Environmental art design course recommendation method and system

Abstract

The invention provides an environmental art design course recommendation method and system, and relates to the technical field of intelligent education, and the method comprises the steps: screening a candidate course set adaptive to the current ability of a student according to the ability matching degree in a multi-dimensional student portrait and the pre-repair relation between courses in a dynamic course knowledge graph; setting double target points in the candidate course set, wherein the double target points comprise the highest matching value of course basic requirements and student abilities and the highest association value of course high-order abilities and occupational planning; constructing a linear evaluation axis, calculating a deviation angle and a midpoint slope of an axis direction and a reference direction, and generating a course display correction parameter; dynamically adjusting the multi-modal display content of the candidate course set according to the course display correction parameters; and generating an interactive recommendation list based on the multi-modal display content. The course selection efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent education technology, and in particular to a method and system for recommending environmental art design courses. Background Art

[0002] With the development of educational informatization, personalized course recommendation solutions have become an important research direction for improving learning efficiency. Current mainstream recommendation methods, some of which rely on student performance, interests, or collaborative filtering algorithms, lack systematic modeling of the logical connections between courses, affecting the coherence of recommended knowledge. Furthermore, some traditional recommendation models rely on static rules, making them difficult to adapt to the dynamic changes in learners' abilities. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and system for recommending environmental art design courses, which can improve course selection efficiency.

[0004] In order to solve the above technical problems, the technical solutions of the present invention are as follows: In a first aspect, a method for recommending an environmental art design course comprises: Step S1: Construct the knowledge graph of the environmental art design course, extract course entities, knowledge point entities and interdisciplinary associations, establish prerequisite and knowledge connection relationships between courses, and generate a dynamic course knowledge graph containing course basic requirements and high-level ability labels; Step S2: Based on the dynamic course knowledge graph, collect student interests, ability test and career planning data, and generate a multi-dimensional student portrait by matching the model to calculate the correlation between ability and basic course requirements, and between career goals and advanced course abilities; Step S3: Based on the ability matching degree in the multi-dimensional student portrait and the prerequisite relationship between courses in the dynamic course knowledge graph, select the candidate course set that is suitable for the student's current ability; Step S4: Set two target points in the candidate course set: the highest matching value between the basic course requirements and student abilities, and the highest correlation value between the advanced course abilities and career planning; construct a linear evaluation axis, calculate the offset angle and midpoint slope of the axis direction from the reference direction, and generate course display correction parameters; Step S5: Dynamically adjust the multimodal display content of the candidate course set according to the course display correction parameters; Step S6: Generate an interactive recommendation list based on the multimodal presentation content.

[0005] Furthermore, step S1 specifically includes: Extracting course entities from the course syllabus, teaching resource library, and teacher files, wherein the course entities include attributes such as course name, credits, and teaching method; Based on the course content text in the course entity, knowledge point entities are extracted through semantic analysis, and the logical association relationship between knowledge points is established based on the semantic similarity calculation of knowledge point entities to form the core knowledge chain of the knowledge graph; Based on the knowledge point entities and their logical associations in the core knowledge chain, combined with the prerequisite course declarations in the teaching plan corresponding to the course entity, the mastery of the knowledge point entities in the student's historical learning data is analyzed, the prerequisite dependency strength and knowledge connection weight between courses are calculated, and the hierarchical relationship between course nodes is constructed; Based on the course nodes in the course hierarchy, we extract the data of cooperating teachers in interdisciplinary course projects, associate them with the subject tags of student works corresponding to the knowledge point entities, and establish the association relationship between interdisciplinary courses to obtain interdisciplinary association data. Based on cross-disciplinary related data, course entity attributes, knowledge point associations, and prerequisite dependency strength, when new courses are added or teachers adjust teaching content, the entities and relationships in the knowledge graph are dynamically and synchronously updated to generate a dynamic course knowledge graph containing basic course requirements and advanced ability labels.

[0006] Furthermore, step S2 specifically includes: Based on the basic course requirements and high-level competency labels in the dynamic course knowledge graph, we collect student data sets including interest preference data, career planning goals, and competency test results. Based on the ability test results, the matching model is used to calculate the matching degree between the student's ability and the basic course requirements in the dynamic course knowledge map. At the same time, based on the career planning goals, the semantic relevance between them and the high-level ability labels in the course knowledge map is calculated. The matching degree and semantic relevance are integrated with the interest preference data in a multi-dimensional weighted manner to generate a multi-dimensional student portrait that includes quantitative values ​​of ability levels, career orientation labels and interest feature vectors.

[0007] Furthermore, step S3 specifically includes: Based on the ability matching degree of the multi-dimensional student portrait, the course nodes with matching degree exceeding the preset threshold are screened out from the dynamic course knowledge graph to generate a basic set of candidate courses that are initially adapted to the student's current ability level; Based on the basic candidate course set and combined with the prerequisite relationships between courses in the dynamic course knowledge graph, we remove course nodes where students do not meet the prerequisite knowledge connection weights, and generate a refined candidate course set that meets the knowledge coherence requirements. Based on the refined candidate course set, high-level ability labels associated with career planning goals in student profiles are extracted, and high-level course nodes matching the labels are retained to form a hybrid candidate course set that is adapted to students' current abilities. The hybrid candidate course set includes basic courses and advanced courses, and the ability improvement paths corresponding to the advanced courses are marked.

