An environmental art design course optimization method and system

Abstract

The application provides an environmental art design course optimization method and system, relates to the field of intelligent education technology, and the method comprises the following steps: collecting new material characteristics, construction technology and design standard data flow of the environmental art design industry in real time, and constructing an industry dynamic database; screening lag knowledge points based on the industry dynamic database, replacing the industry leading technology content, and generating an updated course content library; matching actual construction project requirements based on the updated course content library; decomposing the project requirement task into executable subtasks, establishing a version mapping relationship between the subtasks and the leading technology content through a mirror algorithm, and generating a practice task set with a version identifier. The application realizes the synchronization of the course content and the industry development, improves the practicability of the course by updating the course content, matching the real project requirements, analyzing the student operation data, constructing the ability evaluation system and fusing the enterprise feedback.

Description

Technical Field

[0001] This invention relates to the field of smart education technology, and in particular to a method and system for optimizing environmental art design courses. Background Technology

[0002] With the rapid development of the environmental art design industry, new materials, advanced construction techniques, and design standards are constantly being updated. Traditional curriculum systems may suffer from outdated content and a disconnect between practice and industry needs. On the one hand, the pace of curriculum updates may not keep up with the industry's technological development, potentially leading to a mismatch between students' knowledge and actual job requirements. On the other hand, some practical teaching relies on simulated tasks and lacks integration with real engineering projects, potentially resulting in insufficient development of students' practical skills and comprehensive qualities. Furthermore, the evaluation methods for students' abilities may be too simplistic, with some focusing on theoretical assessments or final results evaluations while neglecting process performance and feedback from actual industry needs, making it difficult to form a closed-loop mechanism of teaching-practice-evaluation-optimization. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide a method and system for optimizing environmental art design courses. By updating course content, matching real project needs, analyzing student operation data, constructing a competency evaluation system, and integrating enterprise feedback, the course content is synchronized with industry development, thereby improving the practicality of the course.

[0004] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: Firstly, a method for optimizing an environmental art design curriculum, the method comprising: Step 1: Collect real-time data streams on the characteristics of new materials, construction techniques, and design standards in the environmental art design industry to build an industry dynamic database; based on the industry dynamic database, filter out outdated knowledge points and replace them with cutting-edge industry technology content to generate an updated course content library; Step 2: Based on the updated course content library, match the actual needs of the construction projects; decompose the project requirements into executable sub-tasks, and establish a version mapping relationship between the sub-tasks and the cutting-edge technology content through the mirroring algorithm to generate a set of practical tasks with version identifiers. Step 3: Based on the execution process data of the practice task set with version identifiers, deploy log tracking technology to collect student operation log streams in real time; Step 4: From the operation log stream, separate the construction decision node dataset using time-series feature extraction technology, analyze the team collaboration interaction dataset using natural language processing technology, and identify the innovative solution adoption dataset using solution identifier matching technology; based on the decision node dataset, interaction dataset, and adoption dataset, apply the entropy weight method to dynamically allocate weight coefficients and construct a capability weight matrix. Step 5: Generate a structured dynamic evaluation report by weighted fusion calculation of the capability weight matrix and the enterprise mentor's score; Step 6: Based on the dynamic evaluation report, perform resource matching calculations through the API interface of the enterprise cooperation platform to generate enterprise feedback data packets; parse the enterprise feedback data packets, extract the demand gap feature vector and equipment parameter set, and trigger the mirror incremental update operation of the industry dynamic database to re-execute the engineering practice task generation process.

[0005] Secondly, an environmental art design curriculum optimization system includes: The replacement module is used to collect data streams on the characteristics of new materials, construction techniques and design standards in the environmental art design industry in real time, and build an industry dynamic database; based on the industry dynamic database, outdated knowledge points are filtered out and replaced with cutting-edge industry technology content to generate an updated course content library; The mapping module is used to match the needs of actual construction projects with the updated course content library; it decomposes the project requirements into executable sub-tasks, and establishes a version mapping relationship between the sub-tasks and the cutting-edge technology content through a mirroring algorithm, generating a set of practical tasks with version identifiers. The data acquisition module is used to collect student operation log streams in real time based on the execution process data of the practice task set with version identifiers and deploy log tracking technology. The module is used to separate the construction decision node dataset from the operation log stream using time-series feature extraction technology, parse the team collaboration interaction dataset using natural language processing technology, and identify the innovative solution adoption dataset using solution identifier matching technology. Based on the decision node dataset, interaction dataset, and adoption dataset, the module applies the entropy weight method to dynamically allocate weight coefficients and construct a capability weight matrix. The fusion module is used to generate a structured dynamic evaluation report by weighted fusion calculation of the capability weight matrix and the enterprise mentor's score; The feedback and update module is used to perform resource matching calculations based on dynamic evaluation reports through the API interface of the enterprise cooperation platform, generate enterprise feedback data packages, parse enterprise feedback data packages, extract demand gap feature vectors and equipment parameter sets, and trigger mirror incremental update operations of the industry dynamic database to re-execute the engineering practice task generation process.

[0006] Thirdly, a computing device includes: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0007] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0008] The above-described solution of the present invention has at least the following beneficial effects: By collecting industry data in real time to build a dynamic database and updating the course content library, we ensure that course content keeps pace with industry trends, discards outdated knowledge, and improves the timeliness and practicality of teaching content. Matching actual engineering project needs, we decompose tasks and establish version mappings with cutting-edge technologies, generating version-identified practical task sets to enhance students' practical skills and narrow the gap between teaching and industry needs. We deploy log tracking technology to collect student operation log streams, providing complete data support for analyzing the learning process and performance. We extract multiple datasets from the log streams and construct a capability weight matrix using the entropy weight method, enabling multi-dimensional objective evaluation of students' decision-making, collaboration, and innovation abilities, providing quantitative basis for comprehensive assessment. We integrate the capability weight matrix with enterprise mentor scores to generate structured dynamic evaluation reports, combining quantitative analysis and professional experience to make evaluations more comprehensive and accurate, assisting students in improving and teachers in personalized teaching. Enterprise feedback triggers incremental database updates, regenerating practical tasks and forming a dynamic iterative closed loop to quickly respond to enterprise needs and industry changes. Attached Figure Description

[0009] Figure 1 This is a flowchart illustrating an environmental art design course optimization method provided by an embodiment of the present invention.

