Method and system for dynamic arrangement of CAD toolbar based on operation chain mining

Abstract

This invention discloses a method and system for dynamic arrangement of CAD toolbars based on operation chain mining, relating to the field of data processing technology. The method includes: monitoring the user's current CAD operation node; performing backward operation chain mining on the operation feature vector of the current CAD operation node to obtain a candidate set of backward CAD operation nodes; setting intermediate resource allocation nodes; obtaining resource indicators that can be used for preloading; and determining whether these indicators are less than the sum of resource requirement indicators in the candidate set of backward CAD operation nodes. If less, tracing back the user's forward CAD operation nodes; re-performing backward operation chain mining on the operation chains of the forward and current CAD operation nodes to obtain an updated set of backward operation nodes; and performing toolbar arrangement preloading on the updated set of backward operation nodes. This invention solves the technical problem in the prior art that CAD toolbars cannot be dynamically adjusted according to user operation behavior, achieving the technical effect of improving CAD operation efficiency.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for dynamic arrangement of CAD toolbars based on operation chain mining. Background Technology

[0002] In CAD systems, toolbars typically employ fixed layouts and static configurations. The functional tools users access vary across different design stages and operational scenarios, but toolbars lack the ability to adjust in real-time based on user behavior. As design processes become more complex and operational chains lengthen, users frequently switch between multiple menus or toolbars, resulting in lengthy operation paths, untimely resource loading, significantly reduced tool access efficiency, and increased system response latency. This makes it difficult to meet the demands for efficient and intelligent CAD design interaction. Summary of the Invention

[0003] This application provides a method and system for dynamic arrangement of CAD toolbars based on operation chain mining, which is used to address the technical problem that CAD toolbars in the prior art cannot be dynamically adjusted according to user operation behavior.

[0004] In view of the above problems, this application provides a method and system for dynamic arrangement of CAD toolbars based on operation chain mining.

[0005] The first aspect of this application provides a method for dynamically arranging CAD toolbars based on operation chain mining, the method comprising:

[0006] The system monitors the user's current CAD operation node and performs backward operation chain mining on the operation feature vector of the current CAD operation node using the historical operation chain mining engine to obtain a candidate set of backward CAD operation nodes. An intermediate resource allocation node is set, and resource indicators available for preloading in the resource pool are obtained according to the intermediate resource allocation node. It is determined whether the resource indicators available for preloading are less than the sum of resource requirements in the candidate set of backward CAD operation nodes. If they are less, the user's forward CAD operation node is traced according to the adaptive context window step size. The system then performs backward operation chain mining again on the operation chains of the forward CAD operation node and the current CAD operation node using the historical operation chain mining engine to obtain an updated set of backward operation nodes. The intermediate resource allocation node preloads the toolbar arrangement of the updated set of backward operation nodes.

[0007] A second aspect of this application provides a dynamic arrangement system for CAD toolbars based on operation chain mining, the system comprising:

[0008] The monitoring module monitors the user's current CAD operation node and performs backward operation chain mining on the operation feature vector of the current CAD operation node based on the historical operation chain mining engine to obtain a candidate set of backward CAD operation nodes. The resource indicator acquisition module sets intermediate resource allocation nodes and acquires resource indicators from the resource pool that can be used for preloading according to the intermediate resource allocation nodes. The judgment module determines whether the resource indicators available for preloading are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes. If they are less, it traces the user's forward CAD operation node according to the adaptive context window step size. The mining module re-mines the backward operation chain based on the operation chains of the forward CAD operation node and the current CAD operation node using the historical operation chain mining engine to obtain an updated set of backward operation nodes. The orchestration preloading module performs toolbar orchestration preloading on the updated set of backward operation nodes by the intermediate resource allocation nodes.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] This application monitors the user's current CAD operation node, performs backward operation chain mining on the operation feature vector of the current CAD operation node using a historical operation chain mining engine, and obtains a candidate set of backward CAD operation nodes. An intermediate resource allocation node is set, and resource indicators available for preloading in the resource pool are obtained according to the intermediate resource allocation node. It is determined whether the resource indicators available for preloading are less than the sum of resource requirements in the candidate set of backward CAD operation nodes. If they are less, the user's forward CAD operation node is traced according to an adaptive context window step size. The operation chains of the forward CAD operation node and the current CAD operation node are re-mined using the historical operation chain mining engine to obtain an updated set of backward operation nodes. The intermediate resource allocation node preloads the toolbar arrangement based on the updated set of backward operation nodes. This invention solves the technical problem in the prior art that the CAD toolbar cannot be dynamically adjusted according to user operation behavior. By mining and predicting the user's operation chain, adaptive toolbar arrangement and resource preloading are achieved, thereby improving the technical effect of CAD operation efficiency. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic diagram of the dynamic arrangement method of CAD toolbar based on operation chain mining provided in this application embodiment;

[0013] Figure 2 A schematic diagram of the structure of a dynamic CAD toolbar arrangement system based on operation chain mining provided in this application embodiment.

