CN114490918A - Dynamic knowledge graph building system and method based on dynamic learning model and application - Google Patents
Dynamic knowledge graph building system and method based on dynamic learning model and application Download PDFInfo
- Publication number
- CN114490918A CN114490918A CN202111666299.0A CN202111666299A CN114490918A CN 114490918 A CN114490918 A CN 114490918A CN 202111666299 A CN202111666299 A CN 202111666299A CN 114490918 A CN114490918 A CN 114490918A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- dynamic
- difficulty
- learning
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 19
- 230000001149 cognitive effect Effects 0.000 claims abstract description 39
- 230000002452 interceptive effect Effects 0.000 claims description 71
- 230000006870 function Effects 0.000 claims description 23
- 238000013507 mapping Methods 0.000 claims description 21
- 230000003993 interaction Effects 0.000 claims description 10
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 2
- 238000010276 construction Methods 0.000 claims description 2
- 238000002372 labelling Methods 0.000 claims description 2
- 238000012216 screening Methods 0.000 claims description 2
- 230000003213 activating effect Effects 0.000 claims 1
- 238000011835 investigation Methods 0.000 abstract description 3
- 238000010801 machine learning Methods 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007786 learning performance Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000007670 refining Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000010972 statistical evaluation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/31—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Electrically Operated Instructional Devices (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of machine learning, and particularly discloses a dynamic knowledge graph building system and method based on a dynamic learning model and application. The dynamic knowledge graph building system based on the dynamic learning model comprises a subject cognitive system building module, a subject difficulty coefficient feedback module and a knowledge graph sample-following dynamic updating module. The dynamic knowledge map building system based on the dynamic learning model takes the learning condition of the learner as an investigation factor to carry out dynamic analysis, thereby adjusting the difficulty coefficient in the dynamic knowledge map and providing a more appropriate learning path for the learner.
Description
Technical Field
The invention relates to the technical field of machine learning, in particular to a dynamic knowledge graph building system and method based on a dynamic learning model and application.
Background
In the traditional teaching, a teacher gives lessons according to teaching outlines of different age groups and related teaching material contents, and students finish homework, exercise and related tests corresponding to the knowledge point, so that the mastering proficiency of different students on the knowledge point is judged. The judgment mode highly depends on experience and qualification of different teacher individuals, so that the corresponding judgment conclusion of students is partially determined by subjective judgment. Limited by the personal energy of the teacher, in the process of teaching one to many lectures, the teacher often hardly catches the individual specificity of each student in the course of academic growth accurately.
For example, in the same test question, for the students who answer the wrong answers, it is difficult for the teacher to accurately judge whether the students are not well understood based on the advanced knowledge points, or are not well understood based on the forgetting curves, or are not well proficient on the advanced knowledge points and the knowledge points, or are only careless and cause errors. For students who answer correctly, it is difficult for teachers to accurately judge whether the students choose to be correct based on deep understanding of knowledge points, temporary memory of the teaching interactive contents, or guessing.
The difficulty coefficients corresponding to different knowledge points are obviously different for students in different years and students in the same age group with different regions, different learning habits and different daily preferences in the growth process of the students. The existing knowledge system, the knowledge map after data processing and the corresponding teaching, practice and evaluation means are all fixed frames, so that the knowledge system can not be dynamically adjusted according to the continuously updated academic state of student groups.
How to dynamically adjust the learning difficulty according to the learning state of the student and form a more appropriate learning path is one of the important subjects of the intelligent development and research of the learning system.
Disclosure of Invention
Aiming at the defect that the learning difficulty of the existing learning system is solidified and cannot be dynamically adjusted according to the learning state of a student, the invention aims to provide a dynamic knowledge graph building system, a dynamic knowledge graph building method and application based on a dynamic learning model, which can dynamically adjust the difficulty coefficient structure of a two-dimensional mapping knowledge graph according to the learning condition of the student to form a new learning path.
