Scene perception learning evaluation method and system
Through the learning evaluation method of scene-awareness, combined with qualitative and quantitative scoring, an immersive evaluation scenario is constructed, which solves the problem that traditional evaluation methods are difficult to comprehensively measure learners' multi-dimensional ability, and achieves a more accurate and personalized evaluation experience.
Patent Information
- Application Number
- CN202410901599.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-05
- Publication Date
- 2024-12-20
AI Technical Summary
The existing learning evaluation methods in the field of education mainly rely on traditional test paper methods, and it is difficult to comprehensively and accurately measure learners' multi-dimensional ability and comprehensive literacy.
Using a scene-aware learning evaluation method, by constructing an evaluation scenario creation module including knowledge goals, knowledge difficulty, knowledge points and evaluation scenario correlation, setting up scene roles to participate in immersive evaluation, combining qualitative and quantitative scoring, to achieve the level of mastering learners’ knowledge. Comprehensive and accurate evaluation.
It provides a more intuitive and interesting evaluation experience, significantly improves the accuracy and scientificity of evaluation, ensures the continuity and pertinence of evaluation, and promotes the continuous progress of learners.
Smart Images
Figure CN119168436A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent teaching technology, and in particular to a scene-aware learning evaluation method and system thereof. Background Art
[0002] At present, the reliance on learning evaluation methods in the education field is indeed largely concentrated on the traditional examination method, which uses a preset set of questions to test students' memory, understanding and application of specific knowledge points, and then quantifies learning outcomes in the form of scores. However, with the renewal of educational concepts and the rapid development of educational technology, more and more educators have realized that this single evaluation method has significant limitations and is difficult to comprehensively and accurately measure learners' multi-dimensional abilities and comprehensive qualities.
[0003] Therefore, it is necessary to develop a new scene-aware learning evaluation method and system. Summary of the invention
[0004] In view of the deficiencies of the above-mentioned related technologies, the purpose of the present invention is to provide a scene-aware learning evaluation method and system to bring learners a better evaluation experience, provide an immersive evaluation method, and more accurately evaluate the learners' knowledge mastery level.
[0005] In a first aspect, a scene perception learning evaluation method according to the present invention comprises the following steps:
[0006] W1. Construct an evaluation scenario creation module including knowledge objectives, knowledge difficulty, knowledge points and evaluation scenarios; wherein the evaluation scenario reflects an immersive evaluation environment including at least the interaction of people, time, space and learning content;
[0007] W2. Setting a knowledge goal with a knowledge difficulty of A, matching the corresponding knowledge points based on the knowledge goal with a knowledge difficulty of A, and generating an evaluation scenario corresponding to the knowledge points;
[0008] W3. Set up scenario roles and make them participate in the current evaluation scenario; scenario roles include learners and interactors;
[0009] W4. The learner completes the learning of the current evaluation scenario, and the interactor evaluates the learner's learning situation and gives an evaluation score;
[0010] W5. Determine whether the evaluation score reaches the preset ability threshold; if so, upgrade the knowledge difficulty to A', A'>A, match new knowledge points and generate new evaluation scenarios, and go to step W3; otherwise, display the evaluation score.
[0011] Optionally, the step W1 is specifically as follows: constructing an evaluation scenario creation module including knowledge objectives, knowledge difficulty, knowledge points, feature nodes, scenario instance nodes and evaluation scenario associations; wherein the feature nodes reflect the scenario features of the corresponding knowledge points; the scenario instance nodes represent the scenario keywords corresponding to the scenario features; the scenario keywords are used to retrieve the corresponding scenario elements in the scenario resource library; and the evaluation scenario is constructed based on the scene elements retrieved from the scenario resource library.
[0012] Optionally, the step W2 is specifically as follows: selecting a knowledge target with a knowledge difficulty of A, matching corresponding knowledge points, feature nodes and scenario instance nodes according to the knowledge target with a knowledge difficulty of A, retrieving corresponding scene elements from a scenario resource library according to the scenario instance nodes, and constructing the evaluation scenario based on the scene elements.