[0014] Figure labeling: Monitoring module 11, resource indicator acquisition module 12, judgment module 13, mining module 14, orchestration preloading module 15. Detailed Implementation

[0015] This application provides a method and system for dynamic arrangement of CAD toolbars based on operation chain mining. It addresses the technical problem that CAD toolbars cannot be dynamically adjusted according to user operation behavior in the prior art. By mining and predicting the user operation chain, it achieves adaptive arrangement and resource preloading of the toolbar, thereby improving the technical effect of CAD operation efficiency.

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0017] It should be noted that any variation of the terms "comprising" and "having" is intended to cover non-exclusive inclusion, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such processes, methods, products, or devices.

[0018] Example 1, as Figure 1 As shown, this application provides a method for dynamically arranging CAD toolbars based on operation chain mining, the method comprising:

[0019] Step S100: Monitor the user's current CAD operation node, and perform backward operation chain mining on the operation feature vector of the current CAD operation node according to the historical operation chain mining engine to obtain a candidate set of backward CAD operation nodes.

[0020] In this embodiment of the application, the user's operation behavior in the CAD interface is monitored in real time, and operation events such as command invocation, parameter input, and object selection are captured in real time. The events are then abstracted into the current CAD operation node according to the order and type of the events.

[0021] Subsequently, the historical operation chain mining engine is used to perform backward operation chain mining on the operation feature vector of the current CAD operation node. The operation feature vector consists of command features, object features, parameter features, time features, and user features. Based on this operation feature vector, the historical operation chain mining engine calculates similar historical sequence samples in the historical operation chain sample library. By comparing the feature similarity with a preset similarity threshold, a candidate historical chain set is obtained, and backward operation chain mining is performed, outputting a candidate set of backward CAD operation nodes.

[0022] Furthermore, in the method provided in the application embodiments, the backward operation chain mining is performed on the operation feature vector of the current CAD operation node based on the historical operation chain mining engine to obtain a candidate set of backward CAD operation nodes, and the method further includes:

[0023] The operation feature vector includes command features, object features, parameter features, time features, and user features. Based on the historical operation chain mining engine, historical sequence samples similar to the operation feature vector are calculated in the historical operation chain sample library to obtain a candidate historical chain set composed of feature similarity greater than a preset similarity threshold. Based on the candidate historical chain set, backward operation chain mining is performed to obtain a candidate set of backward CAD operation nodes.

[0024] In this embodiment, the operation feature vector includes command features, object features, parameter features, time features, and user features. To obtain the operation feature vector of the user's current CAD operation node, different types of features are numerically processed. Command features are converted into numerical representations using a preset function instruction encoding table; object features are mapped to numbered forms based on geometric object types; parameter features are normalized to a fixed range; time features are represented by the relative position and time interval of the operation in the sequence; and user features are converted into statistical values ​​based on usage frequency, call time, and other information. All features are standardized and then concatenated in a fixed order to form the operation feature vector.

[0025] Next, the historical operation chain mining engine calculates historical sequence samples similar to the operation feature vectors in the historical operation chain sample library. The historical operation chain sample library is a data set formed by collecting the actual operation behaviors of different users in the CAD system over a long period of time. It records the complete operation chain of a user from one CAD operation node to another in chronological order. Each operation chain consists of multiple operation nodes, and each operation node corresponds to an operation feature vector.

[0026] When the historical operation chain mining engine calculates similarity between historical sequence samples and operation feature vectors in the historical operation chain sample library, it uses the cosine similarity method for similarity calculation. During the calculation, a corresponding sequence vector is first generated for each historical sequence in the historical operation chain sample library. The operation feature vectors of each node in the sequence are merged according to their actual order to obtain a vector representation representing the entire operation chain. Then, the cosine similarity is calculated between the current operation feature vector and each historical sequence vector to obtain a similarity score. Finally, the similarity score is compared with a preset similarity threshold, and historical sequence samples with feature similarity greater than the preset similarity threshold are retained to form a candidate historical chain set.

[0027] Finally, backward operation chain mining is performed based on the candidate historical chain set. This process begins by locating the current CAD operation node's position within the candidate historical chain set and extracting the node sequence following that position. Then, by statistically analyzing the frequency of subsequent nodes in the historical chain, the conditional probability of their occurrence under the current CAD operation node is calculated. This probability is used to filter subsequent nodes, ultimately yielding the candidate set of backward CAD operation nodes.