[0010] Figure 2 This is a schematic diagram of an environmental art design course optimization system provided by an embodiment of the present invention. Detailed Implementation

[0011] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0012] like Figure 1 As shown in the figure, an embodiment of the present invention proposes a method for optimizing an environmental art design course, the method comprising the following steps: Step 1: Collect real-time data streams on the characteristics of new materials, construction techniques, and design standards in the environmental art design industry to build an industry dynamic database; based on the industry dynamic database, filter out outdated knowledge points and replace them with cutting-edge industry technology content to generate an updated course content library; Step 2: Based on the updated course content library, match the actual needs of the construction projects; decompose the project requirements into executable sub-tasks, and establish a version mapping relationship between the sub-tasks and the cutting-edge technology content through the mirroring algorithm to generate a set of practical tasks with version identifiers. Step 3: Based on the execution process data of the practice task set with version identifiers, deploy log tracking technology to collect student operation log streams in real time; Step 4: From the operation log stream, separate the construction decision node dataset using time-series feature extraction technology, analyze the team collaboration interaction dataset using natural language processing technology, and identify the innovative solution adoption dataset using solution identifier matching technology; based on the decision node dataset, interaction dataset, and adoption dataset, apply the entropy weight method to dynamically allocate weight coefficients and construct a capability weight matrix. Step 5: Generate a structured dynamic evaluation report by weighted fusion calculation of the capability weight matrix and the enterprise mentor's score; Step 6: Based on the dynamic evaluation report, perform resource matching calculations through the API interface of the enterprise cooperation platform to generate enterprise feedback data packets; parse the enterprise feedback data packets, extract the demand gap feature vector and equipment parameter set, and trigger the mirror incremental update operation of the industry dynamic database to re-execute the engineering practice task generation process.

[0013] In this embodiment of the invention, real-time industry data is collected to construct a dynamic database and update the course content library, ensuring that the course content keeps pace with the latest industry developments, discards outdated knowledge, and improves the timeliness and practicality of the content to meet the industry's demand for talent knowledge reserves. Matching actual project needs, tasks are decomposed and then a version mapping with cutting-edge technologies is established through a mirroring algorithm, generating a set of practical tasks with version identifiers. This allows students to participate in real project-related tasks, improving their practical skills and narrowing the gap between teaching and industry needs. Log tracking technology is deployed to collect student operation log streams, providing complete raw data for analyzing students' learning processes and abilities, objectively reflecting their learning trajectories, and avoiding the limitations of evaluating solely based on results. Multiple datasets are extracted from the log streams, and an entropy weighting method is used to construct a capability weight matrix, enabling a multi-dimensional and objective evaluation of students' abilities and comprehensively reflecting their overall capabilities. The capability weight matrix is ​​integrated with enterprise mentor ratings to generate a structured dynamic evaluation report. Combining quantitative analysis and professional experience, the evaluation is more comprehensive and accurate, helping students improve and teachers provide personalized teaching. Feedback is obtained through enterprise cooperation platforms, triggering incremental updates to the database, forming a dynamic iterative closed loop, enabling the course to quickly respond to changes in enterprises and the industry, and strengthening school-enterprise collaboration.

[0014] In a preferred embodiment of the present invention, step 1 above involves real-time data collection of new material characteristics, construction techniques, and design standards in the environmental art design industry to construct an industry dynamic database; based on the industry dynamic database, outdated knowledge points are filtered out and replaced with cutting-edge industry technology content to generate an updated course content library, including: Step 11: Collect data on the characteristics of new materials and construction processes in the environmental art design industry in real time through the Internet of Things sensor group, and simultaneously call the industry data interface to obtain the design standard data stream; Step 12: Based on the characteristics of new materials, construction process data, and design standard data stream, construct an industry dynamic database using distributed storage technology; Step 13: Based on the industry dynamic database, perform semantic feature analysis on the historical course content, and output the set of lagging knowledge point identifiers through similarity matching calculation; Step 14: Based on the lagging knowledge point identifier set, call the cutting-edge technology data flow of the industry dynamic database to perform knowledge graph reconstruction operation and generate an updated course content library.

[0015] In this embodiment of the invention, data is collected through both IoT sensors and industry interfaces to ensure the real-time nature and comprehensiveness of the data, providing high-quality raw data for subsequent database construction. Distributed storage technology improves data storage efficiency and access speed, meeting the dynamic management needs of large amounts of industry data and ensuring the stability of data retrieval. Through semantic feature analysis and similarity matching, outdated knowledge points are accurately identified, providing an objective basis for updating course content. Knowledge graph reconstruction makes the logical structure of course content clearer, seamlessly connecting cutting-edge technologies with existing knowledge, helping students build a complete knowledge system and improving learning efficiency.

[0016] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step 11 above, an IoT sensor group is deployed to collect real-time data on the physical properties of new materials (such as compressive strength and environmental protection coefficient) and the parameters of construction processes (such as construction accuracy and process time), and to determine the collection frequency and data format.

[0017] Connect to the data interfaces of official or authoritative industry platforms, and obtain data streams related to design standards based on preset filtering conditions (such as release time, applicable scenarios, etc.). Synchronize the timestamps with the data collected by sensors to ensure the consistency of the two types of data in the time dimension.

[0018] In step 12 above, the collected data on the characteristics of new materials, construction processes, and design standards are classified and stored according to data type (such as text, numerical data, and image data).

[0019] By employing distributed storage technology, different categories of data are allocated to multiple storage nodes based on their access frequency and correlation, establishing mapping relationships between nodes to achieve distributed data management and rapid retrieval.

[0020] Step 13 above involves extracting keywords (such as material names, process names, design specification clauses, etc.) from the historical course content and constructing semantic feature vectors; performing the same semantic feature extraction on the cutting-edge data in the industry dynamic database to generate comparative feature vectors; calculating the similarity between the historical course knowledge point feature vectors and the cutting-edge data feature vectors, setting a similarity threshold, marking knowledge points below the threshold as lagging knowledge points, and summarizing them to form a set of lagging knowledge point identifiers.

[0021] In step 14 above, based on the lagging knowledge point identifier set, the corresponding cutting-edge technology data streams are selected from the industry dynamic database to extract core concepts and relationships.