[0013] Optionally, the knowledge point includes a plurality of knowledge nodes, which are a knowledge target root node, a plurality of knowledge subject nodes arranged under the knowledge target root node, and a plurality of knowledge concept nodes arranged under each of the knowledge subject nodes;
[0014] The knowledge node is in the form of a quintuple (S id , S name , S level , S dif , S num ) description, where S id is the identifier of the knowledge node; S name is the keyword of the knowledge point; S level is the layer number of the knowledge node; S dif is the knowledge difficulty of the knowledge node; S num is the total number of characteristic nodes of the knowledge node;
[0015] Each of the knowledge concept nodes corresponds to a plurality of feature nodes, and each feature node corresponds to a plurality of scene instance nodes;
[0016] The characteristic node adopts the form of a quaternion (C id , C name , C sid , C num ) description, where C id is the identifier of the feature node, C name is a keyword describing the feature node, C sid is the knowledge concept node number associated with the feature node, C num Describe the total number of scene instance nodes that can reflect the feature node;
[0017] The scene instance node is in the form of a triple (Iid , I name , I cid ) description; wherein, I id is the identifier of the scene instance node, I name is a keyword describing the scene instance node, I cid It is the feature node number associated with the scene instance node.
[0018] Score=(Score 定量 +Score 定性 ) / 2;
[0019] Score 定量 =α1+α2+…+α i +…+α k ;
[0020] Score 定性 =β1+β2+…+β i +…+β k ;
[0021] β i =λ*g 自评 +(1-λ)*g 他评 ;
[0022] Among them, Score represents the evaluation score; Score 定量 Indicates the comprehensive quantitative score, score 定量 The higher the value, the better the learner's knowledge and ability are; 定性 represents the comprehensive qualitative score; α i represents the quantitative score of the i-th evaluation scenario, which is the score obtained based on the number of correct answers; β i represents the qualitative evaluation of the i-th evaluation scenario, and the qualitative evaluation is the grade score; k is the total number of evaluation scenarios; λ represents the weight of self-evaluation; 1-λ represents the weight of other evaluations; g 自评 Indicates the learner's rating of his or her own performance; g 他评 It means that the interactor rates the learner's performance.
[0023] Optionally, the creator of the evaluation scenario creation module and the interactor are both teachers.
[0024] Optionally, the scene resource library needs to be updated incrementally on a regular basis, with the update goal being to search based on new keywords and expand existing keywords.
[0025] Optionally, the evaluation scenario is constructed using VR or AR.
[0026] In a second aspect, a scene-aware learning evaluation system according to the present invention adopts the scene-aware learning evaluation method described above, comprising:
[0027] Evaluation scenario creation module, used for generating evaluation scenarios;
[0028] A scene role generation module, used for generating scene roles;
[0029] The scoring and logic judgment module is used to evaluate the learner's learning situation, determine whether the evaluation score reaches the preset ability threshold, send an instruction to create an evaluation scene, and send an instruction to display the evaluation score;
[0030] The evaluation scene creation module and the scene role generation module are respectively connected to the scoring and logic judgment module, and the scene role generation module is connected to the evaluation scene creation module.
[0031] Optionally, it also includes a scenario resource library, which is connected to the evaluation scenario creation module.
[0032] The present invention has the following advantages:
[0033] The scene-aware learning evaluation method and system described in the present invention, by deeply integrating knowledge points and the construction of evaluation scenes, carefully creates specific scene content, and provides learners with an unprecedented immersive evaluation experience. This method not only enhances the intuitiveness and fun of the evaluation, but also significantly improves the accuracy of evaluating the learner's knowledge mastery level. When the learner's score reaches the preset ability threshold, the system can automatically upgrade the evaluation scene intelligently and achieve smooth switching between scenes based on the scene-aware tree logic, thereby building a highly personalized adaptive learning evaluation environment. This mechanism ensures the continuity and pertinence of the evaluation and promotes the continuous progress of learners.
[0034] In addition, the present invention adopts an innovative scoring method that combines qualitative and quantitative methods to achieve a comprehensive and accurate evaluation of the learner's knowledge mastery level. At the same time, by setting up a multi-level structure such as knowledge theme nodes, knowledge concept nodes, scene feature nodes, and scene instance nodes, the present invention closely links the complex knowledge system with rich scene elements, providing detailed and specific data support for the evaluation, further enhancing the scientificity and effectiveness of the evaluation. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 This is a flow chart of a learning and evaluation method for scene perception according to an embodiment of the present invention;
[0036] Figure 2 A schematic diagram of the relationship between constructing an evaluation scenario according to an embodiment of the present invention;
[0037] Figure 3 A schematic diagram of a scene-aware learning evaluation system according to an embodiment of the present invention.