[0028] Furthermore, in the method provided in the application embodiments, the backward operation chain mining based on the candidate historical chain set to obtain the backward CAD operation node candidate set further includes:

[0029] Extract the node sequence following the position corresponding to the current CAD operation node from the candidate historical chain set; according to the calculated conditional probability of the node sequence appearing under the current CAD operation node, filter from the node sequence based on the conditional probability to obtain the backward CAD operation node candidate set.

[0030] In this embodiment of the application, the current CAD operation node is first located in the candidate historical chain set, and the node sequence following the corresponding position of the node is extracted.

[0031] Next, the conditional probability is calculated using frequency statistics. Taking the current CAD operation node as the conditional event, all node sequences are traversed and statistically analyzed. The number of times each node sequence appears after the current CAD operation node is recorded, and the total number of times the current CAD operation node is located in the candidate historical chain set is also counted. The conditional probability P(node ​​sequence | current CAD operation node) is calculated by dividing the number of occurrences of the node sequence by the total number of occurrences of the current CAD operation node.

[0032] Subsequently, an operation chain graph is constructed based on the node sequence. The operation chain graph consists of multiple operation nodes, and directed connections are established between the current CAD operation node and each node in the node sequence, using the temporal relationships between nodes as directed connections. In this operation chain graph, the operation dependency is calculated by statistically analyzing the frequency of connections between the node sequence and the current CAD operation node.

[0033] Finally, conditional probability and operational dependency are combined to calculate a comprehensive score for each node sequence, and the sequences are sorted in descending order of comprehensive score. Node sequences are then filtered according to preset screening criteria, eliminating those with low comprehensive scores and retaining those that are relevant to the current CAD operation node and have a high frequency of occurrence. This process yields a candidate set of backward CAD operation nodes.

[0034] Furthermore, in the method provided in the application embodiments, obtaining the candidate set of backward CAD operation nodes further includes:

[0035] Construct an operation chain graph, and identify the operation dependency between the current CAD operation node and the node sequence through the operation chain graph; re-update the node sequence and score it according to the conditional probability and the operation dependency to obtain a new candidate set of backward CAD operation nodes.

[0036] In this embodiment, an operation chain graph is first constructed based on a candidate historical chain set. During the construction process, all operation nodes appearing in the candidate historical chain set are used as vertices, and the temporal relationship between operation nodes is used as directed edges. The sequential relationship between the current CAD operation node and each operation node in the node sequence is represented by directed connections, thereby obtaining the operation chain graph.

[0037] Next, the operation dependency between the current CAD operation node and the node sequence is identified through the operation chain graph. In this process, starting from the current CAD operation node, the paths related to the node sequence are traversed along the directed edges. The number of times the current CAD operation node and each operation node in the node sequence co-occur in the candidate historical chain set is counted, and the proportion of this co-occurrence in the overall candidate historical chain set is calculated, thereby obtaining the operation dependency.

[0038] Finally, the node sequences are re-scored based on conditional probability and operational dependency. In this process, the conditional probability of a node sequence is directly multiplied by its operational dependency to obtain a comprehensive score. This comprehensive score is then compared with preset screening criteria. If the comprehensive score of a node sequence is greater than or equal to the preset screening criteria, the node sequence is retained; if the comprehensive score is lower than the preset screening criteria, the node sequence is removed. After completing the comprehensive score calculation and comparison with the preset screening criteria, the retained node sequences are output as the result, yielding the candidate set of backward CAD operation nodes.

[0039] Furthermore, in the method provided in the application embodiments, monitoring the user's current CAD operation node also includes:

[0040] A preset context window step size is used to monitor the set of CAD operation nodes operated by the user within the preset context window step size. Based on the historical operation chain mining engine, backward operation chain mining is performed on the set of operation feature vectors of the set of CAD operation nodes to obtain a candidate set of backward CAD operation nodes.

[0041] In this embodiment, a preset context window step size is first set to define the time range and number of nodes for tracing back the user's historical operations. Then, within the constraints of the preset context window step size, the set of CAD operation nodes within that range is monitored in real time. The set of CAD operation nodes consists of multiple CAD operation nodes arranged chronologically, completely recording the user's continuous operation path prior to the current CAD operation node.

[0042] Next, the historical operation chain mining engine performs backward operation chain mining on the set of operation feature vectors of the CAD operation node set. In this process, the historical operation chain mining engine identifies historical operation chains similar to the set of operation feature vectors by performing similarity matching with historical sequence samples in the historical operation chain sample library, and then performs backward operation chain mining to determine the backward operation chain mining results. Finally, the backward operation chain mining results are compared with preset screening criteria, and through integration and screening, a new candidate set of backward CAD operation nodes is obtained.

