CN114861069A - Knowledge graph-based network learning resource analysis and personalized recommendation method - Google Patents
Knowledge graph-based network learning resource analysis and personalized recommendation method Download PDFInfo
- Publication number
- CN114861069A CN114861069A CN202210633790.1A CN202210633790A CN114861069A CN 114861069 A CN114861069 A CN 114861069A CN 202210633790 A CN202210633790 A CN 202210633790A CN 114861069 A CN114861069 A CN 114861069A
- Authority
- CN
- China
- Prior art keywords
- learning
- knowledge
- value
- analysis
- network
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 78
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000011156 evaluation Methods 0.000 claims abstract description 21
- 238000010276 construction Methods 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims description 30
- 230000008569 process Effects 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 8
- 150000001875 compounds Chemical class 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 239000002131 composite material Substances 0.000 claims description 3
- 238000000354 decomposition reaction Methods 0.000 claims description 3
- 238000009792 diffusion process Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 230000007246 mechanism Effects 0.000 claims description 3
- 238000013508 migration Methods 0.000 claims description 3
- 230000005012 migration Effects 0.000 claims description 3
- 238000005065 mining Methods 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000005295 random walk Methods 0.000 claims description 3
- 238000000547 structure data Methods 0.000 claims description 3
- 238000013526 transfer learning Methods 0.000 claims description 3
- 238000005516 engineering process Methods 0.000 abstract description 7
- 238000013210 evaluation model Methods 0.000 abstract description 3
- 238000012790 confirmation Methods 0.000 description 3
- 230000002457 bidirectional effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research 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
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Educational Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Animal Behavior & Ethology (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a knowledge graph-based network learning resource analysis and personalized recommendation method, which belongs to the technical field of learning resource analysis recommendation methods and specifically comprises the following steps: s1, constructing a network learning resource knowledge graph; s2, analyzing network learning resources; s3, analyzing the learner portrait; s4, personalized recommendation of network learning resources; the invention integrates the individual education knowledge map model of subject knowledge, learning resources and learning strategies and the construction technology thereof, and realizes accurate knowledge pushing and individual learning resources and learning strategy recommendation by taking individual interest and demand driving of learners as the center based on the learning resource concept link, analysis and evaluation technology of knowledge maps and learner knowledge system evaluation model and learning path intelligent planning of knowledge maps and learning targets.
Description
The invention belongs to the technical field of learning resource analysis and recommendation methods, and particularly relates to a knowledge graph-based network learning resource analysis and personalized recommendation method.
Background
With the continuous development of internet technology and the continuous change of application modes, the field of network learning is rapidly developed and widely applied. The rapid development of education informatization, learners can obtain various forms of network education resources in various modes, such as a mu class, Courses, a cyber-Yiyun classroom and the like. However, as the network education resources are explosively increased, the information overload problem faced by learners is increasingly intensified, and meanwhile, the demands and the utilization of the education resources are abnormally complicated due to the large differences of learners in learning objectives, knowledge levels, learning paths, learning styles and the like.
Therefore, how to effectively alleviate the problems of information overload and the like caused by mass resources, so that the conversion of learning activities from the traditional single mode of ' resource finding by people ' into the intelligent bidirectional mode of ' resource finding by people ' and resource finding by people ' becomes an important research subject in the fields of education informatization, artificial intelligence and the like, and in view of the above, the invention provides a knowledge-graph-based network learning resource analysis and personalized recommendation method.
Disclosure of Invention
The invention aims to effectively relieve the problems of information overload and the like caused by mass resources, so that the learning activity is converted from the traditional single mode of ' resource finding by people ' into the intelligent bidirectional mode of ' resource finding by people ' and resource finding by resources ', and the invention provides the knowledge-graph-based network learning resource analysis and personalized recommendation method.