The invention discloses a dynamic knowledge graph building system based on a dynamic learning model, which comprises a subject cognitive system building module, a subject difficulty coefficient feedback module and a knowledge graph sample-following dynamic updating module;
the subject cognitive system building module comprises: the system comprises a primary cognitive branch unit, a secondary cognitive branch unit, a subject cognitive knowledge point unit, a knowledge point mesh structure building unit and a cognitive interactive data acquisition unit;
the subject difficulty coefficient feedback module comprises: a subject difficulty initial coefficient determining unit and a student data fitting learning difficulty function unit;
the knowledge graph sample-following dynamic updating module comprises: the device comprises a data sample magnitude unit, a data sample time unit and a data sample updating and recording unit;
the primary cognitive branch unit is used for extracting primary branches of disciplines from knowledge data;
the secondary cognitive branch unit is used for extracting secondary branches contained in each primary branch from the knowledge data;
the subject cognitive knowledge point unit is used for extracting knowledge points contained in each secondary branch and teaching interactive contents related to each knowledge point from knowledge data;
the knowledge point network structure building unit is used for building a knowledge structure tree of a subject according to the incidence relation of the primary branches, the secondary branches and the knowledge points, building the incidence relation of the network structure for all the knowledge points under the secondary branches according to the front-rear relation, setting a difficulty coefficient attribute for each knowledge point, and finally building a two-dimensional mapping knowledge graph comprising the two dimensions of the secondary branches and the difficulty coefficients;
the cognitive interactive data acquisition module is used for acquiring interactive data generated when a student sample and a student learning knowledge point correspond to the teaching interactive content;
the subject difficulty initial coefficient determining unit is used for endowing each knowledge point with an initial difficulty coefficient;
the student data fitting learning difficulty function unit is used for acquiring interactive data from the cognitive interactive data acquisition module, fitting the interactive data into a learning difficulty curve according to the repetition times and the error rate in the interactive data, and fitting and calculating a new difficulty coefficient according to the difficulty coefficient fitting function;
the data sample magnitude unit is used for collecting the number of students of learning and teaching interactive contents corresponding to each knowledge point from the student samples and the interactive data and forming a student number critical value array;
the data sample time unit is used for calculating time information and calculating a time critical value array from the last difficulty coefficient updating date;
and the data sample updating and recording unit is used for recording the flow steps of data fitting and difficulty coefficient updating when the difficulty coefficient is updated.
Further, the student sample comprises gender, age, grade, region, school category, nationality and learning habit of the student.
Further, the interaction data comprises basic information of knowledge points and operation information of students;
the basic information of the knowledge points comprises the positions of the knowledge points in the two-dimensional mapping knowledge map and the subordinate category characteristics;
the operation information of the students comprises error rate and repetition times when the students learn the teaching interactive contents corresponding to each knowledge point.
Further, the difficulty coefficient fitting function is: w is an=gn n×w；
Wherein, wnAverage error rate after n operations on the same teaching interactive content for m students;
gnthe difficulty coefficient of the knowledge point corresponding to the teaching interactive content is obtained after m students operate the same teaching interactive content for (m multiplied by n) times;
n is the number of operations;
w is an error rate general parameter for teaching interactive content pairs;
and m is the number of student samples and is a positive integer.
The invention also provides a dynamic knowledge graph building method based on the dynamic learning model, and a dynamic knowledge graph capable of dynamically adjusting the difficulty coefficient of each knowledge point is built based on a dynamic knowledge graph building system. The dynamic knowledge graph building method comprises the steps of firstly setting a difficulty coefficient attribute for each knowledge point based on a dynamic learning model, and building an incidence relation of a mesh structure for all knowledge points under a secondary branch according to a preposed and postpositioned relation to form a two-dimensional mapping knowledge graph; then training the two-dimensional mapping knowledge graph by a student sample to obtain a learning difficulty curve which is fit and synthesized by the error rate and the repetition times; and then, calculating a new difficulty coefficient according to the fitting function of the difficulty coefficient for updating, thereby constructing a knowledge graph capable of dynamically adjusting the difficulty coefficient.
The invention also provides a dynamic knowledge map for the self-adaptive learning system, which comprises a primary branch, a secondary branch, knowledge points, difficulty coefficients and difficulty coefficients which are classified according to disciplines; each discipline branch is provided with a plurality of primary branches, each primary branch is divided into a plurality of secondary branches, and each secondary branch covers at least one knowledge point; each knowledge point has a corresponding difficulty coefficient and a difficulty coefficient; a knowledge point belongs to one or more secondary branches.
The present invention also provides an adaptive learning system with a dynamic knowledge graph comprising a dynamic knowledge graph as claimed in claim 9.
The invention has the following beneficial effects.