[0008] Furthermore, step S4 specifically includes: Set two target points in the candidate course set. The first target point is the maximum value of the match between the basic requirements of the course and the students' abilities. The second target point is the maximum value of the correlation between the advanced abilities of the course and career planning. A linear evaluation axis is constructed with the two target points as endpoints, the offset angle of the axis direction relative to the preset reference direction is calculated, and the slope value of the axis midpoint is extracted; According to the combination relationship between the offset angle and the slope value, the course display correction parameters are generated.

[0009] Furthermore, step S5 specifically includes: Determine the display mode type of the candidate course set according to the offset angle and slope value in the course display correction parameter, and generate a display mode instruction set including a priority label; If the priority label in the display mode instruction set points to the career-related direction and the slope exceeds the threshold, the interdisciplinary VR practice scenario resource library is called, and based on the high-level course nodes in the candidate course set, VR scenes and related high-level cases that match the career goals are rendered to generate career-oriented units in the multimodal display content; If the priority label in the display mode instruction set points to the ability adaptation direction and the slope is lower than the threshold, the knowledge point map visualization engine is activated to extract the basic course nodes and their knowledge point logic chains from the candidate course set, generate a dynamic visualization map, and associate the work cases in the student's historical portfolio that match the current ability for prominent display, thus generating the ability adaptation unit in the multimodal display content; Based on the ability adaptation unit and career orientation unit, they together constitute the dynamically adjusted multimodal display content.

[0010] Furthermore, step S6 specifically includes: Based on multimodal display content, the prerequisite relationship weights in the course knowledge graph, the interest tags in the student portrait, and the priority data in the correction parameters are integrated to generate an interactive recommendation list.

[0011] Second, an environmental art design course recommendation system includes: The extraction module is used to construct the knowledge graph of the environmental art design course, extract course entities, knowledge point entities and interdisciplinary associations, establish prerequisite and knowledge connection relationships between courses, and generate a dynamic course knowledge graph containing course basic requirements and high-level ability labels; The matching module is used to collect student interests, ability test and career planning data based on the dynamic course knowledge graph. The matching model calculates the correlation between ability and basic course requirements, and between career goals and advanced course abilities, generating a multi-dimensional student portrait. The screening module is used to select candidate courses that are suitable for students' current abilities based on the ability matching degree in the multi-dimensional student profile and the prerequisite relationships between courses in the dynamic course knowledge graph; The correction module is used to set two target points in the candidate course set: the highest match between the basic course requirements and student abilities, and the highest correlation between the advanced course abilities and career planning. It also constructs a linear evaluation axis, calculates the offset angle between the axis direction and the reference direction, and the midpoint slope, and generates course presentation correction parameters. A dynamic adjustment module is used to dynamically adjust the multimodal display content of the candidate course set according to the course display correction parameters; The interactive module is used to generate interactive recommendation lists based on multimodal display content.

[0012] The above solution of the present invention includes at least the following beneficial effects: By building a dynamic course knowledge graph through multi-source data, the course knowledge is presented in a structured manner and updated in real time. By integrating multi-dimensional student data to accurately generate portraits, candidate courses are screened in layers based on the portraits and knowledge graphs, and an interactive recommendation list is generated based on multimodal content, which can realize intelligent recommendations and improve course selection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The present invention provides a flow chart of an environmental art design course recommendation method.

[0014] Figure 2 This is a schematic diagram of an environmental art design course recommendation system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0015] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0016] like Figure 1 As shown, an embodiment of the present invention provides a method for recommending environmental art design courses, the method comprising the following steps: Step S1: Construct the knowledge graph of the environmental art design course, extract course entities, knowledge point entities and interdisciplinary associations, establish prerequisite and knowledge connection relationships between courses, and generate a dynamic course knowledge graph containing course basic requirements and high-level ability labels; Step S2: Based on the dynamic course knowledge graph, collect student interests, ability test and career planning data, and generate a multi-dimensional student portrait by matching the model to calculate the correlation between ability and basic course requirements, and between career goals and advanced course abilities; Step S3: Based on the ability matching degree in the multi-dimensional student portrait and the prerequisite relationship between courses in the dynamic course knowledge graph, select the candidate course set that is suitable for the student's current ability; Step S4: Set two target points in the candidate course set: the highest matching value between the basic course requirements and student abilities, and the highest correlation value between the advanced course abilities and career planning; construct a linear evaluation axis, calculate the offset angle and midpoint slope of the axis direction from the reference direction, and generate course display correction parameters; Step S5: Dynamically adjust the multimodal display content of the candidate course set according to the course display correction parameters; Step S6: Generate an interactive recommendation list based on the multimodal presentation content.

[0017] In an embodiment of the present invention, a dynamic course knowledge graph is constructed to achieve structured integration and dynamic updating of environmental art design course knowledge, and multi-dimensional data fusion is used to generate accurate student portraits; intelligent course recommendations are achieved using mechanisms such as hierarchical screening and dual-objective optimization, and the course selection experience is enhanced with multimodal dynamic display; at the same time, teaching and career development are closely linked to generate personalized learning paths for students, effectively reducing the blindness of course selection, supporting real-time dynamic adjustment of the teaching process according to changes in student abilities and course knowledge, and improving teaching efficiency and student learning experience.