[0043] Step S200: Set up an intermediate resource allocation node, and obtain the resource indicators in the resource pool that can be used for preloading according to the intermediate resource allocation node.

[0044] In this embodiment, an intermediate resource allocation node is first set up to serve as the execution core for resource status aggregation and scheduling during resource scheduling and preloading. The intermediate resource allocation node establishes a fixed communication interface with the resource pool to collect and calculate the real-time status of the resource pool.

[0045] Subsequently, the intermediate resource allocation node accesses the resource pool, directly retrieving the current availability status data of various resource types from the pool through a combination of periodic polling and real-time querying. The resource pool contains information on command loading resources, rendering resources, memory resources, thread resources, and UI refresh resources. The status data for each resource type includes its current available capacity and total capacity. During the data collection process, the intermediate resource allocation node determines the available capacity of each resource type at the current moment based on the data returned from the resource pool and records these capacity values.

[0046] Finally, the collected available capacity information is organized to obtain resource metrics that can be used for preloading.

[0047] Step S300: Determine whether the resource indicators available for preloading are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes. If they are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes, trace the user's forward CAD operation node according to the adaptive context window step size.

[0048] In this embodiment, the sum of the resource indicators available for preloading and the resource demand indicators of the candidate set of backward CAD operation nodes is first determined. When the resource indicators available for preloading are less than the sum of the resource demand indicators of the candidate set of backward CAD operation nodes, a resource gap ratio is calculated. The resource gap ratio is calculated based on the difference between the resource demand indicators of the candidate set of backward CAD operation nodes and the resource indicators available for preloading. Subsequently, the adaptive context window step size is dynamically determined based on this resource gap ratio. The larger the gap, the larger the adaptive context window step size is, so as to introduce more historical operation information when resources are insufficient.

[0049] Finally, the user's preceding CAD operation nodes are traced according to the adaptive context window step size. Through this process, under resource-constrained conditions, the context tracing range is flexibly adjusted according to the actual resource gap, ultimately obtaining the user's preceding CAD operation node most relevant to the current operation.

[0050] Furthermore, in the method provided in the application embodiments, calculating the resource requirement index of the candidate set of backward CAD operation nodes further includes:

[0051] The command loading cost, rendering load resources, memory usage assessment, thread usage, and interface refresh latency of each operation node in the candidate set of backward CAD operation nodes are collected; the command loading cost, rendering load resources, memory usage assessment, thread usage, and interface refresh latency are comprehensively weighted and calculated to output the resource requirement index of each operation node; the resource requirement index of each operation node is summed to obtain the resource requirement index of the candidate set of backward CAD operation nodes.

[0052] In this embodiment, the resource usage of each operation node in the candidate set of backward CAD operation nodes is first collected. Command loading cost is obtained by recording the time from the start of command loading to the completion of the function call, for example, in milliseconds representing the duration of the loading process. Rendering load resources are obtained by collecting the number of drawing calls and the duration of rendering processing, for example, in milliseconds representing the time occupied by the rendering pipeline. Memory usage assessment is obtained by reading memory usage information during operation node execution, for example, in MB representing the amount of new memory usage and peak usage during operation. Thread utilization is obtained by statistically analyzing thread resource usage during operation node execution, for example, in percentage representing the proportion of available worker threads. Interface refresh latency is obtained by recording the time it takes for the interface to update after the operation node is triggered, for example, in milliseconds representing the interface response latency. Through this process, quantitative data for each operation node in five aspects—command loading cost, rendering load resources, memory usage assessment, thread utilization, and interface refresh latency—is obtained.

[0053] Subsequently, a comprehensive weighted calculation is performed on the collected command loading cost, rendering load resources, memory usage assessment, thread usage, and UI refresh latency. In this process, the collected data on command loading cost, rendering load resources, memory usage assessment, thread usage, and UI refresh latency are first normalized. Each indicator is mapped to a numerical range of 0 to 1 according to its minimum and maximum values ​​in the sample, transforming data of different dimensions and orders of magnitude into a directly comparable unified standard. Then, the normalized command loading cost, rendering load resources, memory usage assessment, thread usage, and UI refresh latency are weighted and summed according to preset weights to obtain the resource requirement indicators for this operation node.

[0054] Finally, the resource requirement indicators of all operation nodes in the candidate set of backward CAD operation nodes are summed to obtain the sum of the resource requirement indicators of the candidate set of backward CAD operation nodes.

[0055] Furthermore, the method provided in the application embodiments also includes:

[0056] The user's forward CAD operation nodes are traced according to an adaptive context window step size, which is dynamically determined adaptively based on the resource gap ratio. The resource gap ratio is the difference between the resource requirement index of the candidate set of backward CAD operation nodes and the resource index used for preloading.