(1) The invention can intelligently and dynamically adjust the difficulty coefficient structure of the knowledge map according to different student samples, so that a set of self-adaptive learning system can automatically match students with different student samples, the timeliness and the accuracy of the self-adaptive learning system can be greatly improved, and the self-adaptive learning mode of teaching according to the factors with the students as the center is really realized.
(2) The invention subdivides and optimizes the knowledge graph, and the knowledge graph not only comprises a primary branch, a secondary branch and knowledge points, but also comprises a difficulty coefficient and a difficulty coefficient, and the difficulty coefficient can be dynamically adjusted according to the student sample.
(3) The invention provides a dynamic knowledge map system architecture for a self-adaptive learning system, which comprises a primary branch, a secondary branch, knowledge points, difficulty coefficients and difficulty coefficients which are classified according to disciplines and have incidence relations, wherein the difficulty coefficients can be dynamically adjusted according to learning conditions.
Drawings
FIG. 1 is a schematic diagram of the architecture of the dynamic knowledge graph building system of the present invention.
FIG. 2 is a schematic diagram of a constructed mesh knowledge point map of a completed mathematical discipline of low age.
FIG. 3 is a diagram illustrating the relationship between the number of repetitions of the same type of teaching interactive exercise and the error rate of the student sample.
Detailed Description
The present invention will be described in further detail with reference to the following examples. This should not be understood as limiting the scope of the above-described subject matter of the present invention to the following examples. Various substitutions and alterations according to the general knowledge and conventional practice in the art are intended to be included within the scope of the present invention without departing from the technical spirit of the present invention as described above.
Example 1:
the embodiment discloses a dynamic knowledge graph building system based on a dynamic learning model, which comprises a subject cognitive system building module, a subject difficulty coefficient feedback module and a knowledge graph sample-following dynamic updating module, wherein the subject cognitive system building module is shown in figure 1;
the subject cognitive system building module comprises: the system comprises a primary cognitive branch unit, a secondary cognitive branch unit, a subject cognitive knowledge point unit, a knowledge point mesh structure building unit and a cognitive interactive data acquisition unit;
the subject difficulty coefficient feedback module comprises: a subject difficulty initial coefficient determining unit and a student data fitting learning difficulty function unit;
the knowledge graph sample-following dynamic updating module comprises: the device comprises a data sample magnitude unit, a data sample time unit and a data sample updating and recording unit;
the primary cognitive branch unit is used for extracting primary branches of disciplines from knowledge data;
the secondary cognitive branch unit is used for extracting secondary branches contained in each primary branch from the knowledge data;
the subject cognitive knowledge point unit is used for extracting knowledge points contained in each secondary branch and teaching interactive contents related to each knowledge point from knowledge data;
the knowledge point network structure building unit is used for building a knowledge structure tree of a subject according to the incidence relation of the primary branches, the secondary branches and the knowledge points, building the incidence relation of the network structure for all the knowledge points under the secondary branches according to the front-rear relation, setting a difficulty coefficient attribute for each knowledge point, and finally building a two-dimensional mapping knowledge graph comprising the two dimensions of the secondary branches and the difficulty coefficients;
the cognitive interactive data acquisition module is used for acquiring interactive data generated when a student sample and a student learning knowledge point correspond to the teaching interactive content;
the subject difficulty initial coefficient determining unit is used for endowing each knowledge point with an initial difficulty coefficient;
the student data fitting learning difficulty function unit is used for acquiring interactive data from the cognitive interactive data acquisition module, fitting the interactive data into a learning difficulty curve according to the repetition times and the error rate in the interactive data, and fitting and calculating a new difficulty coefficient according to the difficulty coefficient fitting function;
the data sample magnitude unit is used for collecting the number of students of learning and teaching interactive contents corresponding to each knowledge point from the student samples and the interactive data and forming a student number critical value array;
the data sample time unit is used for calculating time information and calculating a time critical value array from the last difficulty coefficient updating date;
and the data sample updating and recording unit is used for recording the flow steps of data fitting and difficulty coefficient updating when the difficulty coefficient is updated.
Further, the student sample comprises gender, age, grade, region, school category, nationality and learning habit of the student.
Further, the interaction data comprises basic information of knowledge points and operation information of students;
the basic information of the knowledge points comprises the positions of the knowledge points in the two-dimensional mapping knowledge map and the subordinate category characteristics;
the operation information of the students comprises error rate and repetition times when the students learn the teaching interactive contents corresponding to each knowledge point.