[0018] In a preferred embodiment of the present invention, the above step S1: constructing the environmental art design course knowledge graph, extracting course entities, knowledge point entities and interdisciplinary associations, establishing prerequisite and knowledge connection relationships between courses, and generating a dynamic course knowledge graph containing course basic requirements and high-level ability labels, may include: Step S11, extracting a course entity from the course syllabus, teaching resource library and teacher files, wherein the course entity includes attributes such as course name, credits and teaching method; Step S12: Based on the course content text in the course entity, knowledge point entities are extracted through semantic analysis, and logical association relationships between knowledge points are established based on the semantic similarity calculation of the knowledge point entities to form the core knowledge chain of the knowledge graph; Step S13: Based on the knowledge point entities and their logical associations in the core knowledge chain, combined with the prerequisite course declarations in the teaching plan corresponding to the course entity, the mastery of the knowledge point entities in the student's historical learning data is analyzed, the prerequisite dependency strength and knowledge connection weight between courses are calculated, and the hierarchical relationship between course nodes is constructed; Step S14: extracting the data of cooperating teachers of interdisciplinary course projects based on the course nodes in the course hierarchy, and associating the subject tags of student works corresponding to the knowledge point entities to establish the association relationship between interdisciplinary courses to obtain interdisciplinary association data; Step S15: Based on cross-disciplinary related data, course entity attributes, knowledge point association relationships, and prerequisite dependency strength, when new courses are added or teachers adjust teaching content, the entities and relationships in the knowledge graph are dynamically and synchronously updated to generate a dynamic course knowledge graph containing course basic requirements and high-level ability labels.

[0019] In an embodiment of the present invention, course entity attributes are extracted from multi-source data to achieve accurate integration of course information, laying a solid foundation for knowledge graph construction; core knowledge chains are constructed through semantic analysis, knowledge associations are deeply mined, and the fragmented state of knowledge is changed, which helps students to systematically master knowledge; relevant parameters are calculated based on historical learning data, and learning paths are scientifically planned to avoid knowledge gaps; through teacher cooperation and association with work themes, interdisciplinary integration is promoted, broadening students' knowledge and innovative thinking; dynamic and synchronous updating of knowledge graphs can respond to changes in courses and teaching content in a timely manner, making them fit for teaching reality, providing accurate data for course recommendations and teaching evaluation, and improving teaching quality and efficiency.

[0020] In an embodiment of the present invention, the specific steps include: Step S11: Extract basic information such as course name, credits, and teaching methods from the course syllabus as the core attributes of the course entity; in the teaching resource library, verify and supplement the course entity attributes through a combination of keyword matching and manual verification to ensure the integrity of the information; for teacher files, extract the information of the teacher responsible for course teaching, associate it with the course entity, and clarify the teaching subject of each course.

[0021] Step S12: perform word segmentation on the course content text in the course entity, break down the long text into independent words or phrases, use natural language processing technology to analyze the semantics of each word or phrase, and screen out knowledge point entities with knowledge representation significance; by calculating the semantic similarity between knowledge point entities, such as comparing word meanings, context, etc., the logical relationship between knowledge points is judged. If the semantic similarity is high, it is considered that the two knowledge points are closely related, and then these knowledge point entities are connected in series to form a core knowledge chain.

[0022] Step S13: Based on the logical association between knowledge points in the core knowledge chain and the prerequisite declaration of courses in the teaching plan, the preliminary prerequisite order between courses is sorted out; the students' historical learning data is analyzed, and the mastery of each knowledge point entity in the learning process of different courses is counted. If a certain knowledge point is frequently difficult to master in the subsequent course learning, and the knowledge point has been involved in the previous course, it means that the previous course has a high degree of dependence on the prerequisite of the subsequent course; according to the close connection between knowledge points between courses, the corresponding knowledge connection weight is assigned, and finally a structure with a clear hierarchical relationship between course nodes is constructed.

[0023] Step S14 focuses on the dimension of faculty collaboration, accurately extracting data on cooperating teachers from the data related to interdisciplinary course projects. Specifically, if teachers are found to be co-teaching in courses with different disciplinary attributes, we can preliminarily determine the potential for interdisciplinary collaboration between these courses based on the role of teachers in curriculum development and teaching practice. For example, if a teacher is responsible for teaching both computer science and art design courses, there is a high probability that these two courses will have cross-integration in terms of teaching content and practical projects.

[0024] We conducted an in-depth analysis of student works, carefully examining the topic tags corresponding to knowledge point entities. When student works from different courses share the same or similar themes, this indicates that students applied shared knowledge to solve similar problems during their studies, further clarifying the existence of interdisciplinary connections between courses. For example, if students in architecture and materials science courses create works centered around the application of environmentally friendly building materials in both courses, this confirms a close connection between the two courses at the level of knowledge application.

[0025] The potential correlations obtained from the teacher-student collaboration analysis are fully integrated with the actual correlation information determined by the thematic analysis of student works. After data cleaning, deduplication, correlation verification and other processes, complete and accurate cross-disciplinary correlation data are finally obtained.

[0026] Step S15: When a new course is added, follow the process of steps S11-S14 to extract the entity attributes, knowledge point entities, and association relationships of the new course, and integrate them into the existing knowledge graph; if the teacher adjusts the teaching content, re-analyze the course content text, update the knowledge point entities and their logical association relationships, and at the same time, recalculate the prerequisite dependency strength and knowledge connection weight between courses according to changes in the teaching plan, synchronously update the entities and relationships in the knowledge graph, and add course basic requirements and high-level ability labels to keep the knowledge graph always up to date.

[0027] In a preferred embodiment of the present invention, the above step S2: collecting student interests, ability test and career planning data based on the dynamic course knowledge graph, and generating a multi-dimensional student portrait by matching the model to calculate the correlation between ability and basic course requirements, and between career goals and advanced course abilities, may include: Step S21: Based on the basic course requirements and high-level competency labels in the dynamic course knowledge graph, collect student data sets including interest preference data, career planning goals, and competency test results; Step S22: Based on the ability test results, the matching model is used to calculate the matching degree between the student's ability and the basic course requirements in the dynamic course knowledge graph; at the same time, based on the career planning goals, the semantic relevance between them and the high-level ability labels in the course knowledge graph is calculated; In step S23, the matching degree and semantic relevance are integrated with the interest preference data in a multi-dimensional weighted manner to generate a multi-dimensional student portrait including the ability level quantification value, career tendency label and interest feature vector.