[0022] The course content is reconstructed in the form of a knowledge graph, establishing connections (such as causal relationships, subordinate relationships, etc.) between cutting-edge technology content and existing non-lagging knowledge points, supplementing the logical connections between knowledge nodes, and generating an updated course content library.

[0023] In a preferred embodiment of the present invention, step 2 above involves matching the updated course content library with the actual needs of ongoing construction projects; decomposing the project requirements into executable sub-tasks; and establishing a version mapping relationship between the sub-tasks and cutting-edge technology content through a mirroring algorithm to generate a set of practical tasks with version identifiers, including: Step 21: Based on the updated course content library, extract the course content feature vector; obtain the BIM parameter set of the actual construction project and parse it into the engineering requirement feature vector; calculate the cosine similarity between the course content feature vector and the engineering requirement feature vector using a collaborative filtering algorithm; filter matching items according to the similarity threshold and generate a requirement-content mapping relationship table. Step 22: Based on the requirement-content mapping table, the project requirements are decomposed into a sequence of executable sub-task units using a graph segmentation algorithm; Step 23: Based on the subtask unit sequence, parse the technical requirement description field of each subtask unit and extract the technical requirement identifier; use the technical requirement identifier as the search key to query the version hash value of the corresponding cutting-edge technology content in the course content library; establish a key-value pair mapping between each subtask unit ID and its corresponding version hash value to obtain the ID-hash value mapping table. Step 24: Based on the ID-hash value mapping table, encapsulate the mapping table and embed a timestamp verification code to obtain a version mapping relationship table with verification. Step 25: Integrate the version mapping relationship table with verification and the engineering constraint parameters, including the schedule threshold, cost boundary and specification compliance indicators, to generate a set of practice tasks with version identifiers.

[0024] In this embodiment of the invention, feature vector matching and similarity calculation are used to achieve precise alignment between course content and actual project requirements, avoiding a disconnect between tasks and teaching content; graph segmentation algorithm ensures the rationality of subtask decomposition, transforming complex project requirements into executable unit tasks and reducing the difficulty of practice; ID-hash value mapping clarifies the association between subtasks and technology versions, ensuring that the technologies applied by students in practice strictly correspond to the cutting-edge content of the course; timestamp verification codes enhance the security and reliability of version mapping relationships, preventing data errors from affecting the effectiveness of practical tasks; and the integration of engineering constraint parameters makes practical tasks closer to real project scenarios, cultivating students' ability to solve problems under constraints.

[0025] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step 21 above, each technical content item (such as "new environmentally friendly stone paving technology" and "intelligent lighting system design specifications") is broken down from the updated course content library, and core attributes (including technical category, applicable scenarios, operation difficulty, related materials, etc.) are extracted and transformed into quantifiable feature vectors. For example, quantitative labels such as "material type = environmentally friendly", "operation precision = high" and "related process = sealing treatment" are assigned to "installation of environmentally friendly materials" to form structured feature data.

[0026] Obtain the BIM (Building Information Modeling) parameter set of the actual construction project. This parameter set contains the project's three-dimensional spatial data, structural parameters (such as the location of load-bearing walls and floor load-bearing limits), functional zoning (such as commercial spaces and office areas), and design standards (such as fire resistance ratings and energy-saving indicators). By analyzing the BIM parameters, extract the project's specific technical requirements (such as "the atrium area must use environmentally friendly building materials with a light transmittance of ≥80%" and "the waterproofing construction of the bathroom must comply with the GB50108-2008 standard"), and transform these requirements into engineering requirement feature vectors that are consistent with the feature vector format of the course content.

[0027] The collaborative filtering algorithm is used to compare the feature vectors of course content with the feature vectors of project requirements, and the cosine similarity between the two is calculated. The closer the similarity value is to 1, the higher the matching degree between the course content and the project requirements. For example, if the feature vector of "waterproof membrane construction technology" in the course has a similarity of 0.92 with the requirement vector of "underground garage waterproofing project", it means that the technical content is highly matched with the project requirements.

[0028] Set a similarity threshold (e.g., 0.7), filter out course content and project requirements combinations that exceed the threshold, and eliminate mismatched content (e.g., exclude "ancient building restoration techniques" from the requirements of modern commercial complex projects). Finally, form a mapping relationship table of "project requirements - course content" to clarify the specific technical teaching content corresponding to each project requirement.

[0029] Step 22 above, based on the requirement-content mapping relationship table, treats project requirements as a network composed of multiple related nodes - each node represents a specific requirement (such as "interior wall decoration" or "circuit wiring layout"), the connection between nodes represents the correlation between requirements (such as "wall decoration" must be carried out after "wiring layout" is completed), and each requirement node is bound to the corresponding course content node (such as "wall decoration" is associated with the course content "diatom mud construction process").

[0030] The overall network is split using a graph segmentation algorithm, based on the following criteria: task relevance (separating strongly related requirements such as "water and electricity renovation" and "wall grooving" into the same sub-task unit); balanced complexity (avoiding sub-task units that are too large or too small, such as splitting "whole-house intelligent system deployment" into sub-units like "sensor installation" and "control system debugging"); and completeness of course content coverage (ensuring that each sub-task unit corresponds to at least one course technical content). This ultimately forms an ordered sequence of sub-task units, such as "preliminary survey → pipeline layout design → environmentally friendly material procurement → basic construction → acceptance and commissioning," where each sub-unit can be executed independently and covers clearly defined teaching objectives.

[0031] Step 23 above involves analyzing each subtask in the subtask unit sequence (e.g., "bathroom waterproofing construction") one by one, extracting core keywords from its technical requirement description (e.g., "reactive waterproofing coating must be used, coating thickness ≥ 1.5 mm, curing time ≤ 24 hours") to form technical requirement identifiers; these identifiers must accurately point to the technical points in the course content library, such as "reactive waterproofing coating construction" and "curing time control".

[0032] Using the technical requirement identifier as the search keyword, the corresponding cutting-edge technical content is searched in the course content library, and the version hash value of the content is obtained. This is a unique character code used to identify the version of the technical content (such as the hash value "a7b3c9..." corresponding to the "V2.3" version of "waterproof coating construction process"). This ensures that the technical content associated with the subtask is the latest version (avoiding the use of the obsolete "V1.0" manual brushing process).