[0038] Explanation of the accompanying drawings: 1-evaluation scene creation module; 2-scene role generation module; 3-scoring and logical judgment module. DETAILED DESCRIPTION
[0039] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thoroughly understood.
[0040] See also Figure 1 As shown, the embodiment of the present invention discloses a scene-aware learning evaluation method, the steps comprising:
[0041] W1. Construct an evaluation scenario creation module that includes knowledge objectives, knowledge difficulty, knowledge points, and evaluation scenarios in sequence; wherein the evaluation scenario reflects an immersive evaluation environment that at least includes the interaction of people, time, space, and learning content. Among them, time and space refer to the spatiotemporal space of scenario construction.
[0042] W2. Set the knowledge goal with knowledge difficulty A, match the corresponding knowledge points based on the knowledge goal with knowledge difficulty A, and generate evaluation scenarios for the corresponding knowledge points.
[0043] W3. Set up scenario roles, including learners and interactors, and involve them in the current evaluation scenario.
[0044] W4. The learner completes the learning of the current evaluation scenario, and the interactor evaluates the learner's learning situation and gives an evaluation score;
[0045] W5. Determine whether the evaluation score reaches the preset ability threshold; if so, upgrade the knowledge difficulty to A', and A'>A, match new knowledge points and generate new evaluation scenarios, and go to step W3; otherwise, display the evaluation score.
[0046] By closely linking knowledge points with evaluation scenarios and creating specific scenario content, we can provide learners with a better evaluation experience and provide an immersive evaluation method that can more accurately evaluate learners' knowledge mastery level. If the score reaches the preset capability threshold, the evaluation scenario is upgraded. Based on the scenario perception tree, the evaluation scenario is automatically constructed, and the scenario is switched reasonably, providing a personalized and adaptive evaluation environment that is more intelligent and enables accurate evaluation. If the evaluation score reaches the preset capability threshold, it means that the evaluation scenario needs to be converted. You can also optimize the evaluation scenario by modifying the scenario perception tree, including the knowledge subject node, feature node, and scenario instance node, to enrich the evaluation scenario.
[0047] In a possible embodiment, the knowledge goal may be a course or a specific knowledge goal. The scene roles are set according to the evaluation scene content to enhance the immersion in the evaluation process.
[0048] Learner scenario role: Similar to player roles (AvAars), learners choose their favorite digital avatars to represent themselves, and the digital avatars complete the assessment tasks in the scenario.
[0049] Interactor scene role: Similar to NPC (non-player character), the teacher or learning goal editor can freely set the interaction strategy. For example, in the English dialogue session, the role can be set to communicate with the learner.
[0050] In the embodiment of the present application, step W1 is specifically as follows:
[0051] Construct an evaluation scenario creation module that includes knowledge objectives, knowledge difficulty, knowledge points, feature nodes, scenario instance nodes and evaluation scenarios that are sequentially associated; wherein the feature nodes reflect the scenario features of the corresponding knowledge points; the scenario instance nodes represent the scenario keywords corresponding to the scenario features; the scenario keywords are used to retrieve the corresponding scenario elements in the scenario resource library; and the evaluation scenario is constructed based on the scenario elements retrieved from the scenario resource library.
[0052] In a possible embodiment, the scene resource library is mainly used to store pictures, keywords and addresses; that is, it is maintained through a data table, which contains numbers, keywords, and resource storage addresses. For example, if the keyword for the query is "electrical appliances", all the image addresses related to electrical appliances will be found and then extracted.
[0053] In a possible embodiment, feature nodes are set according to specific knowledge target requirements and are conceptual words that best reflect scene features. For example, scene features include scene location, terrain, scenic spots, folk customs, etc. According to the scene instance node, the corresponding scene resources, such as 3D models or images, are detected in the scene resource library. The initial resource library is obtained by crawling and keyword retrieval URLs. Based on the acquired resources, the perspective is set to organize the scene to complete the initial scene construction.