In an adaptive learning environment, for each teaching interactive content and the same type of teaching interactive exercise, under the condition that students similar to samples participate for multiple times, the error rate of the teaching interactive exercise is inversely related to the participation times (anti-corrected). When a sufficient number of student samples with the general statistical evaluation level being above the B level are subjected to data updating, the inverse correlation presented by different student samples is different. And fitting the inverse correlation data by using an exponential model so as to obtain a dynamic difficulty coefficient under a real-time environment.
As shown in fig. 3, the points of connection are averaged from hundreds of student samples. The horizontal axis represents the repeated times of the same type of teaching interactive exercise, and the vertical axis represents the normal distribution mean value of the error rate in the student sample.
Thus, the difficulty coefficient fitting function is: w is an=gn n×w；
Wherein, wnAverage error rate after n operations on the same teaching interactive content for m students;
gnthe difficulty coefficient of the knowledge point corresponding to the teaching interactive content is obtained after m students operate the same teaching interactive content for (m multiplied by n) times;
n is the number of operations;
w is an error rate general parameter for teaching interactive content pair;
and m is the number of student samples and is a positive integer.
gnAnd w are the original starting parameters g _0, w _0, determined by g 'and w' of the pre-reference teach points. When the teaching interactive practice of each knowledge point in the dynamic knowledge map accumulates sample data exceeding a critical value, a fine-tuning mode of the difficulty coefficient of the knowledge point in the dynamic knowledge map can be initiated at any time, the latest numerical value of g is continuously updated, and the dynamic knowledge map of the self-adaptive learning system is intelligently and dynamically adjusted.
For different student samples s, each knowledge point generates and stores the difficulty coefficient parameter g _ n, s for the sample. The sample parameters s include strong learning state related parameters such as learning behaviors, learning preferences and learning natural tendency of students, and weak related parameters such as age, gender, region and nationality.
By adopting the technical scheme, the self-adaptive learning system can intelligently and dynamically evaluate the semi-real-time learning performance of students instead of a solid state, and can more accurately provide personalized teaching service for each student in each class of student samples, so that the adaptability of the self-adaptive system to students with different learning habits and the real-time accuracy of the optimal learning path are improved to a greater extent.
Compared with the prior art, the dynamic knowledge graph building system based on the dynamic learning model dynamically analyzes the learning condition of the learner as an investigation element, so that the difficulty coefficient in the dynamic knowledge graph is adjusted, and a more appropriate learning path is provided for the learner. The dynamic knowledge map building system based on the dynamic learning model takes the learning condition of the learner as an investigation factor to carry out dynamic analysis, thereby adjusting the difficulty coefficient in the dynamic knowledge map and providing a more appropriate learning path for the learner.
Other parts of this embodiment are the same as those of the above embodiment, and thus are not described again.
Example 2:
the embodiment also provides a dynamic knowledge graph building method based on the dynamic learning model, and a dynamic knowledge graph capable of dynamically adjusting the difficulty coefficient of each knowledge point is built based on a dynamic knowledge graph building system. The dynamic knowledge graph building method comprises the steps of firstly setting a difficulty coefficient attribute for each knowledge point based on a dynamic learning model, and building an incidence relation of a mesh structure for all knowledge points under a secondary branch according to a preposed and postpositioned relation to form a two-dimensional mapping knowledge graph; then training the two-dimensional mapping knowledge graph by a student sample to obtain a learning difficulty curve which is fit and synthesized by the error rate and the repetition times; and then, calculating a new difficulty coefficient according to the fitting function of the difficulty coefficient for updating, thereby constructing a knowledge graph capable of dynamically adjusting the difficulty coefficient.