In order to achieve the purpose, the invention adopts the following technical scheme:
a knowledge graph-based network learning resource analysis and personalized recommendation method specifically comprises the following steps:
s1, constructing a network learning resource knowledge graph: the method is characterized in that unstructured subject knowledge is researched by utilizing a metadata extraction method, the problem of insufficient linguistic data and labeled corpus shortage in the education field is solved, and the construction of a knowledge graph of network learning resources is realized by utilizing a migration learning method on the basis of the existing knowledge graph of teaching resources in each field;
s2, analyzing network learning resources: modeling teaching resources in each field in the network learning resource knowledge map by using an activation diffusion theory based on the network learning resource knowledge map constructed in S1 to obtain a vector model capable of representing knowledge points of the corresponding field, and simultaneously performing feature selection on the vector model through a hierarchical deep network;
s3, learner portrait analysis: acquiring learning analysis data of a learner, realizing construction and real-time dynamic update of a learner portrait according to the acquired learning analysis data, and obtaining learning resource requirements of the learner through analysis of the learner portrait;
s4, personalized recommendation of network learning resources: and (4) combining the contents obtained in the S2 and the S3, carrying out intelligent learning path planning on the learner based on the knowledge point centrality and the learning path recommendation method of the dynamic heterogeneous information network, and meanwhile, realizing accurate pushing and personalized recommendation of the network learning resources according to the learner portrait obtained in the S3.
Preferably, the e-learning resource knowledge graph mentioned in S1 selects a composite structure to represent, and the e-learning resource knowledge graph construction process specifically includes the following steps:
a1, dividing learning contents in knowledge maps of various fields into single knowledge points by using a metadata extraction method and combining the knowledge maps of teaching resources of various fields;
a2, based on the knowledge point data obtained in A1, finding out the dependency relationship between knowledge points in the same field and between knowledge points in different fields, and determining the learning sequence;
and A3, taking the dependency relationship among the knowledge points obtained in A2 as guidance, and linking the knowledge points based on a transfer learning method to construct a complete network learning resource knowledge graph.
Preferably, the learner learning analysis data mentioned in S3 includes explicit data and implicit data, where the explicit data specifically includes learner learning ability structure data, learning style analysis data, online learning preference analysis data and learning resource demand analysis data; the implicit data comprises online learning process analysis data and online learning result analysis data.
Preferably, the web learning resource personalized recommendation mentioned in S4 specifically includes the following steps:
b1, forming a network learning resource database according to the established network learning resource knowledge graph and the knowledge point vector model;
b2, combining the 'online learning resource database' obtained in B1 and the explicit data in S3 to perform learning analysis on learners, wherein the learning analysis specifically comprises learning style analysis, online learning preference analysis and learning resource demand analysis;
b3, after the learning analysis in the B2 is finished, determining and calling the content of the personalized learning resources from the network learning resource database according to the analysis result, wherein the content specifically comprises the steps of selecting the resource type, determining the push content, determining the recommendation time and determining the recommendation frequency;
b4, automatically pushing according to the content determined in the B3, and carrying out personalized learning by a learner according to the pushed content and automatically recording and storing an online learning process;
b5, analyzing the online learning data recorded and stored in the B4, specifically comprising online learning process analysis and online learning result analysis, and evaluating online learning according to the obtained analysis data;
b6, updating and adjusting the pushed content in the B3 in real time based on the online learning evaluation result obtained in the B5;
and B7, after the adjustment is completed, automatically performing intelligent learning path planning and accurate pushing and personalized recommendation of online learning resources to the learner.
Preferably, in the learning path recommendation method based on the centrality of the knowledge point mentioned in S4, the centrality of the knowledge point may be dynamically adjusted, and when the centrality of the knowledge point is evaluated, the value and the contribution value of the network node of the knowledge point are considered at the same time, and based on the value and the contribution value of the knowledge point, a computation model of the centrality of the knowledge point is proposed, and a computation formula of the computation model is as follows:
C i ＝AZ i Q (1)
in the formula, C i Representing the centrality of the node; a is an evaluation coefficient matrix which represents the contribution degree of the knowledge point itself and each order of neighbor nodes to the importance of the knowledge point i; z i Evaluating an index matrix for the value and the contribution value of the knowledge point i, wherein the index matrix comprises the value and the contribution value of the knowledge point i and the neighbor nodes of each order; q is an index weight matrix, and for the calculation method of the degree value and the contribution value centrality, the calculation method of the contribution amount of the same-order neighbor node in the centrality calculated according to the contribution value to the node i centrality is reserved, namely:
in the formula (I), the compound is shown in the specification,
the total contribution of the m-order neighbor node set of the node i to the centrality of the node i,