[0057] In this embodiment, the resource gap ratio is first calculated based on the difference between the resource requirement index of the candidate set of backward CAD operation nodes and the resource index available for preloading. The resource gap ratio represents the degree of difference between currently available resources and actually required resources. Its value is equal to the resource requirement index of the candidate set of backward CAD operation nodes minus the resource index available for preloading, and is consistent with the aforementioned resource calculation process, falling within the range of 0 to 1. For example, when the resource requirement index of the candidate set of backward CAD operation nodes is 0.8, and the resource index available for preloading is 0.5, the resource gap ratio is 0.3, indicating a 30% resource shortage.

[0058] The adaptive context window step size is then dynamically determined based on the resource gap ratio. The adaptive context window step size is a positive number representing the number of nodes or time span for tracing back historical operations; its magnitude is positively correlated with the resource gap ratio. The specific positive step size is obtained by mapping the resource gap ratio from 0 to 1 to preset minimum and maximum step sizes. For example, with a minimum step size of 2 CAD operation nodes and a maximum step size of 10 CAD operation nodes, when the resource gap ratio is 0.3, the step size is calculated using the linear formula: Step Size = Minimum Step Size + (Maximum Step Size - Minimum Step Size) × Resource Gap Ratio, resulting in 4.4. In practical applications, this can be rounded up to 5 operation nodes as needed.

[0059] Finally, based on a single operation node, the user's historical operations are traced forward incrementally according to a defined adaptive context window step size. Starting from the current CAD operation node, adjacent forward CAD operation nodes are retrieved sequentially until the step size limit is reached. For example, when the step size is 5, 5 historical operation nodes are traced, forming a set of forward CAD operation nodes that match the current resource status. Through this process, the tracing depth is precisely controlled when resources are insufficient, ultimately obtaining the user's forward CAD operation nodes adapted to resource conditions.

[0060] Furthermore, the method provided in the application embodiments also includes:

[0061] After obtaining the candidate set of backward CAD operation nodes, if the resource indicators available for preloading are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes, an adaptive solution is performed using the preset context window step size as the adaptive input variable and the resource gap as the adaptive target to obtain the adaptive context window step size.

[0062] In this embodiment, after obtaining the candidate set of backward CAD operation nodes, if it is determined that the resource indicators available for preloading are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes, the resource gap is first calculated. The resource gap is the difference between the resource requirements of the candidate set of backward CAD operation nodes and the resource indicators available for preloading.

[0063] Then, using the preset context window step size as an adaptive input variable, the system traces backwards directly based on the set of operation nodes corresponding to the preset context window step size. Starting from the current CAD operation node, the system extracts the number of preceding CAD operation nodes equal to the preset context window step size in chronological order, forming an operation node set.

[0064] Next, an adaptive solution is performed with the resource gap as the adaptive objective. In this process, the minimum and maximum values ​​of the preset context window step size are first set as the search boundaries. Starting from the initial preset context window step size, the size of the preset context window step size is iteratively adjusted. After each adjustment, a new set of operation nodes corresponding to the preset context window step size is generated, and the resource requirement indicators are recalculated based on this set of operation nodes. These recalculated indicators are then compared with the resource indicators available for preloading to obtain the new resource gap.

[0065] When the resource gap is greater than zero, it indicates that the resource consumption of the operation node set corresponding to the current preset context window step size exceeds the available resources, and the preset context window step size needs to be reduced to decrease the size of the operation node set. When the resource gap is less than zero, it indicates that the resource consumption of the operation node set corresponding to the current preset context window step size is lower than the available resources, and the preset context window step size can be increased to introduce more context information. During this process, the system continuously narrows the range of the preset context window step size using a bisection method or an interval contraction method, using the resource gap as the convergence criterion, until the resource gap is within a preset threshold range or the convergence condition is met. The final output is an adaptive context window step size that satisfies the resource gap adaptive objective.

[0066] Finally, based on the set of operation nodes corresponding to the adaptive context window step size, the user's previous CAD operation nodes are traced back.

[0067] Step S400: Based on the historical operation chain mining engine, re-mining the operation chains of the forward CAD operation node and the current CAD operation node to obtain the updated set of backward operation nodes.

[0068] Furthermore, in the method provided in the application embodiments, the backward operation chain mining is performed again on the operation chains of the forward CAD operation node and the current CAD operation node based on the historical operation chain mining engine, which further includes:

[0069] A multi-scale time window is used to extract features from the operation chains of the forward CAD operation node and the current CAD operation node, and output a multi-scale operation feature vector. The backward operation chain is mined according to the multi-scale operation feature vector to obtain multiple candidate sets of backward CAD operation nodes. The multiple candidate sets of backward CAD operation nodes are weighted and fused to obtain the updated set of backward operation nodes.