Further, the dynamic knowledge graph building method comprises the following steps:
step S1: dividing a discipline cognitive system building module into a primary branch, a secondary branch, a knowledge point and teaching interactive content from wide to narrow according to a discipline knowledge framework;
step S2: setting initial difficulty coefficients for all knowledge points under the secondary branch by a subject difficulty initial coefficient determining unit, wherein all the knowledge points are longitudinally arranged from small to large according to the initial difficulty coefficients;
step S3: establishing an incidence relation of a mesh structure by a knowledge point mesh structure construction unit according to a preposed and postpositioned relation of all knowledge points under the secondary branch to form a two-dimensional mapping knowledge map; for example: a constructed mesh knowledge point map of a completed mathematical discipline of low age as shown in figure 2;
step S4: adjusting the difficulty coefficient of the knowledge points according to the two-dimensional mapping knowledge graph of the mesh structure;
step S5: a cognitive interactive data acquisition module acquires student models and interactive data, and analyzes and records the error rate and the repetition times of the teaching interactive contents corresponding to each knowledge point;
step S6: and fitting the learning difficulty function unit by student data according to the error rate, the repetition times and the difficulty coefficient fitting function of the teaching interaction content corresponding to each knowledge point, calculating a new difficulty coefficient, and feeding the new difficulty coefficient back to the two-dimensional mapping knowledge map to obtain the dynamic knowledge map.
Further, the step S5 specifically includes the following steps:
step S51: recording interaction data (c 1, c2, c 3.., cn) of a plurality of students (s 1, s2, s 3.., sN) learning the same knowledge point; n and N are positive integers; generally n is in the range: 5< n < 10;
step S52: acquiring the number of students learning the same knowledge point and/or the last difficulty coefficient updating date from the student samples and the interactive data; when the number of students participating in learning the knowledge point initially reaches a threshold value in the threshold value series (thr 1, thr2, thr3,.., thrP), or the time threshold value is larger than the last difficulty coefficient updating date, the student data fitting learning difficulty function unit is activated.
Further, the step S6 specifically includes the following steps:
step S61: classifying data of the student samples, and labeling categories;
step S62: screening interaction data (c 1, c2, c3,.., cn) corresponding to each teaching interaction content in all or part of knowledge points, and calculating the average error rate and the repetition times of students in all or part of classes for each teaching interaction content; generally n is in the range: 5< n < 10;
step S63: for each teaching interactive content, fitting the difficulty coefficient fitting function according to the average error rate and the repetition times, and calculating a new difficulty coefficient; if the generated new difficulty coefficient is different from the original value, updating the value of the difficulty coefficient corresponding to the knowledge point;
step S64: and traversing all knowledge points in the two-dimensional mapping knowledge graph, and repeating the processes of S61-S6134 for the knowledge points reaching the activation triggering condition, thereby dynamically adjusting the knowledge graph.
Other parts of this embodiment are the same as those of the above embodiment, and thus are not described again.
Example 4:
the embodiment provides a dynamic knowledge graph for an adaptive learning system, which comprises a primary branch, a secondary branch, knowledge points, difficulty coefficients and difficulty coefficients which are classified according to disciplines; each subject branch is provided with a plurality of primary branches, each primary branch is divided into a plurality of secondary branches, and each secondary branch covers at least one knowledge point; each knowledge point has a corresponding difficulty coefficient and a difficulty coefficient; a knowledge point belongs to one or more secondary branches.
The first-level branch (Strand) is a first level of a tree structure designed for the subject and has global property and refining property, so that the background artificial intelligence algorithm can be well balanced, and the accuracy and the high efficiency are achieved when a model is built and knowledge points are screened through the tree structure.
The secondary branch (Sub Strand) is a supplementary refinement aiming at the first level of the discipline tree structure, and has completeness and detail, so that the first level can be fully and systematically extended, and all knowledge structures of the discipline can be effectively covered.
The knowledge point (Node) is the minimum learning unit in the learning system, is at the end of the tree structure, and has a pre-or post-dependent dependency relationship with each other. Each knowledge point is provided with evaluation links such as teaching and exercise interaction and the like, and is an important reference content of the palm holding degree of students in the system.
The difficulty coefficient (g) is a reference coefficient formed by fitting error rate curves corresponding to the same type of teaching exercises after a large number of times of training is carried out on similar student samples.
In the dynamic knowledge map, the knowledge points and the difficulty coefficients are fitted and longitudinally arranged, initial values of the difficulty coefficients of the knowledge points are set, the difficulty coefficients are in the same value, and the pre-dependency relationship and the post-dependency relationship before the knowledge points are determined, so that an initial reticular knowledge point map is obtained.
Other parts of this embodiment are the same as those of the above embodiment, and thus are not described again.
Example 5:
the present embodiment provides an adaptive learning system with a dynamic knowledge-graph, comprising a dynamic knowledge-graph as described in the above embodiments.