[0028] In an embodiment of the present invention, student interest, career planning and ability test data are collected through multiple channels to comprehensively cover the key dimensions of student learning and development, laying a solid data foundation for accurate portraits; with the help of a matching model, the correlation between deep computing ability and basic course requirements, career goals and high-level abilities is calculated, and the points of fit between students' existing abilities and course requirements and career goals are accurately located, providing quantitative support for personalized teaching and career development guidance; using multi-dimensional weight fusion, various types of data are organically integrated to generate a three-dimensional student portrait covering ability quantification, career tendencies and interest characteristics. Compared with a single-dimensional assessment, it can more comprehensively reflect the overall picture of students, facilitate teachers and systems to formulate teaching and training plans that meet students' needs, improve the pertinence and effectiveness of teaching, and help students achieve their personal development goals.

[0029] In an embodiment of the present invention, the specific steps include: Step S21: Extract the basic requirements and high-level competence labels of the course from the dynamic course knowledge graph as a reference for collecting student data; collect students' interest preference data through online questionnaires, learning platform behavior records, etc., covering artistic style preferences, design tool preferences, project type tendencies, etc.; obtain students' career planning goals, such as specific career directions such as interior designer and landscape planner, through career planning interviews and future career intention filling; extract students' ability test results from the school's examination system and professional skills testing platform, including theoretical knowledge assessment scores, software operation proficiency scores, design practice project scores, and other data, and integrate them to form a complete student data set.

[0030] In step S22, based on the results of the ability test, the student's performance in each test is compared with the knowledge and skills points in the basic requirements of the course one by one; for example, if the basic requirements of the course include "proficient in using AutoCAD software to draw floor plans", then based on the student's score on the AutoCAD software operation test, the student's mastery of the skill is evaluated, and then the matching degree between the student's ability and the basic requirements of the course is determined; for career planning goals, the core competencies required for the target career are analyzed, and they are semantically compared with the high-level competency labels in the course knowledge map; for example, if the career goal is landscape planner, the degree of semantic correlation between the site analysis and planning ability, ecological design concept application ability, etc. in the high-level competency labels of the course and the career goal is determined, and the semantic correlation between the career planning goal and the high-level competency labels of the course is calculated.

[0031] In step S23, different weights are set for matching, semantic relevance, and interest preference data based on educational and teaching experience and professional domain knowledge. For example, for recommendations of professional foundation courses, the matching weight between ability and course foundation requirements can be set higher; while for recommendations of elective courses with a clear career orientation, the semantic relevance weight between career planning goals and higher-level abilities can be appropriately increased. The various data are integrated and calculated according to the set weights, and the student's ability is converted into a specific quantitative value, such as ability level is divided into 1-5 levels. Based on the correlation between career planning goals and higher-level abilities, students are assigned corresponding career tendency labels, such as interior design direction, environmental art engineering direction, etc. The interest preference data is organized into interest feature vectors, and ultimately a comprehensive and detailed multi-dimensional student portrait is generated.

[0032] In a preferred embodiment of the present invention, the above step S3: screening a set of candidate courses that are suitable for the student's current ability based on the ability matching degree in the multi-dimensional student portrait and the prerequisite relationship between courses in the dynamic course knowledge graph, may include: Step S31: Based on the ability matching degree in the multi-dimensional student portrait, select course nodes with matching degrees exceeding a preset threshold from the dynamic course knowledge graph to generate a basic candidate course set that is initially adapted to the student's current ability level; Step S32: Based on the basic candidate course set and in combination with the prerequisite relationships between courses in the dynamic course knowledge graph, remove course nodes where students do not meet the prerequisite knowledge connection weights, and generate a refined candidate course set that meets the knowledge coherence requirements; Step S33: Based on the refined candidate course set, extract the high-level ability labels associated with the career planning goals in the student profile, retain the high-level course nodes that match the labels, and form a hybrid candidate course set that adapts to the student's current ability; wherein, the hybrid candidate course set includes basic courses and advanced courses, and marks the ability improvement paths corresponding to the advanced courses.

[0033] In an embodiment of the present invention, courses are screened by setting an ability matching threshold so that candidate courses are accurately adapted to students' current ability levels, thereby avoiding the impact of inappropriate learning difficulty on enthusiasm and effectively improving learning efficiency; courses with substandard prerequisite knowledge are eliminated based on the course prerequisite relationship and knowledge connection weight, thereby ensuring the continuity of knowledge learning and maintaining the integrity of the knowledge system; high-level courses are screened based on high-level ability labels associated with students' career planning goals, deeply binding course learning with career development, meeting the needs for ability improvement while laying the foundation for career goals and enhancing the targeted nature of learning; the final generated hybrid candidate course set marks the high-level course ability improvement path, which not only points out the direction for students to formulate personalized learning plans, but also provides a reference for teachers' targeted teaching, thereby promoting the improvement of teaching quality.