[0070] In this embodiment, when re-mining the backward operation chain of the operation chains of the preceding and current CAD operation nodes using the historical operation chain mining engine, a multi-scale time window is first used to extract features from the operation chains of the preceding and current CAD operation nodes. During this process, multiple time scales are set on the same operation chain, such as short time windows, medium time windows, and long time windows, each covering different operation ranges. For example, a short time window may include the most recent 3 operation nodes, a medium time window may include the most recent 10 operation nodes, and a long time window may include the most recent 30 operation nodes. Within each time scale, command features, object features, parameter features, time features, and user features are extracted from the included preceding and current CAD operation nodes. By encoding and combining the feature vectors extracted at each time scale, operation feature vectors at the short, medium, and long time scales are output, forming a multi-scale operation feature vector.

[0071] Next, backward operation chain mining is performed based on multi-scale operation feature vectors. In this process, the historical operation chain mining engine calculates the similarity between the operation feature vectors at each of the three scales and historical sequence samples in the historical operation chain sample library. Cosine similarity is used to calculate the similarity between the operation feature vectors and historical samples, and candidate historical chains with similarity greater than a preset threshold are selected. Then, within the candidate historical chain set corresponding to each time scale, the position of the current CAD operation node is located, and node sequences are extracted backward from the current position. The conditional probabilities of these node sequences appearing in the corresponding context are calculated. In this way, corresponding candidate sets of backward CAD operation nodes are obtained at short-term, medium-term, and long-term scales.

[0072] Finally, a weighted fusion of multiple candidate sets of backward CAD operation nodes is performed. In this process, firstly, the conditional probabilities of candidate nodes at each time scale are normalized, mapping scores at different scales to a range of 0 to 1 to ensure comparability. Then, weights are assigned according to the importance of the time scale; for example, shorter time windows are given higher weights to reflect immediacy, while medium and long time windows are given lower weights to reflect global trends. The scores of the same candidate node at different time scales are weighted and summed according to their corresponding weights to obtain the fusion score. If the same node appears simultaneously at multiple scales, its fusion scores are accumulated, and after fusion, the nodes are sorted and deduplicated, removing nodes with low confidence. For example, when a candidate node scores 0.9 in the short time window, 0.7 in the medium time window, and 0.5 in the long time window, with weights of 0.5, 0.3, and 0.2 respectively, its fusion score is 0.9 × 0.5 + 0.7 × 0.3 + 0.5 × 0.2 = 0.76.

[0073] Finally, the merged candidate nodes are sorted from high to low scores, and nodes that meet the preset threshold conditions are selected to obtain the backward operation node update set.

[0074] Step S500: The intermediate resource allocation node preloads the toolbar orchestration of the update set of the backward operation node.

[0075] In this embodiment, the intermediate resource allocation node reads the update set from the operation node and determines the order of operations based on the resource requirement indicators and fusion scores of each operation node, forming a sequence to be orchestrated. Next, the sequence to be orchestrated is compared with the resource indicators available for preloading, and an executable target set is selected within the limits of available preloading resources.

[0076] Then, the toolbar positions are arranged according to the order of the target set, and preloading actions are performed on the corresponding operation nodes, including preparing the data required for command loading, preparing the elements required for rendering, and establishing the call path, so as to reduce the loading and response latency when triggered later.

[0077] After completing the above steps, output the toolbar orchestration preload result that matches the current context, so that the toolbar is presented first and responds quickly to the operation node update set indication operation.

[0078] In summary, the embodiments of this application have at least the following technical effects:

[0079] This application monitors the user's current CAD operation node, performs backward operation chain mining on the operation feature vector of the current CAD operation node using a historical operation chain mining engine, and obtains a candidate set of backward CAD operation nodes. An intermediate resource allocation node is set, and resource indicators available for preloading in the resource pool are obtained according to the intermediate resource allocation node. It is determined whether the resource indicators available for preloading are less than the sum of resource requirements in the candidate set of backward CAD operation nodes. If they are less, the user's forward CAD operation node is traced according to an adaptive context window step size. The operation chains of the forward CAD operation node and the current CAD operation node are re-mined using the historical operation chain mining engine to obtain an updated set of backward operation nodes. The intermediate resource allocation node preloads the toolbar arrangement based on the updated set of backward operation nodes. This invention solves the technical problem in the prior art that the CAD toolbar cannot be dynamically adjusted according to user operation behavior. By mining and predicting the user's operation chain, adaptive toolbar arrangement and resource preloading are achieved, thereby improving the technical effect of CAD operation efficiency.