[0033] Assign a unique ID to each subtask unit (e.g., “ST-001”, “ST-002”), establish key-value pairs between the subtask ID and the corresponding technology version hash value (e.g., “ST-001:a7b3c9...”), and summarize them to form a mapping table, clearly recording the correspondence between each subtask and the cutting-edge technology version, providing a basis for subsequent tracking of technology applications.

[0034] Step 24 above formats the ID-hash value mapping table, encapsulates it using a unified data structure (such as JSON format), and clarifies fields such as subtask ID, technical requirement identifier, version hash value, and associated course content to ensure the standardization of data during transmission and storage.

[0035] Add a timestamp to the encapsulated mapping table (recording the precise time the data was generated, such as "2024-05-2014:30:22"), and generate a checksum (such as an MD5 checksum) based on the timestamp and the mapping table content. When the mapping table is accessed or transmitted, the checksum comparison can quickly determine whether the data has been tampered with (such as the hash value being maliciously modified to an old version encoding). Finally, a version mapping relationship table with verification is formed to ensure the security and accuracy of the data.

[0036] Step 25 above involves collecting the actual constraints of the project, including: time thresholds (e.g., "Subtask ST-003 must be completed within 3 working days"); cost boundaries (e.g., "The material cost of subtask ST-005 shall not exceed 5,000 yuan"); and compliance indicators (e.g., "Must comply with the 'Standard for Acceptance of Construction Decoration and Renovation Engineering Quality' GB50210-2018"). These parameters need to be quantified and determined in conjunction with the actual situation of the project to ensure the enforceability of the constraints.

[0037] The version mapping table with verification is integrated with the engineering constraint parameters, and constraint descriptions are added to each sub-task unit (such as "ST-002: Installation of environmentally friendly board material, version hash value a7b3c9..., cost ≤ 3000 yuan, must meet the LEEDv4 green building evaluation standard"). Finally, a complete set of practical tasks is generated, which includes sub-task ID, technology version identifier, task content, and constraint conditions. This set of tasks ensures that the technology applied by students is in sync with the forefront of the industry, and simulates the constraint environment of real projects, thereby improving the authenticity of practical teaching.

[0038] In a preferred embodiment of the present invention, step 3 above, based on the execution process data of the practice task set with version identification, deploys log tracking technology to collect student operation log streams in real time, including: Step 31: Based on the practice task set with version identifier, extract the task set version metadata and generate a version synchronization event tracking configuration scheme. Step 32: Based on the data entry configuration scheme, deploy a log collection agent on the task execution node; Step 33: Capture student operation events in real time through the log collection agent and encapsulate the operation log stream according to the version identifier; Step 34: Verify the consistency of the version identifier of the operation log stream based on the task set version metadata, and output the operation log stream that has passed the verification.

[0039] In this embodiment of the invention, synchronized tracking configurations are generated through version metadata to ensure accurate matching between log collection targets and task versions, laying the foundation for subsequent data tracing. Collection agents are deployed at key nodes to accurately capture operational behaviors, avoiding the omission of important operational data and reducing the cost of collecting irrelevant data. Log streams are encapsulated by version identifiers, making each operation record traceable to the corresponding task version, facilitating subsequent analysis of the execution status of tasks in different versions. Version consistency verification ensures the validity and accuracy of log data, avoiding analytical biases caused by version inconsistencies and providing reliable data support for student competency assessment.

[0040] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: Step 31 above extracts metadata such as version number, task unit ID, and technology association hash value from the practice task set with version identifier to clarify the version characteristics of each task unit; and designs the tracking rules based on the metadata information, including the type of operation to be collected (such as task submission, parameter modification, scheme upload, etc.), collection frequency (real-time triggering or timed capture), and data field format (including version identifier, timestamp, operator ID, etc.) to form a tracking configuration scheme synchronized with the task version.

[0041] In step 32 above, based on the data collection configuration scheme, a log collection agent is deployed at key nodes in the execution of the practical task (such as the task startup interface, operation submission button, scheme upload entry, etc.); the agent is configured to associate with the interface of the task system to ensure that the agent can recognize the task execution status and start collection only when students perform task-related operations to avoid interference from invalid data.

[0042] In step 33 above, when students perform practical tasks (such as modifying design parameters, submitting construction plans, marking task progress, etc.), the log collection agent captures the details of the operation behavior in real time and records information such as operation type, execution time, and operation result; according to the format set in step 31, each operation record is bound and encapsulated with the corresponding task version identifier to form a structured operation log stream containing version information, ensuring that the log corresponds one-to-one with the task version.

[0043] In step 34 above, the version identifier in the operation log stream is extracted and compared with the obtained task set version metadata to verify the consistency between the two (such as whether the version number matches and whether the hash value corresponds); the log streams that pass the verification are filtered out, and records with version mismatch or abnormal data are removed to ensure that the output operation log stream is completely synchronized with the version of the currently executed practice task.

[0044] In a preferred embodiment of the present invention, step 4 above involves separating the construction decision node dataset from the operation log stream using time-series feature extraction technology, parsing the team collaboration interaction dataset using natural language processing technology, and identifying the innovative solution adoption dataset using solution identifier matching technology. Based on the decision node dataset, interaction dataset, and adoption dataset, a capability weight matrix is ​​constructed by dynamically allocating weight coefficients using the entropy weight method, including: Step 41: Based on the operation log stream, perform time-series feature extraction, identify key decision timestamp sequences, and segment the log stream according to a preset time window to generate a log segment set; extract the construction decision node dataset from the log segment set; Step 42: Based on the operation log stream, perform natural language processing to identify and separate team collaboration communication text segments to obtain a set of communication text segments; based on the set of communication text segments, parse the task assignment instructions and conflict resolution records in the interaction text to generate a team collaboration interaction dataset. Step 43: Perform a scheme identifier scan based on the operation log stream to extract the design scheme submission event and its metadata; extract the feature identifier of the scheme to be reviewed from the metadata; compare the feature identifier of the scheme to be reviewed with the feature code of the historical adopted scheme library to obtain the scheme adoption status set; and fuse the scheme metadata and adoption status based on the scheme adoption status set to generate an innovative scheme adoption dataset. Step 44: Construct a three-dimensional evaluation index matrix based on the decision node dataset, interaction dataset, and adoption dataset, where the X-axis represents the decision response timeliness index, the Y-axis represents the collaborative task completion rate index, and the Z-axis represents the innovative solution adoption rate index; calculate the weight coefficients of each index using the entropy weight method; and generate a capability weight matrix by embedding the weight coefficients.