[0034] In an embodiment of the present invention, the specific steps include: Step S31 extracts ability matching data from the multi-dimensional student portrait. This data reflects the degree of fit between the student's current ability and the basic course requirements. At the same time, all course nodes and their corresponding basic course requirements are obtained from the dynamic course knowledge graph. A reasonable preset threshold for ability matching is then set. This threshold can be determined based on factors such as teaching experience and course difficulty. The basic requirements of each course node are compared with the student's ability matching, and course nodes with an ability matching exceeding the preset threshold are selected. For example, if the preset threshold is 70%, course nodes with a student ability matching greater than 70% are selected and integrated to generate a set of basic candidate courses that are initially suitable for the student's current ability level.

[0035] In step S32, after obtaining the basic candidate course set, the prerequisite courses and corresponding knowledge connection weights for each course are sorted out based on the prerequisite relationships between courses in the dynamic course knowledge graph. For each course in the basic candidate course set, the student is checked to see whether they have mastered the knowledge and skills required by the prerequisite course. This is determined by reviewing data such as the student's grades and knowledge points mastered in the prerequisite courses in their historical learning records. If a student's mastery of the prerequisite course knowledge for a particular course is insufficient, that is, if the required knowledge connection weight is not met, the course is removed from the basic candidate course set. After this screening process, the set of course nodes that remain is a refined candidate course set that meets the knowledge coherence requirements, ensuring that students will not encounter learning difficulties due to knowledge gaps in their subsequent studies.

[0036] Step S33, based on the refined candidate course set, extract the high-level ability labels associated with the career planning goals from the student portrait. These labels represent the key abilities that students need to possess for their future career development; analyze each course node in the refined candidate course set to determine whether it matches the high-level ability labels associated with the career planning goals. If the course content corresponding to the course node can help students develop these high-level abilities, the course node will be retained; otherwise, it will be eliminated. After screening, the resulting course node set is a hybrid candidate course set that is adapted to the students' current abilities. The set includes both basic courses that strengthen students' basic abilities and high-level courses that help students move towards their career goals. Finally, for the high-level courses in the hybrid candidate course set, analyze the correspondence between their course content and high-level ability labels, and mark the specific ability improvement path that each high-level course can help students achieve, such as the ability advancement process from mastering basic design theory to completing complex project design.

[0037] In a preferred embodiment of the present invention, step S4: setting two target points in the candidate course set: the highest matching value between the basic course requirements and the student's ability, and the highest correlation value between the advanced course ability and the career plan; constructing a linear evaluation axis, calculating the offset angle and midpoint slope between the axis direction and the reference direction, and generating course display correction parameters, may include: Step S41: setting two target points in the candidate course set, the first target point being the maximum value of the matching value between the basic course requirements and the student's ability, and the second target point being the maximum value of the correlation value between the advanced course ability and career planning; Step S42: constructing a linear evaluation axis with the two target points as endpoints, calculating the offset angle of the axis direction relative to the preset reference direction, and extracting the slope value of the axis midpoint; Step S43: Generate course display correction parameters according to the combination relationship between the offset angle and the slope value.

[0038] In an embodiment of the present invention, by setting dual target points, focusing on the best match between courses and students' abilities and career plans, the core direction of course recommendations is clarified, and the recommendation results take into account students' current learning needs and future career development goals, thereby improving the accuracy and practicality of recommendations; calculating the offset angle and slope value, quantifying the characteristics of the candidate course set in terms of ability matching and career planning, and presenting course differences and distribution trends with intuitive parameters, providing data support for the system and teachers to analyze course characteristics and make teaching decisions. Based on the offset angle and slope value, course display correction parameters are generated, and the display content and method are dynamically adjusted according to the actual situation of the candidate course set. By focusing on displaying courses related to career planning and implementing differentiated display strategies, students can be helped to clearly recognize the value of the course, improve course selection efficiency and satisfaction, and at the same time help teachers optimize course introductions and teaching resource presentations to enhance teaching effectiveness.

[0039] In an embodiment of the present invention, the specific steps include: Step S41: The degree of match between the basic requirements of each course and the student ability test results is expressed as a percentage; the degree of correlation between the course's advanced abilities and the student's career planning goals is determined through semantic analysis and expert annotation.

[0040] The second target point is to find the course with the highest ability matching value among all courses; for example: Course A (ability matching 90%), Course B (ability matching 75%), Course C (ability matching 80%), the first target point = Course A (90%).

[0041] The second goal is to find the course with the highest career relevance value; for example: Course A (career relevance 60%), Course B (career relevance 85%), Course C (career relevance 70%), the second goal = Course B (85%).

[0042] Consider each course as a point in a two-dimensional coordinate system (X-axis = ability match, Y-axis = career relevance). The dual target points are the two extreme points on the right (highest ability match) and at the top (highest career relevance) of the candidate course set.

[0043] In step S42, a straight line is used to connect the two target points (such as course A and course B) determined in step S41. This line represents the optimal development path of the candidate course set, that is, the optimal transition from current capabilities to career goals.

[0044] Set the ideal development path according to the teaching objectives. For example, a 45-degree diagonal line indicates that ability improvement and career goals develop simultaneously (balanced type). The horizontal direction prioritizes ability improvement and then considers career relevance (foundation consolidation type). The vertical direction prioritizes matching career goals and supplements ability improvement (career-oriented type).

[0045] The offset angle is the angle between the axis direction and the reference direction. A small angle (close to 0 degrees) indicates that the candidate course set is highly consistent with the ideal path. A large angle (such as 45 degrees) indicates that the course set is biased toward ability matching or career association. If the reference direction is 45 degrees and the actual axis direction is 60 degrees, the offset angle is 15 degrees (biased toward career association).