concentrating the value of a node j for an m-order neighbor node of a node i; therefore, the degree value and contribution value evaluation index matrix Z in the formula (1) i Can be expressed as:
in the formula, the first row of elements are values, the second row of elements are contribution values, the magnitude of data is different due to different value ranges of the values and the contribution values and different meanings and measurement units of each index, and an index matrix Z is evaluated on the values and the contribution values i Carrying out normalization processing; to Z i The value of (1) and the contribution value adopt a normalization formula:
in the formula (I), the compound is shown in the specification,
for normalized index value, from
Calculating a normalization value and contribution value evaluation index matrix Z' i Let us order
C i ＝AZ i Q, whereby the centrality of node i can be expressed as C i ＝AZ′ i Q, namely:
the knowledge point and the centrality of the m-order neighbor node under the constraints of the values and the contribution values are comprehensively considered, so that an accurate result can be obtained; in summary, the algorithm of knowledge point centrality based on the value and the contribution value can be summarized as the following steps:
c1, extracting a1 to m-order neighbor node set of knowledge points i according to the topological structure of the directed acyclic graph:
c2, calculating the values of the neighbors of each order of the knowledge point i and the contribution values, and determining the value of the knowledge point iAnd the contribution value evaluation index matrix Z i ；
C3 for Z i Normalizing the value and the contribution value, and calculating a normalized value and contribution value evaluation index matrix Z' i ；
C4, calculating the centrality of each node according to the formula (5), and outputting C i 。
Preferably, the learning path recommendation method based on the dynamic heterogeneous information network mentioned in S4 specifically includes the following steps:
d1, extracting sub-graphs from the knowledge graph to construct a heterogeneous information network related to the learning resources of the learner;
d2, fully mining the complex relation hidden by semantic association in the heterogeneous information network in D1 by adopting a random walk strategy guided by a symmetric element path;
d3, introducing a fusion strategy based on an attention mechanism, and organically fusing preference features generated by different weight element paths;
d4, solving the expandability problem by using the attribute information through a joint optimization matrix decomposition model and a fusion function;
d5, carrying out experimental analysis on the processing flows proposed in the D1-D4 on a real large-scale data set, and further constructing a learning path recommendation model based on a dynamic heterogeneous information network.
Compared with the prior art, the invention provides a knowledge graph-based network learning resource analysis and personalized recommendation method, which has the following beneficial effects:
the invention integrates the individual education knowledge map model of subject knowledge, learning resources and learning strategies and the construction technology thereof, and realizes accurate knowledge pushing and individual learning resources and learning strategy recommendation by taking individual interest and demand driving of learners as the center based on the learning resource concept link, analysis and evaluation technology of knowledge maps and learner knowledge system evaluation model and learning path intelligent planning of knowledge maps and learning targets.
Drawings
Fig. 1 is a schematic flow chart of a web learning resource personalized recommendation method based on a knowledge graph and a web learning resource analysis and personalized recommendation method provided by the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments.
Example 1:
referring to fig. 1, a method for analyzing and recommending network learning resources based on a knowledge graph specifically includes the following steps:
s1, constructing a network learning resource knowledge graph: the method is characterized in that unstructured subject knowledge is researched by utilizing a metadata extraction method, the problem of insufficient linguistic data and labeled corpus shortage in the education field is solved, and the construction of a knowledge graph of network learning resources is realized by utilizing a migration learning method on the basis of the existing knowledge graph of teaching resources in each field;
the network learning resource knowledge graph mentioned in the step S1 is represented by selecting a composite structure, and the network learning resource knowledge graph construction process specifically includes the following steps:
a1, dividing learning contents in knowledge maps of various fields into single knowledge points by using a metadata extraction method and combining the knowledge maps of teaching resources of various fields;
a2, based on the knowledge point data obtained in A1, finding out the dependency relationship between knowledge points in the same field and between knowledge points in different fields, and determining the learning sequence;
a3, taking the dependency relationship among the knowledge points obtained in A2 as guidance, linking the knowledge points based on a transfer learning method, and constructing a complete network learning resource knowledge graph;
s2, analyzing network learning resources: modeling teaching resources in each field in the network learning resource knowledge map by using an activation diffusion theory based on the network learning resource knowledge map constructed in S1 to obtain a vector model capable of representing knowledge points of the corresponding field, and simultaneously performing feature selection on the vector model through a hierarchical deep network;
s3, learner portrait analysis: acquiring learning analysis data of a learner, realizing construction and real-time dynamic update of a learner portrait according to the acquired learning analysis data, and obtaining learning resource requirements of the learner through analysis of the learner portrait;
the learner learning analysis data mentioned in the step S3 includes explicit data and implicit data, and the explicit data specifically includes learner learning energy structure data, learning style analysis data, online learning preference analysis data and learning resource demand analysis data; the implicit data comprises online learning process analysis data and online learning result analysis data;