[0045] In this embodiment of the invention, construction decision-making node data is accurately separated by extracting time-series features, objectively reflecting students' decision-making logic and timeliness, and providing targeted data for assessing decision-making ability; natural language processing technology enables structured parsing of team collaboration data, avoiding the subjectivity of manual analysis and comprehensively presenting collaboration efficiency and communication effectiveness; scheme identification and matching technology quantifies the adoption of innovative schemes, clearly reflecting students' innovation ability and the practicality of schemes, and providing an objective basis for innovation evaluation; the entropy weight method dynamically allocates weights to make ability evaluation more in line with the actual data distribution, and the three-dimensional matrix intuitively presents the ability structure, providing a quantitative basis for comprehensively evaluating students' abilities.

[0046] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step 41 above, the operation log stream is sorted in chronological order, the timestamp information of each log is extracted, and the time points marked as key construction decision behaviors such as "decision submission" and "scheme confirmation" are identified to form a key decision timestamp sequence.

[0047] Set a fixed time window (e.g., 30 minutes) or a dynamic time window (adjusted according to the interval of task nodes), and divide the log stream based on the key decision timestamp to obtain a set of log segments corresponding to the decision nodes.

[0048] Information such as decision type, decision parameters, and execution results are extracted from each log segment and summarized to form a construction decision node dataset, ensuring that the data accurately corresponds to the decision time node.

[0049] In step 42 above, records containing text information (such as team chat logs, task messages, collaborative annotations, etc.) are filtered from the operation log stream. Using a text classification model in natural language processing technology, communication text segments related to team collaboration are identified and separated to form a set of communication text segments.

[0050] Semantic analysis is performed on the communication text segments to extract task allocation instructions containing keywords such as "assign", "responsible", and "transfer", as well as conflict resolution records containing keywords such as "resolve", "negotiate", and "reach an agreement".

[0051] The extracted instructions and records are categorized and integrated according to team number and task unit to generate a structured team collaboration interaction dataset, which includes key information such as collaboration objects, content, and time.

[0052] In step 43 above, the operation log stream is scanned to identify events marked as "Solution Submission" and metadata such as the solution name, submission time, submitter, and solution content summary corresponding to the event is extracted.

[0053] Extract core features (such as design concept, technical method, material selection, etc.) from the metadata of the proposal to generate feature identifiers for the proposal to be reviewed; compare the feature identifiers of the proposal to be reviewed with the feature codes in the historical adopted proposal database, calculate the matching degree between the two, and determine whether the proposal is adopted based on the matching degree threshold, forming a set of proposal adoption statuses (such as "adopted", "partially adopted", "not adopted").

[0054] By associating and integrating the metadata of the proposed solutions with their adoption status, and arranging them in the order of submission, an innovative solution adoption dataset is generated.

[0055] Step 44 above establishes a three-dimensional evaluation index matrix based on the three datasets: the construction decision node dataset, the team collaboration interaction dataset, and the innovative solution adoption dataset.

[0056] Extract the response time of each decision node from the construction decision node dataset, as well as the required completion time for that decision task. Calculate the ratio of the response time to the required task time, and use this ratio as the indicator data on the X-axis to measure whether the student's response speed in the decision-making process meets the task expectations.

[0057] In the team collaboration interaction dataset, the number of completed collaboration tasks and the total number of collaboration tasks are counted. The ratio of completed tasks to total tasks is calculated and used as the indicator data on the Y-axis to reflect the overall situation of student teams in completing tasks during the collaboration process.

[0058] From the dataset of adopted innovative solutions, we count the number of adopted innovative solutions and the total number of submitted innovative solutions, calculate the ratio of adopted solutions to total submitted solutions, and use this ratio as the indicator data on the Z-axis to reflect the degree to which the innovative solutions proposed by students are recognized.

[0059] The entropy weight method was applied to analyze the constructed three-dimensional evaluation index matrix. The core of the entropy weight method is to determine the weights based on the dispersion of the index data. Specifically, the greater the difference in the value of an index among different samples, i.e., the higher the dispersion, the more effectively the index can distinguish the abilities of different students, and therefore a higher weight coefficient is assigned to it. By analyzing the data distribution characteristics of each of the three-dimensional indicators, such as the data dispersion range and fluctuation, the weight coefficients corresponding to the three indicators—decision response timeliness, collaborative task completion rate, and innovative solution adoption rate—were dynamically calculated.

[0060] The calculated weight coefficients are then incorporated into the previously constructed three-dimensional evaluation index matrix, so that each index value in the matrix corresponds to its corresponding weight. This generates an ability weight matrix containing index values ​​and corresponding weights, which can clearly reflect the specific proportion of students in the three aspects of decision-making ability, collaboration ability, and innovation ability through quantitative means.

[0061] In a preferred embodiment of the present invention, step 5 above, which generates a structured dynamic evaluation report by weighted fusion calculation of the capability weight matrix and the enterprise mentor's score, includes: Step 51: Call the rating service of the enterprise cooperation platform through the API interface to obtain the enterprise mentor rating value in real time; Step 52: Based on the capability weight matrix and the enterprise mentor rating, perform a timestamp synchronization operation to generate an aligned data pair set; Step 53: Based on the aligned data pair set, extract the three-dimensional indicator weight coefficients in the capability weight matrix, calculate the normalized score of the enterprise mentor's score, and generate a comprehensive score vector according to the preset fusion ratio. Step 54: Based on the comprehensive scoring vector, embed dynamic timestamps and version identifiers to generate a machine-readable structured dynamic evaluation report.

[0062] In this embodiment of the invention, enterprise mentor ratings are obtained in real time through an API interface, ensuring the timeliness and authority of the rating data and incorporating industry-leading evaluations into the student competency assessment system. Timestamp synchronization ensures the consistency between competency data and mentor ratings, avoiding fusion bias caused by time discrepancies and improving the accuracy of data matching. Weighted fusion combines objective data indicators with subjective professional evaluations, and normalization eliminates differences in rating scales. The comprehensive rating vector more fully reflects student competency, and the fusion ratio can be flexibly adjusted to adapt to different evaluation emphases. The structured report facilitates automated machine processing and subsequent data analysis, while dynamic timestamps and version identifiers ensure report traceability, providing a standardized and clear reference for teaching improvement and enterprise talent selection.