[0046] The midpoint is the coordinate of the center point of the axis (average value of ability matching, average value of occupational correlation), and the slope is the steepness of the curve at the midpoint, which indicates the rate of increase in difficulty from basic courses to advanced courses. A large slope (such as 2.0) means that the course difficulty increases rapidly and is suitable for students with strong abilities; a small slope (such as 0.5) means that the course gradient is gentle and suitable for step-by-step learning.

[0047] In step S43, a correction strategy is formulated according to a combination of the offset angle and the slope value, wherein the correction strategy is shown in Table 1 below.

[0048] Table 1 Correction strategy:

[0049] Adjust the sorting, size, and color of course cards based on the correction parameters. For example, the career-enhancing parameters will increase the size of the Landscape Planning Principles course card by 30% and add a "Career Express" icon. Personalize recommendation copy and dynamically generate recommendation reasons. For example, based on your career goals, it is recommended to study this course that can enhance the innovation ability of landscape design. Learning path planning optimizes the order of course arrangement. For example, when the slope is large, it is recommended to take basic course A first, and then challenge advanced course B.

[0050] In a preferred embodiment of the present invention, the above step S5: dynamically adjusting the multimodal display content of the candidate course set according to the course display correction parameter may include: Step S51, determining the presentation mode type of the candidate course set according to the offset angle and slope value in the course presentation correction parameter, and generating a presentation mode instruction set including a priority tag; Step S52: If the priority tag in the presentation mode instruction set points to a career-related direction and the slope exceeds the threshold, the interdisciplinary VR practice scenario resource library is called, and based on the high-level course nodes in the candidate course set, VR scenarios and related high-level cases that match the career goals are rendered to generate career-oriented units in the multimodal presentation content. Step S53: If the priority tag in the display mode instruction set points to the ability adaptation direction and the slope is lower than the threshold, the knowledge point map visualization engine is activated to extract the basic course nodes and their knowledge point logic chains in the candidate course set, generate a dynamic visualization map, and associate the student's historical work collection with the current ability of the work case for prominent display, thereby generating the ability adaptation unit in the multimodal display content; Step S54: Based on the capability adaptation unit and the career orientation unit, dynamically adjusted multimodal display content is jointly constructed.

[0051] In the embodiment of the present invention, parameter mapping is used to achieve accurate pattern recognition, avoiding manual configuration errors, and converting abstract parameters into standardized executable display instructions to provide guidance for subsequent operations. VR technology is used to create an immersive career experience, and real project cases are used to enhance practicality and appeal, breaking through the limitations of traditional display. The learning path is visualized using knowledge point maps, and personalized references are achieved through historical work comparisons. Tiered learning incentives are formed through step-by-step case presentations. At the same time, it ensures that ability adaptation and career-oriented content complement each other, supports multiple layouts to ensure device compatibility, and enhances user engagement with dynamic interactive elements.

[0052] In an embodiment of the present invention, the specific steps include: In step S51, the offset angle and slope value are input into a preset decision tree model. If the offset angle is biased toward the career-related direction (e.g., >30°) and the slope exceeds the threshold (e.g., >1.5), it is determined to be a career-enhancing display mode; if the offset angle is biased toward the ability-adapting direction (e.g., <15°) and the slope is lower than the threshold (e.g., <0.8), it is determined to be a foundation-consolidating display mode; in other cases, it is determined to be a balanced development mode.

[0053] Assign a corresponding priority label to each display mode: PRIORITY_PROFESSIONAL Foundation consolidation type → PRIORITY_FOUNDATION; Balanced development type → PRIORITY_BALANCE.

[0054] According to the display mode type, the corresponding display instruction set is called from the preset template library. The instruction set includes: course sorting rules, content highlighting strategies, interactive element configuration, etc.

[0055] In step S52, when the PRIORITY_PROFESSIONAL tag is detected, content matching the advanced course node is filtered from the interdisciplinary VR practice scene resource library. For example, if the advanced course is commercial space design, resources such as shopping mall VR roaming and brand store design case library are retrieved.

[0056] Extract keywords from students' career planning goals (such as commercial space, sustainable design, etc.), perform secondary rendering of VR scenes based on these keywords, and highlight design elements and solutions related to their career goals.

[0057] Real business cases were retrieved from the corporate cooperation project library and associated according to the following dimensions: industry matching (such as retail, catering), design complexity (elementary / intermediate / advanced), and innovative technology applications (such as smart lighting and interactive devices).

[0058] Step S53: extract the core knowledge points in the candidate basic courses, and construct a hierarchical relationship diagram between the knowledge points based on the knowledge chain. For example, plane composition is a prerequisite knowledge point of space design, which is represented by arrow lines.

[0059] The map display is dynamically adjusted according to the student's current ability level: the mastered knowledge points are displayed in gray semi-transparent color, with the mastery time and score marked; the knowledge points being learned are highlighted, showing the current progress bar; the knowledge points that have not been learned are displayed in light color, with the recommended learning order marked.

[0060] Works that match the current knowledge points are selected from the student historical work library and displayed using the ability gradient principle: basic cases are qualified works that demonstrate the minimum ability requirements; advanced cases are excellent works that reach the current ability level; and challenge cases are demonstration works that are slightly higher than the current ability.

[0061] Step S54: Determine the display weights of the two units according to the display mode type: Career-enhancing type: Career-oriented units account for 70%, and ability adaptation units account for 30%; Basic consolidation type: ability adaptation units account for 70%, and career-oriented units account for 30%; Balanced development type: 50% each.

[0062] Design three display layout templates: left and right column type (suitable for PC), top and bottom scrolling type (suitable for mobile devices) and card switching type (suitable for interactive devices), and automatically select the optimal layout according to the user's device type; add a dynamic switch button between two units, add pop-up detail windows for key knowledge points and cases, embed learning progress tracking components, and update ability improvement data in real time.