[0054] In a possible embodiment, the step W2 is specifically as follows:
[0055] Select a knowledge goal with a knowledge difficulty of A, match the corresponding knowledge points, feature nodes and scene instance nodes according to the knowledge goal with a knowledge difficulty of A, retrieve the corresponding scene elements from the scene resource library according to the scene instance nodes, and construct the evaluation scene based on the scene elements.
[0056] In a possible embodiment, the knowledge difficulty is upgraded to A', A'>A, and new knowledge points are matched and new evaluation scenarios are generated. Specifically, the following steps are performed:
[0057] Upgrade the knowledge difficulty to A', and A'>A, and match the corresponding knowledge points, feature nodes and scenario instance nodes, retrieve the corresponding scenario elements from the scenario resource library according to the scenario instance nodes, and build an evaluation scenario based on the scenario elements.
[0058] like Figure 2 As shown, in a possible embodiment, the knowledge point includes multiple knowledge nodes, namely a knowledge target root node, multiple knowledge subject nodes arranged under the knowledge target root node, and multiple knowledge concept nodes arranged under each knowledge subject node.
[0059] In a possible embodiment, the knowledge node is in the form of a quintuple (S id , S name , S level , S dif , S num ) description, where S id is the identifier of the knowledge node; S name is the keyword of the knowledge point; S level is the layer number of the knowledge node; S dif is the knowledge difficulty of the knowledge node; S num is the total number of feature nodes of the knowledge node. For example, see Figure 2 As shown in the figure, the knowledge target root node is identified as S0, the number of layers is 1, and the total number is 1. The knowledge subject node is identified as S n ; Where n≥1, n is an integer; the number of layers is 2. The identifier of the knowledge concept node is S 10n+m, m≥1, m is an integer, and the number of layers is 3.
[0060] like Figure 2 As shown, in a possible embodiment, each knowledge concept node corresponds to multiple feature nodes, and each feature node corresponds to multiple scene instance nodes. Knowledge subject nodes, knowledge concept nodes, scene feature nodes, and scene instance nodes are set to closely link the knowledge points with the evaluation scene elements.
[0061] In a possible embodiment, the feature node is in the form of a quadruple (C id , C name , C sid , C num ) description, where C id is the identifier of the feature node, C name is a keyword describing a feature node, C sid is the knowledge concept node number associated with the feature node, C num Describes the total number of scene instance nodes that can reflect this feature node.
[0062] In a possible embodiment, the scene instance node is in the form of a triple (I id , I name , I cid ) description; wherein, I id I is the identifier of the scene instance node; name It is a keyword describing the scene instance node; cid It is the feature node number associated with the scene instance node.
[0063] like Figure 2 As shown, in a possible embodiment, the evaluation scenario creation module includes a knowledge layer, a feature layer and a scenario layer, wherein the knowledge layer includes a knowledge target root node (S0), knowledge subject nodes (S1 and S2) and knowledge concept nodes (S11 to S16). The feature layer includes feature nodes (C1 to C12); the scenario layer includes scenario instance nodes (I1 to I24). The scenario layer is composed of scenario instance nodes, which describe scenario keywords related to a given feature. For example, C1 is described by two keywords, I1 and I2. The scenario keywords can directly retrieve the corresponding scenario elements in the resource library.
[0064] In one example, take the course book "Safety Education for College Students" published by People's Posts and Telecommunications Press as an example:
[0065]
[0066]
[0067] In a possible embodiment, the calculation formula of the evaluation score is:
[0068] Score=(Score 定量 +Score 定性 ) / 2;
[0069] Score 定量 =α1+α2+…+α i +…+α k ;
[0070] Score 定性 =β1+β2+…+β i +…+β k ;
[0071] β i =λ*g 自评 +(1-λ)*g 他评 ;
[0072] Among them, Score represents the evaluation score; Score 定量 Indicates the comprehensive quantitative score, score 定量 The higher the value, the better the learner's knowledge and ability are; 定性 represents the comprehensive qualitative score; α i represents the quantitative score of the i-th evaluation scenario, which is the score obtained based on the number of correct answers; β i represents the qualitative evaluation of the i-th evaluation scenario, and the qualitative evaluation is the grade score; k is the total number of evaluation scenarios; λ represents the weight of self-evaluation; 1-λ represents the weight of other evaluations; g 自评 Indicates the learner's rating of his or her own performance; g 他评 Indicates that the interactor scores the learner's performance. A scoring method that combines qualitative and quantitative methods is used to evaluate the learner's knowledge level as accurately as possible.