[0080] Example 2, based on the same inventive concept as the CAD toolbar dynamic arrangement method based on operation chain mining in the previous examples, such as... Figure 2 As shown, this application provides a dynamic CAD toolbar arrangement system based on operation chain mining. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0081] The monitoring module 11 monitors the user's current CAD operation node and performs backward operation chain mining on the operation feature vector of the current CAD operation node based on the historical operation chain mining engine to obtain a candidate set of backward CAD operation nodes. The resource indicator acquisition module 12 sets intermediate resource allocation nodes and acquires resource indicators that can be used for preloading in the resource pool according to the intermediate resource allocation nodes. The judgment module 13 judges whether the resource indicators that can be used for preloading are less than the sum of the resource demand indicators of the candidate set of backward CAD operation nodes. If they are less than the sum of the resource demand indicators of the candidate set of backward CAD operation nodes, the user's forward CAD operation node is traced according to the adaptive context window step size. The mining module 14 re-mines the backward operation chain on the operation chains of the forward CAD operation node and the current CAD operation node based on the historical operation chain mining engine to obtain a new set of backward operation nodes. The orchestration preloading module 15 is used by the intermediate resource allocation node to perform toolbar orchestration preloading on the new set of backward operation nodes.

[0082] Furthermore, the system is also used to implement the following functions:

[0083] The user's forward CAD operation nodes are traced according to an adaptive context window step size, which is dynamically determined adaptively based on the resource gap ratio. The resource gap ratio is the difference between the resource requirement index of the candidate set of backward CAD operation nodes and the resource index used for preloading.

[0084] Furthermore, the system is also used to implement the following functions:

[0085] The operation feature vector includes command features, object features, parameter features, time features, and user features. Based on the historical operation chain mining engine, historical sequence samples similar to the operation feature vector are calculated in the historical operation chain sample library to obtain a candidate historical chain set composed of feature similarity greater than a preset similarity threshold. Based on the candidate historical chain set, backward operation chain mining is performed to obtain a candidate set of backward CAD operation nodes.

[0086] Furthermore, the system is also used to implement the following functions:

[0087] Extract the node sequence following the position corresponding to the current CAD operation node from the candidate historical chain set; according to the calculated conditional probability of the node sequence appearing under the current CAD operation node, filter from the node sequence based on the conditional probability to obtain the backward CAD operation node candidate set.

[0088] Furthermore, the system is also used to implement the following functions:

[0089] Construct an operation chain graph, and identify the operation dependency between the current CAD operation node and the node sequence through the operation chain graph; re-update the node sequence and score it according to the conditional probability and the operation dependency to obtain a new candidate set of backward CAD operation nodes.

[0090] Furthermore, the system is also used to implement the following functions:

[0091] The command loading cost, rendering load resources, memory usage assessment, thread usage, and interface refresh latency of each operation node in the candidate set of backward CAD operation nodes are collected; the command loading cost, rendering load resources, memory usage assessment, thread usage, and interface refresh latency are comprehensively weighted and calculated to output the resource requirement index of each operation node; the resource requirement index of each operation node is summed to obtain the resource requirement index of the candidate set of backward CAD operation nodes.

[0092] Furthermore, the system is also used to implement the following functions:

[0093] A preset context window step size is used to monitor the set of CAD operation nodes operated by the user within the preset context window step size. Based on the historical operation chain mining engine, backward operation chain mining is performed on the set of operation feature vectors of the set of CAD operation nodes to obtain a candidate set of backward CAD operation nodes.

[0094] Furthermore, the system is also used to implement the following functions:

[0095] After obtaining the candidate set of backward CAD operation nodes, if the resource indicators available for preloading are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes, an adaptive solution is performed using the preset context window step size as the adaptive input variable and the resource gap as the adaptive target to obtain the adaptive context window step size.

[0096] Furthermore, the system is also used to implement the following functions:

[0097] A multi-scale time window is used to extract features from the operation chains of the forward CAD operation node and the current CAD operation node, and output a multi-scale operation feature vector. The backward operation chain is mined according to the multi-scale operation feature vector to obtain multiple candidate sets of backward CAD operation nodes. The multiple candidate sets of backward CAD operation nodes are weighted and fused to obtain the updated set of backward operation nodes.

[0098] It should be noted that the order of the embodiments described above is for descriptive purposes only and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0099] 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 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 method for dynamically arranging CAD toolbars based on operation chain mining, characterized in that, The method includes: Monitor the user's current CAD operation node, and perform backward operation chain mining on the operation feature vector of the current CAD operation node based on the historical operation chain mining engine to obtain a candidate set of backward CAD operation nodes; Set up intermediate resource allocation nodes, and obtain resource indicators from the resource pool that can be used for preloading according to the intermediate resource allocation nodes; Determine whether the resource indicators available for preloading are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes. If they are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes, trace the user's forward CAD operation node according to the adaptive context window step size. Based on the historical operation chain mining engine, the operation chains of the forward CAD operation node and the current CAD operation node are re-mined to obtain the updated set of backward operation nodes. The intermediate resource allocation node preloads the toolbar arrangement of the update set of the backward operation node; The user's forward CAD operation nodes are traced according to an adaptive context window step size, which is dynamically determined adaptively based on the resource gap ratio. The resource gap ratio is the difference between the resource requirement index of the candidate set of backward CAD operation nodes and the resource index used for preloading.