[0063] In a preferred embodiment of the present invention, the above step S6: generating an interactive recommendation list based on the multimodal presentation content may include: Based on multimodal display content, the prerequisite relationship weights in the course knowledge graph, the interest tags in the student portrait, and the priority data in the correction parameters are integrated to generate an interactive recommendation list.

[0064] In this embodiment of the present invention, by integrating the four dimensions of ability matching, career relevance, interest matching, and prerequisite weight, a three-dimensional recommendation dimension is achieved, preventing the one-sidedness of a single indicator. Dynamic adjustment functions and visual decision-making assistance enhance the personalized interactive experience and reduce blind course selection. Real-time collection of student interaction behavior data forms a closed loop of "behavioral data-model optimization-precise recommendation", achieving a closed-loop teaching feedback loop. The introduction of prerequisite weights recommends courses based on knowledge progression logic and annotates the preceding and following relationships, helping students build a systematic knowledge system and realize intelligent learning paths.

[0065] In an embodiment of the present invention, the specific steps include: Extract the prerequisite course dependency of each course from the course knowledge graph (for example, if Course B requires Course A as a prerequisite, then the prerequisite weight of Course B = the mastery of Course A × 0.8); semantically match the interest tags in the student portrait (such as sustainable design) with the course content keywords (such as course title, introduction) to generate an interest matching degree (0-100 points).

[0066] Based on the priority label of the display mode (such as career-enhancing and foundation-building), recommendation weights are assigned to different types of courses (for example, career-oriented courses have a weight of +30%). For each course in the candidate course set, the comprehensive recommendation score is calculated based on the following weighted dimensions: Ability matching (from student portraits): 40%; Career relevance (from student portraits): 30%; Interest matching: 20%; Prerequisite relationship weight: 10%.

[0067] For example: If the ability matching degree of a course is 85 points, the career relevance degree is 90 points, the interest matching degree is 75 points, and the prerequisite weight is 0.9, then the comprehensive score = 85×0.4+90×0.3+75×0.2+0.9×10=84.9 points.

[0068] Courses are sorted from high to low by comprehensive scores to form a basic recommendation sequence. A capability gap analysis button is added under each course entry. Clicking it will display the differences between students' current capabilities and course requirements (such as "need to supplement advanced CAD skills"). A career path preview button is added for advanced courses, which is linked to the VR scene resource library to display the corresponding career scene clips of the course (for example, click on "Principles of Landscape Planning" to watch a 1-minute VR animation of garden design). An intelligent adjustment slider is set up so that students can manually adjust the recommendation focus (for example, sliding left to increase the interest priority weight, sliding right to increase the career priority weight).

[0069] like Figure 2 As shown, an embodiment of the present invention further provides an environmental art design course recommendation system, comprising: The extraction module is used to construct the knowledge graph of the environmental art design course, extract course entities, knowledge point entities and interdisciplinary associations, establish prerequisite and knowledge connection relationships between courses, and generate a dynamic course knowledge graph containing course basic requirements and high-level ability labels; The matching module is used to collect student interests, ability test and career planning data based on the dynamic course knowledge graph. The matching model calculates the correlation between ability and basic course requirements, and between career goals and advanced course abilities, generating a multi-dimensional student portrait. The screening module is used to select candidate courses that are suitable for students' current abilities based on the ability matching degree in the multi-dimensional student profile and the prerequisite relationships between courses in the dynamic course knowledge graph; The correction module is used to set two target points in the candidate course set: the highest match between the basic course requirements and student abilities, and the highest correlation between the advanced course abilities and career planning. It also constructs a linear evaluation axis, calculates the offset angle between the axis direction and the reference direction, and the midpoint slope, and generates course presentation correction parameters. A dynamic adjustment module is used to dynamically adjust the multimodal display content of the candidate course set according to the course display correction parameters; The interactive module is used to generate interactive recommendation lists based on multimodal display content.

[0070] It should be noted that this system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0071] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for recommending environmental art design courses, characterized in that: The method comprises: Step S1: Construct the knowledge graph of the environmental art design course, extract course entities, knowledge point entities and interdisciplinary associations, establish prerequisite and knowledge connection relationships between courses, and generate a dynamic course knowledge graph containing course basic requirements and high-level ability labels; Step S2: Based on the dynamic course knowledge graph, collect student interests, ability test and career planning data, and generate a multi-dimensional student portrait by matching the model to calculate the correlation between ability and basic course requirements, and between career goals and advanced course abilities; Step S3: Based on the ability matching degree in the multi-dimensional student portrait and the prerequisite relationship between courses in the dynamic course knowledge graph, select the candidate course set that is suitable for the student's current ability; Step S4: Set two target points in the candidate course set: the highest matching value between the basic course requirements and student abilities, and the highest correlation value between the advanced course abilities and career planning; construct a linear evaluation axis, calculate the offset angle and midpoint slope of the axis direction from the reference direction, and generate course display correction parameters; Step S5: Dynamically adjust the multimodal display content of the candidate course set according to the course display correction parameters; Step S6: Generate an interactive recommendation list based on the multimodal presentation content.

