that, The compilation, deployment, and execution module supports differentiated deployment based on the performance of cloud-based smart screen devices. When device hardware parameters are detected, it automatically optimizes and generates a lightweight code package to ensure stable operation of the operation chain on low-performance devices, while retaining the complete functional code package for high-performance devices.

5. The visual compilation system for cloud-based intelligent screen operation chain tracing according to claim 3, characterized in that, In the compilation, deployment, and execution module, for offline scenarios of the cloud smart screen, the compiled code packages of commonly used operation chains are pre-cached. When the cloud smart screen is disconnected from the server network, the cached code packages are automatically activated to complete the deployment test. The cached packages are also updated regularly with the server to ensure that the operation chain logic of offline deployment is consistent with the latest dependencies.

6. The visual compilation system for cloud-based intelligent screen operation chain tracing according to claim 1, characterized in that, After identifying an anomalies in the operation chain, the intelligent anomaly diagnosis module automatically generates a visualization map of the anomaly propagation path. The propagation trajectory of the anomaly from the initial node to the affected node is marked with red dynamic lines and overlaid on the operation chain flowchart on the cloud smart screen to help users intuitively locate the scope of the anomaly spread. The propagation path map supports gesture zooming to view details.

7. The visual compilation system for cloud-based intelligent screen operation chain tracing according to claim 1, characterized in that, The determination of abnormal nodes in the abnormal intelligent diagnosis module is based on the abnormal propagation entropy diagnosis model. It calculates the abnormal propagation entropy value of each operation node by combining the extracted abnormal features. In the calculation process, the abnormal features of the node itself and the degree of influence on downstream nodes are taken into account. The nodes are sorted from high to low entropy value, and the node with the highest entropy value is marked as the suspected abnormal root source node. The suspected abnormal root source node is subjected to multi-dimensional cross-validation, including logical verification, historical verification, and correlation verification. After cross-validation, the abnormal root source node is determined.

Images