[0073] Before quantitative scoring, set the questions and standard answers in the corresponding evaluation scenarios. The system directly calculates the scores based on the learners' answers. Note that since there may be more than one scenario, the total score is not a fixed value. The full score of each scenario is fixed, that is, the learner's quantitative evaluation score.
[0074] The grade includes multiple performance grades, each performance grade corresponds to a score, and the higher the grade evaluation, the higher the score. For example, the performance grades include very good, good, average and poor, and the score of the very good grade is 10 points, the score of the good grade is 8 points, the score of the average grade is 5 points, and the score of the poor grade is 3. The specific score corresponding to the performance grade is set according to the actual situation and is not limited here.
[0075] Score refers to the overall score of the learner. The higher the score, the better the learner's performance, the stronger the ability, and the more difficult the knowledge evaluation scene. The preset ability threshold for scene transition is set according to the score, but the determination of the preset ability threshold requires two factors: the performance score and the knowledge difficulty in the scene; the score value required for different knowledge difficulties is different. Generally speaking, the larger the score value, the more scene transition is needed.
[0076] In one possible embodiment, the creator and the interactor of the evaluation scenario creation module are the same, for example, both are teachers or parents or professionals or other types of people related to the learning content, which is not limited here. In other embodiments, the creator and the interactor of the evaluation scenario creation module have different roles. For example, the creator of the evaluation scenario creation module is the manager of the knowledge goal; the interactor is a teacher or a professional or other type of person related to the learning content, which is not limited here.
[0077] In a possible embodiment, the scene resource library needs to be updated incrementally on a regular basis, with the update goal being to retrieve based on new keywords and to expand existing keywords.
[0078] In a possible embodiment, the evaluation scene is constructed using VR or AR.
[0079] See also Figure 3 As shown, the present invention also discloses a scene-aware learning evaluation system, which adopts the scene-aware learning evaluation method, including an evaluation scene creation module 1, a scene role generation module 2 and a scoring and logic judgment module 3; the evaluation scene creation module 1 and the scene role generation module 2 are respectively connected to the scoring and logic judgment module 3; the scene role generation module 2 is connected to the evaluation scene creation module 1. Among them, the evaluation scene creation module 1 is used for the generation of evaluation scenes. The scene role generation module 2 is used for the generation of scene roles. The scoring and logic judgment module 3 is used to evaluate the learning situation of the learner, and to judge whether the evaluation score reaches the preset ability threshold, send an instruction to create an evaluation scene, and send an instruction to display the evaluation score.
[0080] In a possible embodiment, the evaluation scenario creation module 1 includes four parts: evaluation scenario generation, evaluation scenario conversion, evaluation scenario optimization and scenario resource management; wherein:
[0081] Evaluation scenario generation: Input knowledge objectives and knowledge difficulty (taking A as an example), match the corresponding knowledge nodes, obtain feature nodes and scenario instance nodes, and detect the corresponding scenario resources in the scenario resource library. Based on this, set the perspective, organize the scenario, and complete the initial scenario creation.
[0082] Evaluation scenario conversion: In the current evaluation scenario, when the learner's comprehensive score is higher than the preset ability threshold, the knowledge nodes connected to it with a value greater than A are retrieved according to the difficulty of the knowledge nodes in the scenario perception graph, and a new evaluation scenario is constructed based on its related scenario instance nodes.
[0083] Evaluation scenario optimization: responsible for the modification of the scenario perception tree, including the modification of knowledge nodes, feature nodes and scenario instance nodes, to enrich the evaluation scenario.
[0084] Scene resource management: responsible for retrieving scene resources related to scene instance nodes, including images or 3D models. The initial scene resources are retrieved using crawlers and keyword I name Retrieve the URL to obtain the initial resource library. Update it incrementally at regular intervals. Two strategies are used for updating: based on new keywords. name Search and expansion of existing keywords.
[0085] In a possible embodiment, the evaluation scene is constructed using virtual reality technology (VR) or augmented reality (AR). Virtual reality technology uses computers to create scenes that make users feel as if they are in the scene, while augmented reality creates a space that integrates the real and the virtual by superimposing virtual information in the real world. The support of VR and AR technology has changed the traditional single learning method of multimedia playback and text interaction only, allowing learners and teachers to perceive real experiences in the scene and deepen learners' understanding of knowledge points. The use of VR and AR technology in the evaluation scene has changed the traditional evaluation method and can better stimulate the learners' real level and other advantages.
[0086] In a possible embodiment, the scene-aware learning evaluation system has the following objectives:
[0087] Immersion: The evaluation scene constructed with VR or AR allows the evaluator to deeply participate in the scene and obtain the learner's true level.
[0088] Interactivity: refers to the relationship between scene objects, evaluation scenes and evaluation targets, which can interact naturally.
[0089] Guidance: Emphasis on the behavioral orientation provided by the evaluation scenario for learners. The evaluation scenario provides clues through the various artistic conceptions it contains. Learners are influenced by the evaluation scenario, which triggers the association, memory, and application of relevant knowledge.
[0090] Extensibility: The knowledge content of the evaluation will change, and the system provides an expansion interface to facilitate subsequent knowledge expansion.
[0091] Adaptability: Establish personalized and adaptive evaluation scenarios based on learners’ abilities. The scenarios perceive the evaluator’s ability level and can automatically upgrade or switch to ensure that scenario-based accurate evaluation is effective and achieve real-time dynamic matching between knowledge and ability evaluation and scenarios.
[0092] In a possible embodiment, a scenario-aware learning evaluation system further includes a scenario resource library, which is connected to the evaluation scenario creation module.
[0093] The scene-aware learning evaluation method and system of the present invention, by closely linking knowledge points with evaluation scene construction and creating specific scene content, brings a better evaluation experience to learners, provides an immersive evaluation method, and can more accurately evaluate the learner's knowledge mastery level; if the score reaches the preset ability threshold, the evaluation scene is upgraded, and based on the scene perception tree, the evaluation scene is automatically constructed, and the scene switching is performed reasonably, providing a personalized and adaptive evaluation environment, which is more intelligent and achieves accurate evaluation. A scoring method that integrates qualitative and quantitative methods is adopted to evaluate the learner's knowledge mastery level as accurately as possible. Knowledge subject nodes, knowledge concept nodes, scene feature nodes, and scene instance nodes are set to closely link the knowledge points with the evaluation scene elements.
[0094] In the description of the embodiments of the present invention, “plurality” means two or more than two, unless otherwise clearly and specifically defined.
[0095] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "illustrative embodiments", "examples", "specific examples" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0096] It should be understood that the application of the present invention is not limited to the above examples. For those skilled in the art, improvements or changes can be made according to the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention. Those skilled in the art can understand that all or part of the processes of the above embodiments can be implemented, and equivalent changes made according to the claims of the present invention still fall within the scope of the present invention.
Claims
1. A scene-aware learning evaluation method, characterized in that the steps include: W1. Construct an evaluation scenario creation module including knowledge objectives, knowledge difficulty, knowledge points and evaluation scenarios; wherein the evaluation scenario reflects an immersive evaluation environment including at least interaction of people, time, space and learning content; W2. Setting a knowledge goal with a knowledge difficulty of A, matching corresponding knowledge points based on the knowledge goal with a knowledge difficulty of A, and generating evaluation scenarios for the corresponding knowledge points; W3. Set up scenario roles and make them participate in the current evaluation scenario; wherein the scenario roles include learners and interactors; W4. The learner completes the learning of the current evaluation scenario, and the interactor evaluates the learner's learning situation and gives an evaluation score; W5. Determine whether the evaluation score reaches the preset ability threshold; if so, upgrade the knowledge difficulty to A', and A'>A, match new knowledge points and generate new evaluation scenarios, and go to step W3; otherwise, display the evaluation score.