In a dynamic knowledge map under the environment of a self-adaptive Learning system, after a student participates in Learning of the Learning system, the Learning system deduces the mastery degree of the student on different knowledge points according to a background judgment standard and an algorithm, and intelligently and dynamically evaluates an optimal Path (affected Optimized Learning Path) which accords with the Learning behavior habit of the student and a next optimal knowledge point corresponding to the Path in the system architecture.
When new students continuously join the learning system for learning, the background of the learning system continuously updates and expands learning behavior performance data of the students in respective knowledge points (nodes), accumulates more sufficient student samples for each knowledge point, records the performance of each student in similar teaching exercises and respective accuracy error rate, and prepares data support for subsequent intelligent dynamic adjustment knowledge graph architecture. And then the optimal learning path is judged for different students more accurately, so that the real-time learning efficiency of the students is obviously enhanced.
Other parts of this embodiment are the same as those of the above embodiment, and thus are not described again.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and any simple modifications and equivalent variations of the above embodiments according to the technical spirit of the present invention are within the scope of the present invention.
Claims (10)
1. A dynamic knowledge graph building system based on a dynamic learning model is characterized by comprising a subject cognitive system building module, a subject difficulty coefficient feedback module and a knowledge graph sample-following dynamic updating module;
the subject cognitive system building module comprises: the system comprises a primary cognitive branch unit, a secondary cognitive branch unit, a subject cognitive knowledge point unit, a knowledge point mesh structure building unit and a cognitive interactive data acquisition unit;
the subject difficulty coefficient feedback module comprises: a subject difficulty initial coefficient determining unit and a student data fitting learning difficulty function unit;
the knowledge graph sample-following dynamic updating module comprises: the device comprises a data sample magnitude unit, a data sample time unit and a data sample updating and recording unit;
the primary cognitive branch unit is used for extracting primary branches of disciplines from knowledge data;
the secondary cognitive branch unit is used for extracting secondary branches contained in each primary branch from the knowledge data;
the subject cognitive knowledge point unit is used for extracting knowledge points contained in each secondary branch and teaching interactive contents related to each knowledge point from knowledge data;
the knowledge point network structure building unit is used for building a knowledge structure tree of a subject according to the incidence relation of the primary branches, the secondary branches and the knowledge points, building the incidence relation of the network structure for all the knowledge points under the secondary branches according to the front-rear relation, setting a difficulty coefficient attribute for each knowledge point, and finally building a two-dimensional mapping knowledge graph comprising the two dimensions of the secondary branches and the difficulty coefficients;
the cognitive interactive data acquisition module is used for acquiring interactive data generated when a student sample and a student learning knowledge point correspond to the teaching interactive content;
the subject difficulty initial coefficient determining unit is used for endowing each knowledge point with an initial difficulty coefficient;
the student data fitting learning difficulty function unit is used for acquiring interactive data from the cognitive interactive data acquisition module, fitting the interactive data into a learning difficulty curve according to the repetition times and the error rate in the interactive data, and fitting and calculating a new difficulty coefficient according to the difficulty coefficient fitting function;
the data sample magnitude unit is used for collecting the number of students of learning and teaching interactive contents corresponding to each knowledge point from the student samples and the interactive data and forming a student number critical value array;
the data sample time unit is used for calculating time information and calculating a time critical value array from the last difficulty coefficient updating date;
and the data sample updating and recording unit is used for recording the flow steps of data fitting and difficulty coefficient updating when the difficulty coefficient is updated.
2. The dynamic knowledge graph building system based on the dynamic learning model according to claim 1, wherein the student samples comprise gender, age, grade, region, school category, ethnicity and learning habits of students.
3. The dynamic knowledge graph building system based on the dynamic learning model according to claim 1, wherein the interaction data comprises basic information of knowledge points and operation information of students;
the basic information of the knowledge points comprises the positions of the knowledge points in the two-dimensional mapping knowledge map and the subordinate category characteristics;
the operation information of the students comprises error rate and repetition times when the students learn the teaching interactive contents corresponding to each knowledge point.
4. The dynamic knowledge graph building system based on the dynamic learning model according to claim 1, wherein the difficulty coefficient fitting function is: w is an=gn n×w；
Wherein, wnAverage error rate after n operations on the same teaching interactive content for m students;
gnthe difficulty coefficient of a knowledge point corresponding to the teaching interactive content is obtained after m students operate the same teaching interactive content for (m multiplied by n) times;
n is the number of operations;
w is an error rate general parameter for teaching interactive content pair;
and m is the number of student samples and is a positive integer.