2. The environmental art design course recommendation method according to claim 1, characterized in that: Step S1 specifically includes: Extracting course entities from the course syllabus, teaching resource library and teacher files, wherein the course entities include course name, credits and teaching method attributes; Based on the course content text in the course entity, knowledge point entities are extracted through semantic analysis, and the logical association relationship between knowledge points is established based on the semantic similarity calculation of knowledge point entities to form the core knowledge chain of the knowledge graph; Based on the knowledge point entities and their logical associations in the core knowledge chain, combined with the prerequisite course declarations in the teaching plan corresponding to the course entity, the mastery of the knowledge point entities in the student's historical learning data is analyzed, the prerequisite dependency strength and knowledge connection weight between courses are calculated, and the hierarchical relationship between course nodes is constructed; Based on the course nodes in the course hierarchy, we extract the data of cooperating teachers in interdisciplinary course projects, associate them with the subject tags of student works corresponding to the knowledge point entities, and establish the association relationship between interdisciplinary courses to obtain interdisciplinary association data. Based on cross-disciplinary related data, course entity attributes, knowledge point associations, and prerequisite dependency strength, when new courses are added or teachers adjust teaching content, the entities and relationships in the knowledge graph are dynamically and synchronously updated to generate a dynamic course knowledge graph containing basic course requirements and advanced ability labels.

3. The environmental art design course recommendation method according to claim 2, characterized in that: Step S2 specifically includes: Based on the basic course requirements and high-level competency labels in the dynamic course knowledge graph, we collect student data sets including interest preference data, career planning goals, and competency test results. Based on the ability test results, the matching model is used to calculate the matching degree between the student's ability and the basic course requirements in the dynamic course knowledge map. At the same time, based on the career planning goals, the semantic relevance between them and the high-level ability labels in the course knowledge map is calculated. The matching degree and semantic relevance are integrated with the interest preference data in a multi-dimensional weighted manner to generate a multi-dimensional student portrait that includes quantitative values ​​of ability levels, career orientation labels and interest feature vectors.

4. The environmental art design course recommendation method according to claim 3, characterized in that: Step S3 specifically includes: Based on the ability matching degree of the multi-dimensional student portrait, the course nodes with matching degree exceeding the preset threshold are screened out from the dynamic course knowledge graph to generate a basic set of candidate courses that are initially adapted to the student's current ability level; Based on the basic candidate course set and combined with the prerequisite relationships between courses in the dynamic course knowledge graph, we remove course nodes where students do not meet the prerequisite knowledge connection weights, and generate a refined candidate course set that meets the knowledge coherence requirements. Based on the refined candidate course set, high-level ability labels associated with career planning goals in student profiles are extracted, and high-level course nodes matching the labels are retained to form a hybrid candidate course set that is adapted to students' current abilities. The hybrid candidate course set includes basic courses and advanced courses, and the ability improvement paths corresponding to the advanced courses are marked.

5. The environmental art design course recommendation method according to claim 4, characterized in that: Step S4 specifically includes: Set two target points in the candidate course set. The first target point is the maximum value of the match between the basic requirements of the course and the students' abilities. The second target point is the maximum value of the correlation between the advanced abilities of the course and career planning. A linear evaluation axis is constructed with the two target points as endpoints, the offset angle of the axis direction relative to the preset reference direction is calculated, and the slope value of the axis midpoint is extracted; According to the combination relationship between the offset angle and the slope value, the course display correction parameters are generated.

6. The environmental art design course recommendation method according to claim 5, characterized in that: Step S5 specifically includes: Determine the display mode type of the candidate course set according to the offset angle and slope value in the course display correction parameter, and generate a display mode instruction set including a priority label; If the priority label in the display mode instruction set points to the career-related direction and the slope exceeds the threshold, the interdisciplinary VR practice scenario resource library is called, and based on the high-level course nodes in the candidate course set, VR scenes and related high-level cases that match the career goals are rendered to generate career-oriented units in the multimodal display content; If the priority label in the display mode instruction set points to the ability adaptation direction and the slope is lower than the threshold, the knowledge point map visualization engine is activated to extract the basic course nodes and their knowledge point logic chains from the candidate course set, generate a dynamic visualization map, and associate the work cases in the student's historical portfolio that match the current ability for prominent display, thus generating the ability adaptation unit in the multimodal display content; Based on the ability adaptation unit and career orientation unit, they together constitute the dynamically adjusted multimodal display content.

7. The environmental art design course recommendation method according to claim 6, characterized in that: Step S6 specifically includes: Based on multimodal display content, the prerequisite relationship weights in the course knowledge graph, the interest tags in the student portrait, and the priority data in the correction parameters are integrated to generate an interactive recommendation list.

8. An environmental art design course recommendation system, which implements the method according to any one of claims 1 to 7, characterized in that: include: The extraction module is used to construct the knowledge graph of the environmental art design course, extract course entities, knowledge point entities and interdisciplinary associations, establish prerequisite and knowledge connection relationships between courses, and generate a dynamic course knowledge graph containing course basic requirements and high-level ability labels; The matching module is used to collect student interests, ability test and career planning data based on the dynamic course knowledge graph. The matching model calculates the correlation between ability and basic course requirements, and between career goals and advanced course abilities, generating a multi-dimensional student portrait. The screening module is used to select candidate courses that are suitable for students' current abilities based on the ability matching degree in the multi-dimensional student profile and the prerequisite relationships between courses in the dynamic course knowledge graph; The correction module is used to set two target points in the candidate course set: the highest matching value between the basic requirements of the course and the students' abilities, and the highest correlation value between the advanced abilities of the course and career planning; Construct a linear evaluation axis, calculate the offset angle and midpoint slope between the axis direction and the reference direction, and generate course display correction parameters; A dynamic adjustment module is used to dynamically adjust the multimodal display content of the candidate course set according to the course display correction parameters; The interactive module is used to generate interactive recommendation lists based on multimodal display content.

Images