[0063] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: In step 51 above, the system calls the pre-defined API interface of the enterprise cooperation platform to send a scoring request to the platform. The request includes the student identifier, the corresponding practice task number, and the task version information for which a score is required.

[0064] After receiving the request, the enterprise collaboration platform queries the student's enterprise mentor rating record in the corresponding task and returns the rating value (such as a percentage score, quantitative score corresponding to the grade assessment, etc.) through the interface. The local system receives and stores these rating values ​​in real time.

[0065] In step 52 above, the timestamp when the capability weight matrix was generated and the scoring timestamp in the enterprise mentor scoring value record are extracted.

[0066] If there is a time discrepancy between the two timestamps, time calibration is performed according to the task execution cycle and scoring rules (such as aligning with the task submission time) to ensure that the capability weight matrix and the corporate mentor's score correspond to the same practical task stage in the time dimension.

[0067] The calibrated capability weight matrix is ​​paired with the corresponding corporate mentor ratings to form a set of one-to-one aligned data pairs.

[0068] In step 53 above, the three-dimensional indicator weight coefficients of the X-axis (decision response time), Y-axis (collaborative task completion rate), and Z-axis (innovative solution adoption rate) are extracted from the capability weight matrix in the aligned data pair set.

[0069] Collect enterprise mentor ratings and normalize them according to the rating range (e.g., 0-100 points), converting the ratings into standardized scores between 0 and 1 to eliminate the influence of different rating scales.

[0070] According to the preset integration ratio (e.g., the capability weight matrix accounts for 60% and the corporate mentor score accounts for 40%), the quantitative results of the three-dimensional indicators are weighted and calculated with the normalized mentor score, and the comprehensive score vector is obtained. The vector contains the comprehensive score of each dimension and the total score.

[0071] In step 54 above, add a dynamic timestamp (to record the report generation time) and the corresponding task version identifier to the comprehensive scoring vector (to ensure consistency with the practice task version).

[0072] The comprehensive scoring vector, timestamp, version identifier and other information are encapsulated according to a preset structured format (such as JSON, XML), the meaning and data type of each field are clarified, and a structured dynamic evaluation report that can be directly read and parsed by the machine is generated.

[0073] In a preferred embodiment of the present invention, step 6 above, based on the dynamic evaluation report, performs resource matching calculations through the API interface of the enterprise cooperation platform to generate an enterprise feedback data packet; parses the enterprise feedback data packet, extracts the demand gap feature vector and equipment parameter set, and triggers the mirror incremental update operation of the industry dynamic database, re-executes the engineering practice task generation process, including: Step 61: Based on the dynamic evaluation report, perform resource matching calculations through the API interface of the enterprise cooperation platform to generate enterprise feedback data packets; Step 62: Based on the enterprise feedback data packet, parse the data packet structure and extract the demand gap feature vector and equipment parameter set; Step 63: Based on the demand gap feature vector and equipment parameter set, trigger the mirror incremental update operation of the industry dynamic database to generate the updated database with version identifier; Step 64: Based on the updated database with the version identifier, re-execute the engineering practice task generation process.

[0074] In this embodiment of the invention, real-time data interaction with the enterprise cooperation platform is achieved through an API interface, enabling rapid transmission of enterprise feedback and providing direct industry basis for course optimization, thus enhancing the timeliness of school-enterprise collaboration. Accurate extraction of demand gaps and equipment parameter information clarifies the direction and focus of course optimization, providing specific targets for subsequent database updates and task generation, avoiding blind adjustments. Mirror incremental update operations reduce the workload of database updates, improve update efficiency, and the version identifier facilitates tracking of database changes, ensuring data manageability and enabling the industry dynamic database to reflect changes in enterprise needs in a timely manner. Based on the updated database, a new set of practical tasks is generated, ensuring that practical tasks always remain consistent with actual enterprise needs and the latest industry conditions, forming a closed loop of "evaluation-feedback-update-practice," continuously improving the practicality of the course and students' job adaptability.

[0075] In this embodiment of the invention, when applied in a specific way, it can be implemented through the following technical solutions, for example: Step 61 above extracts key information from the dynamic evaluation report, including the student's overall score, scores for each ability dimension, and the corresponding version of the practical task.

[0076] Through the API interface of the enterprise cooperation platform, this information is sent to the platform's resource matching mechanism to initiate a resource matching calculation request. The enterprise cooperation platform performs matching calculations on the student's ability and the enterprise's resource needs according to the preset matching rules (such as the correspondence between enterprise job requirements and student abilities, the relevance between task requirements and enterprise actual business, etc.), generates an enterprise feedback data package containing information such as the enterprise's evaluation of the student's ability, the demand gap, and the required equipment parameters, and returns it to the local end through the interface.

[0077] In step 62 above, after receiving the data packet from the enterprise, the structure of the data packet is parsed and the field identifiers (such as "demand gap" and "equipment parameters") are identified.

[0078] Extract the demand gap feature vector from the data packet. This vector contains the specific dimensions and degree of the gap that enterprises believe students lack in terms of knowledge and skills. At the same time, extract the equipment parameter set, including detailed parameters such as the equipment models, performance indicators, and operating requirements actually used by the enterprises. Perform format conversion and validation on the extracted demand gap feature vector and equipment parameter set.

[0079] Step 63 above involves breaking down the extracted demand gap feature vector into specific knowledge and skill dimensions, such as material knowledge, construction process skills, and mastery of design specifications. These dimensions are then compared one by one with the existing knowledge items and technical content in the industry dynamic database to see if the corresponding content in the database can meet the needs proposed by the enterprise. For the parts with gaps, it is determined whether it is necessary to supplement new knowledge points and technical content or to update and upgrade existing content, thereby determining the specific list of supplements and updates.

[0080] For the extracted equipment parameter set, firstly, sort out the information entries related to these devices in the database, covering the equipment's operating specifications, applicable engineering scenarios, and corresponding construction process steps; then, modify and improve these information entries based on the latest data in the equipment parameter set; for example, if the equipment parameter set mentions new changes in the operating procedures of a certain device, update the operating specifications of that device in the database; if the applicable scenarios of the device have expanded, supplement the corresponding applicable scenario descriptions to ensure the accuracy and timeliness of the equipment-related information in the database.

[0081] Based on the previously determined supplementary and updated list, a mirrored incremental update operation is triggered in the industry dynamic database. This operation will not modify the content in the database that is not involved in the update, but only process the parts that need to be supplemented and updated. During the update process, new storage entries will be created for newly added knowledge points and technical content, and the content that needs to be updated will be directly modified on the existing entries, and the modification records will be retained.

[0082] After the update is completed, a new version identifier is generated for the industry dynamics database. This version identifier includes information such as the update time and a summary of the update content. By comparing it with the previous version identifier, the state of the database before and after the update can be clearly distinguished, which facilitates the subsequent tracking of database changes and provides a clear version basis for other processes based on the database.

[0083] Step 64 above restarts the engineering practice task generation process based on the updated database with version identifiers. Following the previous task generation logic, the updated course content library (generated from the updated industry dynamic database) is matched with the actual needs of the engineering projects under construction. Tasks are decomposed, mapping relationships are established, and a new set of practice tasks with version identifiers is generated to ensure that the practice tasks reflect the latest industry needs and equipment requirements.

[0084] like Figure 2 As shown, embodiments of the present invention also provide an environmental art design course optimization system, comprising: The replacement module is used to collect data streams on the characteristics of new materials, construction techniques and design standards in the environmental art design industry in real time, and build an industry dynamic database; based on the industry dynamic database, outdated knowledge points are filtered out and replaced with cutting-edge industry technology content to generate an updated course content library; The mapping module is used to match the needs of actual construction projects with the updated course content library; it decomposes the project requirements into executable sub-tasks, and establishes a version mapping relationship between the sub-tasks and the cutting-edge technology content through a mirroring algorithm, generating a set of practical tasks with version identifiers. The data acquisition module is used to collect student operation log streams in real time based on the execution process data of the practice task set with version identifiers and deploy log tracking technology. The module is used to separate the construction decision node dataset from the operation log stream using time-series feature extraction technology, parse the team collaboration interaction dataset using natural language processing technology, and identify the innovative solution adoption dataset using solution identifier matching technology. Based on the decision node dataset, interaction dataset, and adoption dataset, the module applies the entropy weight method to dynamically allocate weight coefficients and construct a capability weight matrix. The fusion module is used to generate a structured dynamic evaluation report by weighted fusion calculation of the capability weight matrix and the enterprise mentor's score; The feedback and update module is used to perform resource matching calculations based on dynamic evaluation reports through the API interface of the enterprise cooperation platform, generate enterprise feedback data packages, parse enterprise feedback data packages, extract demand gap feature vectors and equipment parameter sets, and trigger mirror incremental update operations of the industry dynamic database to re-execute the engineering practice task generation process.

[0085] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0086] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing an environmental art design curriculum, characterized in that, The method includes: Step 1: Collect real-time data streams on the characteristics of new materials, construction techniques, and design standards in the environmental art design industry to build an industry dynamic database; based on the industry dynamic database, filter out outdated knowledge points and replace them with cutting-edge industry technology content to generate an updated course content library; Step 2: Based on the updated course content library, match the actual needs of the construction projects; decompose the project requirements into executable sub-tasks, and establish a version mapping relationship between the sub-tasks and the cutting-edge technology content through the mirroring algorithm to generate a set of practical tasks with version identifiers. Step 3: Based on the execution process data of the practice task set with version identifiers, deploy log tracking technology to collect student operation log streams in real time; Step 4: From the operation log stream, separate the construction decision node dataset using time-series feature extraction technology, analyze the team collaboration interaction dataset using natural language processing technology, and identify the innovative solution adoption dataset using solution identifier matching technology; based on the decision node dataset, interaction dataset, and adoption dataset, apply the entropy weight method to dynamically allocate weight coefficients and construct a capability weight matrix. Step 5: Generate a structured dynamic evaluation report by weighted fusion calculation of the capability weight matrix and the enterprise mentor's score; Step 6: Based on the dynamic evaluation report, perform resource matching calculations through the API interface of the enterprise cooperation platform to generate enterprise feedback data packets; parse the enterprise feedback data packets, extract the demand gap feature vector and equipment parameter set, and trigger the mirror incremental update operation of the industry dynamic database to re-execute the engineering practice task generation process.

2. The method for optimizing environmental art design courses according to claim 1, characterized in that, Step 1: Collect real-time data streams on the characteristics of new materials, construction techniques, and design standards in the environmental art design industry to build a dynamic industry database; Based on an industry dynamics database, outdated knowledge points are filtered out and replaced with cutting-edge industry technology content to generate an updated course content library, including: The IoT sensor array collects data on the characteristics of new materials and construction processes in the environmental art design industry in real time, and simultaneously calls the industry data interface to obtain the design standard data stream. Based on the characteristics of new materials, construction process data, and design standard data streams, a dynamic industry database is constructed using distributed storage technology. Based on an industry dynamic database, semantic feature analysis is performed on the historical course content, and a set of lagging knowledge point identifiers is output through similarity matching calculation. Based on the lagging knowledge point identifier set, the knowledge graph reconstruction operation is performed by calling the cutting-edge technology data flow of the industry dynamic database to generate an updated course content library.

3. The method for optimizing the environmental art design curriculum according to claim 2, characterized in that, Step 2: Based on the updated course content library, match the actual needs of ongoing construction projects; decompose the project requirements into executable sub-tasks, and establish a version mapping relationship between the sub-tasks and cutting-edge technology content through a mirroring algorithm, generating a set of practical tasks with version identifiers, including: Based on the updated course content library, extract course content feature vectors; obtain the BIM parameter set of actual construction projects and parse it into engineering requirement feature vectors; calculate the cosine similarity between the course content feature vectors and the engineering requirement feature vectors using a collaborative filtering algorithm; filter matching items according to the similarity threshold and generate a requirement-content mapping relationship table. Based on the requirement-content mapping table, a graph segmentation algorithm is used to decompose project requirements into a sequence of executable sub-task units. Based on the subtask unit sequence, the technical requirement description field of each subtask unit is parsed to extract the technical requirement identifier; using the technical requirement identifier as the search key, the version hash value of the corresponding cutting-edge technology content in the course content library is queried; and a key-value pair mapping is established between each subtask unit ID and its corresponding version hash value to obtain an ID-hash value mapping table. Based on the ID-hash value mapping table, the mapping table is encapsulated and a timestamp verification code is embedded to obtain a version mapping relationship table with verification. By integrating a version mapping table with verification and engineering constraint parameters, including schedule thresholds, cost boundaries, and specification compliance indicators, a set of practice tasks with version identifiers is generated.

4. The method for optimizing the environmental art design curriculum according to claim 3, characterized in that, Step 3: Based on the execution process data of the practice task set with version identifiers, deploy log tracking technology to collect student operation log streams in real time, including: Based on the practice task set with version identifier, extract the task set version metadata and generate a version synchronization event configuration scheme. Based on the data entry configuration scheme, a log collection agent is deployed at the task execution node; The log collection agent captures student operation events in real time and encapsulates the operation log stream according to version identifier; Based on the version metadata of the task set, verify the consistency of the version identifier of the operation log stream and output the operation log stream that has passed the verification.

5. The method for optimizing the environmental art design curriculum according to claim 4, characterized in that, Step 4: Separate the construction decision node dataset from the operation log stream using time-series feature extraction technology, analyze the team collaboration interaction dataset using natural language processing technology, and identify the innovative solution adoption dataset using solution identifier matching technology; Based on the decision node dataset, interaction dataset, and adoption dataset, an entropy weighting method is applied to dynamically allocate weight coefficients to construct a capability weight matrix, including: Based on the time sequence feature extraction of the operation log stream, the key decision timestamp sequence is identified, and the log stream is segmented according to a preset time window to generate a log segment set; the construction decision node dataset is extracted from the log segment set. Based on the operation log stream, natural language processing is performed to identify and separate team collaboration communication text segments to obtain a set of communication text segments; based on the set of communication text segments, task assignment instructions and conflict resolution records in the interaction text are parsed to generate a team collaboration interaction dataset. Based on the operation log stream, a scheme identifier scan is performed to extract the design scheme submission event and its metadata; the feature identifier of the scheme to be reviewed is extracted from the metadata; the feature identifier of the scheme to be reviewed is compared with the feature code of the historical adopted scheme library to obtain the scheme adoption status set; based on the scheme adoption status set, the scheme metadata and adoption status are fused to generate an innovative scheme adoption dataset; A three-dimensional evaluation index matrix is ​​constructed based on the decision node dataset, interaction dataset, and adoption dataset. The X-axis represents the decision response timeliness index, the Y-axis represents the collaborative task completion rate index, and the Z-axis represents the innovative solution adoption rate index. The entropy weight method is applied to calculate the weight coefficients of each index. The weight coefficients are then embedded to generate a capability weight matrix.

6. The method for optimizing the environmental art design curriculum according to claim 5, characterized in that, Step 5: Generate a structured dynamic evaluation report by weighted fusion calculation of the capability weight matrix and the corporate mentor scores, including: By calling the rating service of the enterprise cooperation platform through the API interface, the rating value of the enterprise mentor can be obtained in real time. Based on the capability weight matrix and the corporate mentor rating, a timestamp synchronization operation is performed to generate an aligned data pair set. Based on the aligned data set, the weight coefficients of the three-dimensional indicators in the capability weight matrix are extracted, the normalized score of the enterprise mentor's score is calculated, and a comprehensive score vector is generated according to the preset fusion ratio. Based on the comprehensive scoring vector, dynamic timestamps and version identifiers are embedded to generate a machine-readable structured dynamic evaluation report.

7. The method for optimizing the environmental art design curriculum according to claim 6, characterized in that, Step 6: Based on the dynamic evaluation report, perform resource matching calculations through the API interface of the enterprise cooperation platform to generate enterprise feedback data packets; parse the enterprise feedback data packets, extract the demand gap feature vector and equipment parameter set, and trigger the mirror incremental update operation of the industry dynamic database to re-execute the engineering practice task generation process, including: Based on the dynamic evaluation report, resource matching calculations are performed through the API interface of the enterprise cooperation platform to generate enterprise feedback data packets; Based on enterprise feedback data packets, the data packet structure is parsed to extract demand gap feature vectors and equipment parameter sets; Based on the demand gap feature vector and equipment parameter set, a mirror incremental update operation of the industry dynamic database is triggered to generate an updated database with version identifier; The updated database, based on version identifiers, re-executes the engineering practice task generation process.

8. An environmental art design curriculum optimization system, wherein the system implements the method as described in any one of claims 1 to 7, characterized in that, include: The replacement module is used to collect real-time data streams of new material properties, construction techniques and design standards in the environmental art design industry, and to build a dynamic industry database. The course content library is updated by filtering outdated knowledge points from the industry dynamics database and replacing them with cutting-edge industry technology content. The mapping module is used to match the needs of actual construction projects with the updated course content library; it decomposes the project requirements into executable sub-tasks, and establishes a version mapping relationship between the sub-tasks and the cutting-edge technology content through a mirroring algorithm, generating a set of practical tasks with version identifiers. The data acquisition module is used to collect student operation log streams in real time based on the execution process data of the practice task set with version identifiers and deploy log tracking technology. The module is used to separate the construction decision node dataset from the operation log stream using time-series feature extraction technology, parse the team collaboration interaction dataset using natural language processing technology, and identify the innovative solution adoption dataset using solution identifier matching technology. Based on the decision node dataset, interaction dataset, and adoption dataset, the entropy weight method is applied to dynamically allocate weight coefficients to construct a capability weight matrix. The fusion module is used to generate a structured dynamic evaluation report by weighted fusion calculation of the capability weight matrix and the enterprise mentor's score; The feedback and update module is used to perform resource matching calculations based on dynamic evaluation reports through the API interface of the enterprise cooperation platform, generate enterprise feedback data packages, parse enterprise feedback data packages, extract demand gap feature vectors and equipment parameter sets, and trigger mirror incremental update operations of the industry dynamic database to re-execute the engineering practice task generation process.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images