5. A dynamic knowledge graph building method based on a dynamic learning model is characterized in that a dynamic knowledge graph capable of dynamically adjusting difficulty coefficients of all knowledge points is built based on the dynamic knowledge graph building system according to any one of claims 1 to 4; the method is characterized in that a difficulty coefficient attribute is set for each knowledge point based on a dynamic learning model, and the incidence relation of a mesh structure is established for all knowledge points under a secondary branch according to a preposed and postpositional relation to form a two-dimensional mapping knowledge map; then training the two-dimensional mapping knowledge graph by a student sample to obtain a learning difficulty curve which is fit and synthesized by the error rate and the repetition times; and then, calculating a new difficulty coefficient according to the fitting function of the difficulty coefficient for updating, thereby constructing a knowledge graph capable of dynamically adjusting the difficulty coefficient.
6. The dynamic knowledge graph building method based on the dynamic learning model according to claim 5, characterized by comprising the following steps:
step S1: dividing a discipline cognitive system building module into a primary branch, a secondary branch, a knowledge point and teaching interactive content from wide to narrow according to a discipline knowledge framework;
step S2: setting initial difficulty coefficients for all knowledge points under the secondary branch by a subject difficulty initial coefficient determining unit, wherein all the knowledge points are longitudinally arranged from small to large according to the initial difficulty coefficients;
step S3: establishing an incidence relation of a mesh structure by a knowledge point mesh structure construction unit according to a preposed and postpositioned relation of all knowledge points under the secondary branch to form a two-dimensional mapping knowledge map;
step S4: adjusting the difficulty coefficient of the knowledge points according to the two-dimensional mapping knowledge graph of the mesh structure;
step S5: a cognitive interactive data acquisition module acquires student models and interactive data, and analyzes and records the error rate and the repetition times of the teaching interactive contents corresponding to each knowledge point;
step S6: and fitting the learning difficulty function unit by student data according to the error rate, the repetition times and the difficulty coefficient fitting function of the teaching interaction content corresponding to each knowledge point, calculating a new difficulty coefficient, and feeding the new difficulty coefficient back to the two-dimensional mapping knowledge map to obtain the dynamic knowledge map.
7. The method for building a dynamic knowledge graph based on a dynamic learning model according to claim 6, wherein the step S5 specifically comprises the following steps:
step S51: recording interactive data of a plurality of students for learning the same knowledge point;
step S52: acquiring the number of students learning the same knowledge point and/or the last difficulty coefficient updating date from the student samples and the interactive data; and when the number of students participating in the learning of the knowledge point initially reaches a critical value in a critical value number series, or the time critical value is larger than the last time difficulty coefficient updating date, activating a student data fitting learning difficulty function unit.
8. The method for building a dynamic knowledge graph based on a dynamic learning model according to claim 7, wherein the step S6 specifically comprises the following steps:
step S61: classifying data of the student samples, and labeling categories;
step S62: screening interactive data corresponding to each teaching interactive content in all or part of knowledge points, and calculating the average error rate and the repetition times of all or part of classes of students for each teaching interactive content;
step S63: for each teaching interactive content, fitting the difficulty coefficient fitting function according to the average error rate and the repetition times, and calculating a new difficulty coefficient; if the generated new difficulty coefficient is different from the original value, updating the value of the difficulty coefficient corresponding to the knowledge point;
step S64: and traversing all knowledge points in the two-dimensional mapping knowledge graph, and repeating the processes of S61-S6134 for the knowledge points reaching the activation triggering condition, thereby dynamically adjusting the knowledge graph.
9. A dynamic knowledge map for an adaptive learning system comprises a primary branch, a secondary branch, knowledge points, difficulty coefficients and difficulty coefficients which are classified according to disciplines; each discipline branch is provided with a plurality of primary branches, each primary branch is divided into a plurality of secondary branches, and each secondary branch covers at least one knowledge point; each knowledge point has a corresponding difficulty coefficient and a difficulty coefficient; a knowledge point belongs to one or more secondary branches.