s4, personalized recommendation of network learning resources: combining the contents obtained in S2 and S3, performing intelligent learning path planning on the learner by a learning path recommendation method based on knowledge point centrality, and meanwhile, realizing accurate pushing and personalized recommendation of online learning resources according to the learner portrait obtained in S3;
the personalized recommendation of the web learning resources mentioned in S4 specifically includes the following steps:
b1, forming a network learning resource database according to the established network learning resource knowledge graph and the knowledge point vector model;
b2, combining the 'e-learning resource database' obtained in B1 and the explicit data in S3 to carry out learning analysis on the learner, wherein the learning analysis specifically comprises learning style analysis, online learning preference analysis and learning resource demand analysis;
after learning analysis in B3 and B2 is finished, determining and calling the content of the personalized learning resources from the network learning resource database according to the obtained analysis result, wherein the content specifically comprises resource type selection, push content confirmation, recommendation time confirmation and recommendation frequency confirmation;
b4, automatically pushing according to the content determined in B3, and the learner performs personalized learning according to the pushed content and automatically records and stores the online learning process;
b5, analyzing the online learning data recorded and stored in the B4, specifically comprising online learning process analysis and online learning result analysis, and evaluating online learning according to the obtained analysis data;
b6, updating and adjusting the pushed content in B3 in real time based on the online learning evaluation result obtained in B5;
b7, after the adjustment is completed, automatically performing intelligent learning path planning and accurate pushing and personalized recommendation of online learning resources to the learner;
in the learning path recommendation method based on the knowledge point centrality mentioned in S4, when the centrality of the knowledge point is evaluated, the value and the contribution value of the knowledge point network node are considered at the same time, and based on the knowledge point value and the contribution value, a knowledge point centrality calculation model is proposed, which has a calculation formula as follows:
C i ＝AZ i Q (1)
in the formula, C i Representing the centrality of the node; a is an evaluation coefficient matrix which represents the contribution degree of the knowledge point itself and each order of neighbor nodes to the importance of the knowledge point i; z i Evaluating an index matrix for the value and the contribution value of the knowledge point i, wherein the index matrix comprises the value and the contribution value of the knowledge point i and the neighbor nodes of each order; q is an index weight matrix, and for the calculation method of the degree value and the contribution value centrality, the calculation method of the contribution amount of the same-order neighbor node in the centrality calculated according to the contribution value to the node i centrality is reserved, namely:
in the formula (I), the compound is shown in the specification,
the total contribution of the m-order neighbor node set of the node i to the centrality of the node i,
concentrating the value of a node j for an m-order neighbor node of a node i; therefore, the degree value and contribution value evaluation index matrix Z in the formula (1) i Can representComprises the following steps:
in the formula, the first row of elements are values, the second row of elements are contribution values, the magnitude of data is different due to different value ranges of the values and the contribution values and different meanings and measurement units of each index, and an index matrix Z is evaluated on the values and the contribution values i Carrying out normalization processing; to Z i The value of (1) and the contribution value adopt a normalization formula:
in the formula (I), the compound is shown in the specification,
for normalized index value, from
Calculating a normalization value and contribution value evaluation index matrix Z' i Let us order
C i ＝AZ i Q, whereby the centrality of node i can be expressed as C i ＝AZ′ i Q, namely:
the knowledge point and the centrality of the m-order neighbor node under the constraints of the values and the contribution values are comprehensively considered, so that an accurate result can be obtained; in summary, the algorithm of knowledge point centrality based on the value and the contribution value can be summarized as the following steps:
c1, extracting a1 to m-order neighbor node set of knowledge points i according to the topological structure of the directed acyclic graph:
c2, calculating the value and contribution value of each order neighbor of the knowledge point i, and determining the evaluation index matrix Z of the value and contribution value of the knowledge point i i ；
C3 for Z i Normalizing the value and the contribution value, and calculating a normalized value and contribution value evaluation index matrix Z' i ；
C4, calculating the centrality of each node according to the formula (5), and outputting C i 。
The learning path recommendation method based on the dynamic heterogeneous information network mentioned in S4 specifically includes the following steps:
d1, extracting sub-graphs from the knowledge graph to construct a heterogeneous information network related to the learning resources of the learner;
d2, fully mining the complex relation hidden by semantic association in the heterogeneous information network in D1 by adopting a random walk strategy guided by a symmetric element path;
d3, introducing a fusion strategy based on an attention mechanism, and organically fusing preference features generated by different weight element paths;
d4, solving the expandability problem by using the attribute information through a joint optimization matrix decomposition model and a fusion function;
d5, carrying out experimental analysis on the processing flows proposed in the D1-D4 on a real large-scale data set, and further constructing a learning path recommendation model based on a dynamic heterogeneous information network.