2. The method as described in claim 1, characterized in that, Based on the historical operation chain mining engine, backward operation chain mining is performed on the operation feature vector of the current CAD operation node to obtain a candidate set of backward CAD operation nodes. include: The operation feature vector includes command features, object features, parameter features, time features, and user features; Based on the historical operation chain mining engine, historical sequence samples similar to the operation feature vector are calculated in the historical operation chain sample library to obtain a candidate historical chain set composed of feature similarity greater than a preset similarity threshold. Based on the candidate historical chain set, backward operation chain mining is performed to obtain a candidate set of backward CAD operation nodes.

3. The method as described in claim 2, characterized in that, Based on the candidate historical chain set, backward operation chain mining is performed to obtain a candidate set of backward CAD operation nodes. The method includes: Extract the node sequence from the candidate historical chain set after the position corresponding to the current CAD operation node; Based on the calculated conditional probability of the node sequence appearing under the current CAD operation node, a candidate set of backward CAD operation nodes is obtained by filtering from the node sequence according to the conditional probability.

4. The method as described in claim 3, characterized in that, The method for obtaining the candidate set of backward CAD operation nodes also includes: Construct an operation chain graph, and use the operation chain graph to identify the operation dependency between the current CAD operation node and the node sequence; The node sequence is re-scored based on the conditional probability and the operation dependency to obtain a new candidate set of backward CAD operation nodes.

5. The method as described in claim 1, characterized in that, The method for calculating the resource requirement indices of the candidate set of backward CAD operation nodes includes: The command loading cost, rendering load resources, memory usage, thread usage, and interface refresh latency of each operation node in the candidate set of backward CAD operation nodes are collected. The resource requirement index for each operation node is output by comprehensively weighting the command loading cost, rendering load resources, memory usage assessment, thread usage, and interface refresh latency. The resource requirement indicators of each operation node are summed to obtain the resource requirement indicators of the candidate set of backward CAD operation nodes.

6. The method as described in claim 1, characterized in that, Methods for monitoring the user's current CAD operation node also include: A preset context window step size is used to monitor the set of CAD operation nodes operated by the user within the preset context window step size. Based on the historical operation chain mining engine, backward operation chain mining is performed on the operation feature vector set of the CAD operation node set to obtain a candidate set of backward CAD operation nodes.

7. The method as described in claim 6, characterized in that, After obtaining the candidate set of backward CAD operation nodes, if the resource indicators available for preloading are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes; Using the preset context window step size as the adaptive input variable and the resource gap as the adaptive target, an adaptive solution is performed to obtain the adaptive context window step size.

8. The method as described in claim 1, characterized in that, Based on the historical operation chain mining engine, the backward operation chain is re-mined for the operation chains of the forward CAD operation node and the current CAD operation node. The method includes: A multi-scale time window is used to extract features from the operation chains of the forward CAD operation node and the current CAD operation node, and outputs a multi-scale operation feature vector. Backward operation chain mining is performed based on the multi-scale operation feature vector to obtain multiple candidate sets of backward CAD operation nodes; The multiple candidate sets of backward CAD operation nodes are weighted and fused to obtain the updated set of backward operation nodes.

9. A CAD toolbar dynamic arrangement system based on operation chain mining, characterized in that, The system is used to execute the CAD toolbar dynamic arrangement method based on operation chain mining as described in any one of claims 1-8, the system comprising: The monitoring module is used to monitor the user's current CAD operation node and perform backward operation chain mining on the operation feature vector of the current CAD operation node based on the historical operation chain mining engine to obtain a candidate set of backward CAD operation nodes. The resource indicator acquisition module is used to set intermediate resource allocation nodes and acquire resource indicators in the resource pool that can be used for preloading according to the intermediate resource allocation nodes. The judgment module is used to determine whether the resource indicators available for preloading are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes. If they are less than the sum of the resource requirements of the candidate set of backward CAD operation nodes, the user's forward CAD operation node is traced according to the adaptive context window step size. The mining module is used to re-mint the backward operation chain of the operation chain of the forward CAD operation node and the current CAD operation node based on the historical operation chain mining engine, so as to obtain the backward operation node update set. The orchestration preloading module is used by the intermediate resource allocation node to perform toolbar orchestration preloading on the update set of the backward operation node.

Images