10. An adaptive learning system with a dynamic knowledge graph comprising the dynamic knowledge graph of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111666299.0A CN114490918A (en) | 2021-12-31 | 2021-12-31 | Dynamic knowledge graph building system and method based on dynamic learning model and application |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111666299.0A CN114490918A (en) | 2021-12-31 | 2021-12-31 | Dynamic knowledge graph building system and method based on dynamic learning model and application |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114490918A true CN114490918A (en) | 2022-05-13 |
Family
ID=81507586
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111666299.0A Pending CN114490918A (en) | 2021-12-31 | 2021-12-31 | Dynamic knowledge graph building system and method based on dynamic learning model and application |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114490918A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116541480A (en) * | 2023-07-05 | 2023-08-04 | 中国科学院文献情报中心 | Thematic data construction method and system based on multi-label driving |
| CN117252047A (en) * | 2023-11-20 | 2023-12-19 | 深圳市联特微电脑信息技术开发有限公司 | Teaching information processing method and system based on digital twinning |
| CN117575862A (en) * | 2023-12-11 | 2024-02-20 | 广州番禺职业技术学院 | Knowledge graph-based student personalized practical training guiding method and system |
-
2021
- 2021-12-31 CN CN202111666299.0A patent/CN114490918A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116541480A (en) * | 2023-07-05 | 2023-08-04 | 中国科学院文献情报中心 | Thematic data construction method and system based on multi-label driving |
| CN116541480B (en) * | 2023-07-05 | 2023-09-01 | 中国科学院文献情报中心 | Thematic data construction method and system based on multi-label driving |
| CN117252047A (en) * | 2023-11-20 | 2023-12-19 | 深圳市联特微电脑信息技术开发有限公司 | Teaching information processing method and system based on digital twinning |
| CN117252047B (en) * | 2023-11-20 | 2024-03-19 | 深圳市联特微电脑信息技术开发有限公司 | Teaching information processing method and system based on digital twinning |
| CN117575862A (en) * | 2023-12-11 | 2024-02-20 | 广州番禺职业技术学院 | Knowledge graph-based student personalized practical training guiding method and system |
| CN117575862B (en) * | 2023-12-11 | 2024-05-24 | 广州番禺职业技术学院 | Knowledge graph-based student personalized practical training guiding method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114490918A (en) | Dynamic knowledge graph building system and method based on dynamic learning model and application | |
| CN109767366A (en) | A kind of computer radar system of real time scan analysis vocabulary memorization effect | |
| KR101182390B1 (en) | Apparatus and Method for learning foreign language words seamlessly | |
| CN109388744A (en) | A kind of adaptive learning recommended method and device | |
| US20060166174A1 (en) | Predictive artificial intelligence and pedagogical agent modeling in the cognitive imprinting of knowledge and skill domains | |
| CN111062842A (en) | Method and device for dynamically generating personalized questions | |
| CN111125640A (en) | Knowledge point learning path recommendation method and device | |
| CN113487248B (en) | Personalized course customization system and method based on big data | |
| Pradina et al. | The Effect of Picture and Picture Learning model towards science outcomes for students with hearing impairment in the class VII | |
| CN112885172B (en) | Network teaching method and system | |
| Santos et al. | A taxonomy of exercises to support individual learning paths in initial programming learning | |
| Commeford et al. | Characterizing active learning environments in physics using latent profile analysis | |
| CN109635869A (en) | On-line study interfering system | |
| CN111311997B (en) | Interaction method based on network education resources | |
| Castillo et al. | Adaptive Bayes for a student modeling prediction task based on learning styles | |
| CN113190747A (en) | Personalized resource recommendation method based on learning style and cognitive level | |
| Al-Hammadi et al. | A neuro-fuzzy classification approach to the assessment of student performance | |
| CN113961773A (en) | Intelligent course selection recommendation system for college sports ball course | |
| Yarandi et al. | Personalised mobile learning system based on item response theory | |
| Javadi et al. | Improving student's modeling framework in a tutorial-like system based on Pursuit learning automata and reinforcement learning | |
| CN118037003B (en) | Online learning course optimization ordering method and system | |
| CN116453387B (en) | AI intelligent teaching robot control system and method | |
| Wahyono et al. | A novel intelligent learning for teaching using artificial neural network | |
| CN116433427A (en) | Personalized learning career data image drawing method and system | |
| Syahrial | Review/Turnitin Pedagogic Competence and Indonesia Language Competence Pre Service Teacher of Elementary Program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |