A method of precision farming design

Abstract

The application provides a method for accurate homework design, comprising the following steps: S1, receiving homework information input by a user; S2, constructing a directed acyclic graph according to a homework range input by the user and a teaching knowledge graph; and S3, calculating and outputting a key node set as a homework arrangement point according to a homework target and a homework amount input by the user. Through the method, the accuracy and pertinence of homework arrangement point selection are effectively improved according to the homework range and target.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of smart education, and particularly relates to a method for precise homework design. BACKGROUND

[0002] As an important part of the learning process, homework has always been the focus of attention in the field of education research and practice. Since ancient times, there has been the saying "learn and practice at the right time" and "review the past and know the new", among which "practice at the right time" and "review the past" are the embryonic form of homework. For a long time, homework has always been responsible for the key task of testing the quality of classroom teaching and consolidating students' knowledge and skills, and provides an inexhaustible source of power for the improvement of teaching and the growth of students.

[0003] Homework refers to learning tasks arranged for students to complete outside of teaching time in accordance with certain teaching objectives, and its functions in education and teaching are usually as follows: (1) as an indispensable learning activity: students practice the knowledge learned in class through homework, thereby promoting the internalization of subject knowledge and continuously improving their ability level; (2) as an important window reflecting the effectiveness of teaching: analyzing homework data using appropriate methods can help schools and teachers understand the learning situation in a timely manner, so as to adjust teaching strategies and achieve efficient and precise educational goals.

[0004] Currently, in schools, the common homework design method is as follows: teachers select relevant chapter exercises from existing learning resources (such as textbooks and exercise books) in combination with the knowledge learned in class, and uniformly arrange them for students to complete after class; after collecting the students' answers, the teachers complete the review by a certain means, and statistics the overall homework situation of the class. In recent years, with the continuous advancement of education informatization, some schools have begun to use online learning platforms to arrange homework, and after students have completed the homework, the machine automatically reviews the objective questions to reduce the workload of teachers, and automatically gathers the answer data to statistics the class learning situation.

[0005] Although the current operation mode is not limited to offline arrangement and manual correction, there are still some problems in the design of homework, especially in the selection of questions and the control of question quantity. On the one hand, the chapter homework designed in the learning resources usually contains questions of different difficulties and knowledge points, which may not be completely suitable for the actual situation of the class. If part of the questions do not match the teaching progress and the learning situation of the class, the effect of the homework will be greatly weakened, and the learning interest of the students will be reduced. On the other hand, education should guarantee the healthy growth of students' body and mind, especially in the era of "double reduction", more emphasis is placed on the balance between "quality" and "quantity" of homework. For example, the policy clearly states that the average completion time of junior high school written homework should not exceed 90 minutes. According to the calculation of five working days per week, the total homework time of a junior high school student per week should be controlled within 450 minutes; while the subjects involved in the junior high school stage reach nine, the rough calculation of the average after-school homework time of each subject is only about 50 minutes per week. In this case of limited time and quantity, if the homework questions are still arranged without selection, it will seriously limit the accuracy of homework in knowledge consolidation and learning situation diagnosis.

[0006] Therefore, there is an urgent need for a homework design mechanism that closely combines the actual needs of education and teaching, accurately selects and arranges homework questions, and fully plays the role of homework in consolidating knowledge, so as to improve the pertinence and effectiveness of homework. SUMMARY

[0007] In view of the above problems, the present application provides a method for precise homework design, which can adaptively select homework arrangement points based on education domain knowledge graph and user personalized information.

[0008] The present application provides a method for precise homework design, comprising the following steps:

[0009] S1, receiving homework information input by the user;

[0010] S2, constructing a directed acyclic graph according to the homework range input by the user and the teaching knowledge graph;

[0011] S3, calculating and outputting a key node set as the homework arrangement point according to the homework target and homework quantity input by the user.

[0012] Further, in step S1, the homework information includes homework quantity, homework range, homework target, wherein the homework range includes core concepts, and the homework target is "finding problems (diagnosis)", "improving advantages (consolidation)", and "overall investigation (learning situation analysis)".

[0013] Further, the step S2 comprises:

[0014] S21, extracting core concepts according to the homework range;

[0015] S22, acquire nodes corresponding to the core concept and nodes directly connected with the nodes in the educational knowledge graph, to form a node set V, and gather all directed edges with nodes in the node set V as the starting point and the ending point to form an edge set E, thereby constructing a directed acyclic graph G=(V, E) of the core concept.

[0016] Further, the teaching knowledge graph is a multi-entity teaching knowledge graph, which is constructed based on the curriculum standards of each subject of primary school, junior high school and high school issued by the Ministry of Education, wherein the entity is a knowledge point in the curriculum standard of each subject, and the attribute is a cognitive state of a learner to the entity.

[0017] Further, the teaching knowledge graph comprises: unit module entities, core concept entities, cognitive state attributes of learners on the core concepts, a containing relationship between the unit modules and the core concepts, a predecessor-successor relationship between the core concepts under the same unit module, a parallel relationship between the unit modules, and a predecessor-successor relationship between the core concepts in different unit modules.

[0018] Further, the step S3 comprises:

[0019] S31, based on the homework target, calculate the importance score of the node in the directed acyclic graph G;

[0020] S32, determine the node of homework arrangement according to the homework amount.

[0021] Further, in step S31, if the homework target is "finding problems (diagnosis)", the calculation formula of the importance score score(v) of the node v is as follows:

[0022]

[0023] Wherein, topo_Level(v) represents the topological order of node v in G, max(topo_Level) represents the maximum topological order in G, and master_Rate(v) represents the proportion of students who have mastered the core concept on node v;

[0024] Further, in step S31, if the homework target is "improving advantages (consolidation)", the calculation formula of the importance score score(v) of the node v is as follows:

[0025]

[0026] Wherein, topo_Level(v) represents the topological order of node v in G, max(topo_Level) represents the largest topological order in G, and master_Rate(v) represents the proportion of students who have mastered the core concept on node v.

[0027] Further, in step S31, if the job target is "overall investigation (analysis of learning situation)", the calculation formula of the importance score score(v) of node v is as follows:

[0028] score(v) = (1 - master_Rate(v)) x degree_out(v) + (master_Rate(v)) x degree_in(v)

[0029] Wherein, master_Rate(v) represents the proportion of students who have mastered the core concept on node v, degree_in(v) represents the in-degree of node v, and degree_out(v) represents the out-degree of node v.

[0030] Further, the calculation method of the topological order of the node comprises:

[0031] (1) Initialize the variable topo_Level = 1;

[0032] (2) Assign the topological order of the node with an in-degree of 0 in the directed acyclic graph to topo_Level;

[0033] (3) Delete the node with an in-degree of 0 in the directed acyclic graph, delete the edge starting from the node with an in-degree of 0, update the directed acyclic graph, and increase the value of topo_Level by 1;

[0034] (4) Repeat (2) and (3) until the graph is empty, thereby obtaining the topological order value of all nodes.

[0035] The present application has the advantages that the present application improves the existing job design and job arrangement method, in addition to considering the subjective constraint conditions of users in terms of job quantity, job target, knowledge range and the like, the topological shape, node position, connectivity and the like graph structure information of the teaching knowledge graph are also included in the job design method, thereby enhancing the accuracy and pertinence of the job arrangement point selection. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 It is a flowchart of a precise job design method according to an embodiment of the present application;

[0037] Figure 2 It is a schematic diagram of a teaching knowledge graph according to an embodiment of the present application;

[0038] Figure 3 A schematic diagram of a directed acyclic graph according to another embodiment of the application. DETAILED DESCRIPTION

[0039] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of the present application.

[0040] The present application provides a method for precise operation design, as shown in the figure, based on user input information, precise selection of key nodes in the teaching knowledge graph and arrangement of operation thereon. Specifically, the following steps are included: Figure 1

[0041] S1, receiving user input operation information;

[0042] S2, constructing a directed acyclic graph according to the user input operation range and the teaching knowledge graph;

[0043] S3, calculating and outputting a key node set as the arrangement point of the operation according to the user input operation target and operation amount.

[0044] In step S1, the operation information includes operation amount, operation range and operation target. Among them,

[0045] (1) The operation amount refers to the number of test tasks contained in one operation, for example, "a certain number of mathematics homework needs to contain 9 questions", and the operation amount is 9.

[0046] (2) The operation range refers to the set of teaching content mainly focused on by one operation, for example, "a certain number of mathematics homework mainly focuses on the learning situation of two core concepts of <definition of right triangle, Pythagorean theorem>".

[0047] (3) The operation target refers to the implementation purpose of this operation formulated by the teacher in combination with the classroom teaching target and the learning situation of the students. According to the "Opinions on Further Reducing the Burden of Homework and Extracurricular Training of Compulsory Education Students", homework should better "play the functions of homework diagnosis, consolidation, learning situation analysis, etc.", accordingly, the present application refers to "diagnosis, consolidation, learning situation analysis" three aspects to determine "find problems (diagnosis), enhance advantages (consolidation), overall survey (learning situation analysis)" three types of operation targets.

[0048] In step S2, it specifically includes: ​

[0049] S21, according to the working range, extract core concepts; for example, "a certain number of mathematics homework mainly focuses on the learning situation of two core concepts <definition of right triangle, Pythagorean theorem>", the core concepts extracted are <definition of right triangle> and <Pythagorean theorem> two entities (i.e. knowledge points);

[0050] S22, obtain the nodes corresponding to the core concepts in the educational knowledge graph, and the nodes directly connected with the core concept nodes, to form a node set V, and all directed edges with the nodes in the set V as the starting point and the ending point are converged together to form an edge set E, thereby constructing a directed acyclic graph G=(V,E) of the core concept.

[0051] The knowledge graph includes entities, attributes and relationships. The entity is the node in the knowledge graph, each entity represents a specific thing or concept in the real world, and in the education field, it can be knowledge, skill, exercise, course, etc.; the attribute represents the various internal characteristics of the entity, such as the course entity can have teaching goal, ability requirement, mastery level, etc.; the relationship is the edge in the knowledge graph, which represents the external connection between entities, such as containing, parallel, predecessor and successor, etc.

[0052] The teaching knowledge graph of the application is a multi-entity teaching knowledge graph, which is constructed by experts and machines based on the curriculum standards of primary school, junior high school and high school issued by the Ministry of Education.

[0053] The artificial construction means that experts read the curriculum standards, identify the core concepts, teaching sequence and other elements in the curriculum standard text, and put them into the entities, attributes and relationships of the teaching knowledge graph.

[0054] The machine automatic construction means that the natural language processing technology is used to automatically identify and extract the core concepts, teaching sequence and other elements from the curriculum standard text, and put them into the entities, attributes and relationships of the teaching knowledge graph.

[0055] The teaching knowledge graph in the application contains the following elements: entity (unit module, core concept), attribute (cognitive state of learners on core concept), relationship (containing relationship between unit module and core concept, predecessor-successor relationship between core concepts under the same unit module, parallel relationship between unit modules, predecessor-successor relationship between core concepts in different unit modules).

[0056] Among them, the unit module entity and the core concept entity are the knowledge points in the curriculum standards of various disciplines, such as the "number and formula" module in the "Compulsory Education Mathematics Curriculum Standard (2022 Edition)" is a unit module entity, and the content of this module is divided into "rational number", "real number" and "algebraic expression" three chapters, so these three chapters are core concept entities in the teaching knowledge graph, and the relationship between the "number and formula" unit module entity and the three core concept entities is "contains" and is directed. The attribute of the core concept entity is the learner's cognitive state, that is, the mastery level of the student group on the core concept, which can be further divided into "proportion of students who have mastered" attribute and "proportion of students who have not mastered" attribute, such as in a certain student group, the attribute of the core concept "real number" can be "proportion of students who have mastered = 60%" and "proportion of students who have not mastered = 40%", and the relationship between the entity and the attribute is constructed as the edge (also directed) of "mastered degree" and "not mastered degree". The requirement of the new curriculum standard for "real number" is "know that real numbers are composed of rational numbers and irrational numbers", from which the prerequisite knowledge of the core concept entity "real number" can be summarized as "rational number" entity and "irrational number" entity, and then the "predecessor-successor" relationship is constructed between the three entities, as shown in Figure 2 .

[0057] In one embodiment, the user input assignment range is "<determination of right triangle, Pythagorean theorem>", and the nodes and edges that meet the conditions are found in the neighborhood of the adjacent nodes of the two nodes in the graph to construct a directed acyclic graph G. As shown in Figure 3 , "right triangle" (v3), "square root" (v4) and "application of Pythagorean theorem" (v7) are directly connected to the nodes in the assignment range, so they and the connected edges (e3, e4, e5, e6) will be included in the graph G, and the remaining nodes and edges will not be included in the graph G because they are not directly connected to the core concept nodes in the assignment range. Therefore, the directed acyclic graph G constructed in this embodiment contains the node set <v3, v4, v5, v6, v7> and the directed edge set <e3, e4, e5, e6> between these nodes.

[0058] In step S3, specifically comprising:

[0059] S31, based on the assignment target, calculating the node importance score of the directed acyclic graph;

[0060] S32, determining the nodes of the assignment arrangement according to the assignment amount.

[0061] In step S31, for different assignment targets, there are different node evaluation principles, as shown in Table 1.

[0062] Table 1

[0063] Serial number Job target Node importance evaluation principle 1 Find problems (diagnosis) More basic core concepts are more important 2 Promote advantages (consolidate) More high-level core concepts are more important 3 Overall investigation (learning situation analysis) Learning situation information diagnosed by learning situation jobs needs to cover as many core concepts as possible

[0064] The importance score of each node in the directed acyclic graph G is calculated as follows:

[0065] (1) If the job target is "discovery of problems (diagnosis)", the importance score score(v) of node v is calculated as follows:

[0066]

[0067] Wherein, topo_Level(v) represents the topological order of node v in G, max(topo_Level) represents the largest topological order in G, and master_Rate(v) represents the proportion of students who have mastered this core concept on node v. The former part of the formula is used to represent the topological order of the node, and the more important the core concept is, the earlier the topological order is, and the greater the value of this part is; the latter part is used to represent the superimposed cognitive state of the learner on the node, and the more important the core concept is, the lower the correct rate is, and the greater the value of this part is. The multiplication of the two parts comprehensively obtains the node score under the "discovery of problems (diagnosis)" target.

[0068] (2) If the job target is "improving advantages (consolidation)", the importance score score(v) of node v is calculated as follows:

[0069]

[0070] Wherein, topo_Level(v) represents the topological order of node v in G, max(topo_Level) represents the largest topological order in G, and master_Rate(v) represents the proportion of students who have mastered this core concept on node v. The former part of the formula is used to represent the topological order of the node, and the more important the core concept is, the later the topological order is, and the greater the value of this part is; the latter part is used to represent the superimposed cognitive state of the learner on the node, and the more important the core concept is, the lower the correct rate is, and the greater the value of this part is. The multiplication of the two parts comprehensively obtains the node score under the "improving advantages (consolidation)" target.

[0071] (3) If the job target is "overall investigation (analysis of learning situation)", the importance score score(v) of node v is calculated as follows:

[0072] score(v) = (1 - master_Rate(v)) x degree_out(v) + (master_Rate(v)) x degree_in(v)

[0073] where master_Rate(v) represents the proportion of students on node v who have mastered this core concept, degree_in(v) represents the in-degree of node v, i.e. how many nodes in graph G are direct predecessors of v, and degree_out(v) represents the out-degree of node v, i.e. how many nodes in graph G are direct successors of v. This formula quantifies the influence of node v on the analysis of the learning situation of the core concepts of the predecessors and successors. According to the internal order of knowledge acquisition, students who have not mastered a certain core concept cannot master the successor nodes, and students who have mastered a certain core concept have already mastered the predecessor nodes. Therefore, the part before the plus sign represents the amount of learning situation analysis information that node v can provide for the successor nodes, and the part after the plus sign represents the amount of learning situation analysis information that node v can provide for the predecessor nodes, and the sum of the two represents the total amount of information that node v can provide under the overall survey target, and thus the score of the importance of node v.

[0074] In the above formula, (1) in-degree and out-degree: the in-degree refers to the number of edges pointing to the core concept node, and the out-degree refers to the number of edges pointing to other nodes from the core concept node. For example, Figure 3 The in-degree of node v6 is degree_in(v6) = 2, and the out-degree of node v6 is degree_out(v6) = 1.

[0075] (2) Topological order of nodes: the topological order is a linear sequence of all nodes in a directed acyclic graph, each node in the graph appears in the sequence and appears only once, and if there is a directed edge from point A to point B, then in the topological sequence, point A should appear before point B. In the present application, the topological order of the nodes is calculated according to the following steps:

[0076] (1) Initialize the variable topo_Level = 1;

[0077] (2) In the current directed acyclic graph, find all nodes with an in-degree of 0, and assign their topological order to topo_Level;

[0078] (3) Remove all nodes with an in-degree of 0 in the current graph, and delete the edges starting from these nodes, update the directed acyclic graph, and increase the value of topo_Level by 1;

[0079] (4) Repeat Step 2 and Step 3 until all nodes in the graph are removed, and obtain the topological order values of all nodes.

[0080] For example, for Figure 3The in-degree of nodes 1, 2, 8, 9, and 10 in the directed acyclic graph is 0, so the topological order values of these nodes are all 1; after deleting nodes 1, 2, 8, 9, and 10 and the edges (i.e., e1 and e2) connected to these nodes, the in-degree of nodes 3 and 4 becomes 0, so the topological order values of nodes 3 and 4 are 2; after deleting nodes 3 and 4 and the edges (i.e., e3 and e4) connected to these nodes, the in-degree of node 5 becomes 0, so the topological order value of node 5 is 3; in this way, nodes 6 (topological order value 4) and 7 (topological order value 5) are sequentially deleted, and all nodes in the directed acyclic graph are removed, and the calculation ends. Figure 3 The master_Rate(v3) of node v3 in the directed acyclic graph is 64%.

[0081] In step S32, after calculating the score(v) of all core concept nodes in the graph G, the top N core concept nodes (N is the workload) are taken as the arrangement points of the current assignment. For example, if the workload is 2, the nodes with the top two importance scores are taken as the arrangement points of the assignment.

[0082] In one embodiment, the teaching knowledge graph is as shown in FIG. 5. Figure 3 The user input is "workload=2, assignment goal=discover problems, and assignment range=<Pythagorean theorem, application of Pythagorean theorem>". First, based on the assignment range, the core concepts are extracted and a directed acyclic graph G=(V, E) is constructed, where V=<v4, v5, v6, v7> and E=<e4, e5, e6>. Under the assignment goal of "discover problems", the variables involved in the importance score calculation of the nodes are as follows:

[0083] topo_Level(v4)=1, master_Rate(v4)=0.8

[0084] topo_Level(v5)=1, master_Rate(v5)=0.6

[0085] topo_Level(v6)=2, master_Rate(v6)=0.5

[0086] topo_Level(v7)=3, master_Rate(v7)=0.45

[0087] max(topo_Level)=3

[0088] Then, the above variables are substituted into the node importance calculation formula for the assignment goal of "discover problems" to obtain the importance scores of core concept nodes v8, v9, and v10. 10Importance score of the node:

[0089]

[0090]

[0091]

[0092]

[0093] The top 2 core concepts (2 is the amount of work) are taken as the arrangement points of this work. Therefore, this work is arranged at core concept nodes v5 and v6. Compared with the traditional work arrangement method, the present application selects nodes with topological order in the front and less ideal cognitive state of learners, locks the weak basic knowledge of the group, and arranges work on it, so it can more effectively play the role of "discovering problems". Core concept nodes v5 and v6 have relatively high importance scores because of their relatively high importance scores. Although node v4 has the topological order in the front, it has a good group learner cognitive state (80% of students have mastered it), so its importance score is not high. Similarly, node v7 has a weak group learner cognitive state (only 25% of students have mastered it), but because of the relatively late topological order, it is not the most needed problem to be discovered at this stage according to the internal order of knowledge, so its importance score is not high. Therefore, nodes v4 and v7 will not be selected in this work arrangement.

[0094] In another embodiment, the teaching knowledge map is as shown in Figure 3 The user input is "work amount = 2, work target = overall investigation, work range = <Pythagorean theorem, application of Pythagorean theorem>". First, based on the work range, extract the core concepts and construct a directed acyclic graph G = (V, E), where V = <v4, v5, v6, v7>, E = <e4, e5, e6>. Under the work target of "overall investigation", the variables participating in the node importance score calculation are as follows:

[0095] degree_in(v4) = 0, degree_out(v4) = 1, master_Rate(v4) = 0.8

[0096] degree_in(v5) = 0, degree_out(v5) = 1, master_Rate(v5) = 0.6

[0097] degree_in(v6) = 2, degree_out(v6) = 1, master_Rate(v6) = 0.5

[0098] degree_in(v7)=1, degree_out(v7)=0, master_Rate(v7)=0.45

[0099] Then, substituting the above variables into the node importance calculation formula for the task objective of "overall investigation", we obtain the importance scores of core concept nodes v4, v5, v6, and v7:

[0100] score(v4)=(1-0.8)×1+0.8×0=0.2

[0101] score(v5)=(1-0.6)×1+0.6×0=0.4

[0102] score(v6)=(1-0.5)×1+0.5×2=1.5

[0103] score(v7)=(1-0.45)×0+0.45×1=0.45

[0104] The top two core concepts (2 representing the workload) are selected as the assignment points for this task. Therefore, this task is assigned to core concept nodes v6 and v7. Compared with traditional task assignment methods, this invention considers the amount of information that the connectivity between nodes can provide to the learning situation, and quantifies this amount of information through in-degree, out-degree, and node cognitive state. The greater the amount of information, the more learning situation information the node can provide, thus more effectively playing the role of "overall assessment".

[0105] The above-described method for precise homework design effectively overcomes the traditional reliance on manual experience in homework design, saving manpower and resources and improving efficiency. Simultaneously, this method considers the topological structure of the teaching knowledge graph and the objective learning information of student groups during the homework design process. It can tailor homework design to the actual learning situations of different student groups, and without changing the total amount of homework, the relevance and diagnostic effectiveness of the homework are often better than traditional methods. It is an effective way to balance the "quality" and "quantity" of homework under the "double reduction" policy. The precise homework design method proposed in this invention has higher efficiency and automation than manual homework assignment, making it more suitable for practical application. As a precise homework design method under the "double reduction" policy, this method has been tested by tens of thousands of primary and secondary school teachers and students in multiple districts and counties of Beijing, gaining widespread recognition from teachers.

[0106] The technical contents not described in detail in this invention are common knowledge to those skilled in the art.

[0107] While the foregoing specific embodiments of the application have been described in some detail to provide a clear understanding thereof, it will be apparent to those of ordinary skill in the art that numerous modifications can be made to the specific embodiments described without departing from the spirit and scope of the application as defined by the claims which are appended hereto.

Claims

1. A method of precision job design, characterized by, The method comprises the following steps: S1, receiving user inputted job information; S2, constructing a directed acyclic graph according to the user inputted job range and teaching knowledge graph; S3, calculating and outputting a key node set as a job arrangement point according to the user inputted job target and job amount; In step S1, the job information includes job amount, job range and job target; Step S3 comprises: S31, calculating a node importance score in the directed acyclic graph G based on the job target; S32, determining a node for job arrangement according to the job amount; In step S31, if the job target is "discovering problems-diagnosis", the calculation formula of the importance score score(v) of the node v is as follows: Wherein, topo_Level(v) represents the topological order of the node v in G, max(topo_Level) represents the maximum topological order in G, and master_Rate(v) represents the proportion of students who have mastered the core concept on the node v. In S32, after calculating the score(v) of all core concept nodes in the graph G, the top N core concepts are taken as the arrangement points of the job, and N is the job amount.

2. The method of precision job design of claim 1, wherein, The job range includes core concepts, and the job target is "discovering problems-diagnosis", "improving advantages-consolidation" and "overall investigation-learner analysis".

3. The method of precision job design of claim 1, wherein, The step S2 comprises: S21, extracting core concepts according to the job range; S22, obtaining the nodes corresponding to the core concepts and the nodes directly connected with the nodes in the educational knowledge graph to form a node set V, and converging all directed edges with the nodes in the node set V as the starting point and the ending point to form an edge set E, thereby constructing a directed acyclic graph G=(V,E) of the core concepts.

4. The method of precision job design of claim 3, wherein, The teaching knowledge graph is a multi-entity teaching knowledge graph, which is constructed based on the curriculum standards of primary school, junior high school and high school issued by the Ministry of Education, wherein the entity is a knowledge point in the curriculum standards of each subject, and the attribute is the cognitive state of the learner to the entity.

5. The method of precision job design of claim 3, wherein, The teaching knowledge graph comprises: unit module entity, core concept entity, cognitive state attribute of learners on core concepts, containing relationship between unit modules and core concepts, predecessor-successor relationship between core concepts under the same unit module, parallel relationship between unit modules, and predecessor-successor relationship between core concepts in different unit modules.

6. The method of precision job design of claim 1, wherein, In step S31, if the job target is "improving advantages-consolidation", the calculation formula of the importance score score(v) of the node v is as follows: Wherein, topo_Level(v) represents the topological order of the node v in G, max(topo_Level) represents the maximum topological order in G, and master_Rate(v) represents the proportion of students who have mastered the core concept on the node v.

7. The method of precision job design of claim 1, wherein, In step S31, if the job target is "overall investigation-learner analysis", the calculation formula of the importance score score(v) of the node v is as follows: score(v) = (1 - master_Rate(v)) x degree_out(v) + master_Rate(v) x degree_in(v) where master_Rate(v) represents the proportion of students on the node v have mastered this core concept, degree_in(v) represents the in-degree of node v, and degree_out(v) represents the out-degree of node v.

8. The method of precision job design of claim 1 or 6, wherein, The calculation method of the topological order of the nodes includes: (1): initialize the variable topo_Level = 1; (2): assign the topological order of the nodes with an in-degree of 0 in the directed acyclic graph to topo_Level; (3): delete the nodes with an in-degree of 0 in the directed acyclic graph, and simultaneously delete the edges starting from the nodes with an in-degree of 0, update the directed acyclic graph, and increase the value of topo_Level by 1; (4): repeat the execution of (2) and (3) until the graph is empty, thereby obtaining the topological order values of all nodes.

Images