2. The scene perception learning evaluation method according to claim 1, characterized in that: The step W1 is specifically as follows: constructing an evaluation scenario creation module including knowledge objectives, knowledge difficulty, knowledge points, feature nodes, scenario instance nodes and evaluation scenario associations; wherein the feature nodes reflect the scenario features of the corresponding knowledge points; the scenario instance nodes represent the scenario keywords corresponding to the scenario features; the scenario keywords are used to retrieve the corresponding scenario elements in the scenario resource library; and the evaluation scenario is constructed based on the scenario elements retrieved from the scenario resource library.
3. The scene perception learning evaluation method according to claim 2, characterized in that: The step W2 is specifically as follows: selecting a knowledge target with a knowledge difficulty of A, matching corresponding knowledge points, feature nodes and scenario instance nodes according to the knowledge target with a knowledge difficulty of A, retrieving corresponding scenario elements from the scenario resource library according to the scenario instance nodes, and constructing the evaluation scenario based on the scenario elements.
4. The scene perception learning evaluation method according to claim 2, characterized in that: The knowledge point includes a plurality of knowledge nodes, which are a knowledge target root node, a plurality of knowledge subject nodes arranged under the knowledge target root node, and a plurality of knowledge concept nodes arranged under each of the knowledge subject nodes; The knowledge node is in the form of a quintuple (S id , S name , S level , S dif , S num ) description, where S id is the identifier of the knowledge node; S name is the keyword of the knowledge point; S level is the layer number of the knowledge node; S dif is the knowledge difficulty of the knowledge node; S num is the total number of feature nodes of knowledge nodes; A plurality of feature nodes are arranged under each of the knowledge concept nodes, and a plurality of scene instance nodes are arranged under each feature node; The characteristic node is in the form of a quaternion (C id , C name , C sid , C num ) description, where C id is the identifier of the feature node, C name is a keyword describing the feature node, C sid is the knowledge concept node number associated with the feature node, C num Describe the total number of scene instance nodes that can reflect the feature node; The scene instance node is in the form of a triple (I id , I name , I cid ) description; where I id is the identifier of the scene instance node, I name is a keyword describing the scene instance node, I cid It is the feature node number associated with the scene instance node.
5. The scene perception learning and evaluation method according to any one of claims 1 to 4, characterized in that: The calculation formula of the evaluation score is: Score = (Score 定量 + Score 定性 ) / 2; Score 定量 =α1+α2+…+α i +…+a k ; Score 定性 =β1+β2+…+β i +…+b k ; b i =λ*g 自评 +(1-λ)*g 他评 ; Among them, Score represents the evaluation score; Score 定量 Indicates the comprehensive quantitative score, score 定量 The higher the value, the better the learner's knowledge and ability are; 定性 represents the comprehensive qualitative score; α i represents the quantitative score of the i-th evaluation scenario, which is the score obtained based on the number of correct answers; β i represents the qualitative evaluation of the i-th evaluation scenario, and the qualitative evaluation is the grade score; k is the total number of evaluation scenarios; λ represents the weight of self-evaluation; 1-λ represents the weight of other evaluations; g 自评 Indicates the learner's rating of his or her own performance; g 他评 It means that the interactor rates the learner's performance.
6. The scene perception learning and evaluation method according to any one of claims 1 to 4, characterized in that: The creator of the evaluation scenario creation module and the interactor are both teachers.
7. The scene perception learning evaluation method according to claim 2 or 3, characterized in that: The scene resource library needs to be updated regularly and incrementally, and the update goal is to search based on new keywords and expand existing keywords.
8. The scene perception learning evaluation method according to claim 1, characterized in that: The evaluation scenario is constructed using VR or AR.
9. A scene-aware learning evaluation system, characterized in that: A learning and evaluation method for scene perception as claimed in any one of claims 1 to 8, comprising: Evaluation scenario creation module, used for generating evaluation scenarios; A scene role generation module, used for generating scene roles; The scoring and logic judgment module is used to evaluate the learner's learning situation, determine whether the evaluation score reaches the preset ability threshold, send an instruction to create an evaluation scene, and send an instruction to display the evaluation score; The evaluation scene creation module and the scene role generation module are respectively connected to the scoring and logic judgment module, and the scene role generation module is connected to the evaluation scene creation module.
10. The scene-aware learning evaluation system according to claim 9, characterized in that: It also includes a scene resource library, which is connected to the evaluation scene creation module.