The invention integrates the individual education knowledge map model of subject knowledge, learning resources and learning strategies and the construction technology thereof, and realizes accurate knowledge pushing and individual learning resources and learning strategy recommendation by taking individual interest and demand driving of learners as the center based on the learning resource concept link, analysis and evaluation technology of knowledge maps and learner knowledge system evaluation model and learning path intelligent planning of knowledge maps and learning targets.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be able to cover the technical scope of the present invention and the equivalent alternatives or modifications according to the technical solution and the inventive concept of the present invention within the technical scope of the present invention.
Claims (6)
1. A knowledge graph-based network learning resource analysis and personalized recommendation method is characterized by comprising the following steps:
s1, constructing a network learning resource knowledge graph: the method is characterized in that unstructured subject knowledge is researched by utilizing a metadata extraction method, the problem of insufficient linguistic data and labeled corpus shortage in the education field is solved, and the construction of a knowledge graph of network learning resources is realized by utilizing a migration learning method on the basis of the existing knowledge graph of teaching resources in each field;
s2, analyzing network learning resources: modeling teaching resources in each field in the network learning resource knowledge map by using an activation diffusion theory based on the network learning resource knowledge map constructed in S1 to obtain a vector model capable of representing knowledge points of the corresponding field, and simultaneously performing feature selection on the vector model through a hierarchical deep network;
s3, learner portrait analysis: acquiring learning analysis data of a learner, realizing construction and real-time dynamic update of a learner portrait according to the acquired learning analysis data, and obtaining learning resource requirements of the learner through analysis of the learner portrait;
s4, personalized recommendation of network learning resources: and (4) combining the contents obtained in the S2 and the S3, carrying out intelligent learning path planning on the learner based on the knowledge point centrality and the learning path recommendation method of the dynamic heterogeneous information network, and meanwhile, realizing accurate pushing and personalized recommendation of the network learning resources according to the learner portrait obtained in the S3.
2. The method for analyzing and recommending personalized knowledge domain based on web learning resources of claim 1, wherein the web learning resources knowledge domain mentioned in S1 is represented by selecting a composite structure, and the web learning resources knowledge domain construction process specifically includes the following steps:
a1, dividing learning contents in knowledge maps of various fields into single knowledge points by using a metadata extraction method and combining the knowledge maps of teaching resources of various fields;
a2, based on the knowledge point data obtained in A1, finding out the dependency relationship between knowledge points in the same field and between knowledge points in different fields, and determining the learning sequence;
and A3, taking the dependency relationship among the knowledge points obtained in A2 as guidance, and linking the knowledge points based on a transfer learning method to construct a complete network learning resource knowledge graph.
3. The method as claimed in claim 1, wherein the learner learning analysis data in S3 includes explicit data and implicit data, and the explicit data includes learner known energy structure data, learning style analysis data, online learning preference analysis data and learning resource requirement analysis data; the implicit data comprises online learning process analysis data and online learning result analysis data.
4. The method for analyzing and personalized recommendation of web learning resources based on knowledge graph according to claim 1 or 3, wherein the personalized recommendation of web learning resources mentioned in S4 specifically includes the following steps:
b1, forming a network learning resource database according to the established network learning resource knowledge graph and the knowledge point vector model;
b2, combining the 'online learning resource database' obtained in B1 and the explicit data in S3 to perform learning analysis on learners, wherein the learning analysis specifically comprises learning style analysis, online learning preference analysis and learning resource demand analysis;
b3, after the learning analysis in the B2 is finished, determining and calling the content of the personalized learning resources from the network learning resource database according to the analysis result, wherein the content specifically comprises the steps of selecting the resource type, determining the push content, determining the recommendation time and determining the recommendation frequency;
b4, automatically pushing according to the content determined in the B3, and carrying out personalized learning by a learner according to the pushed content and automatically recording and storing an online learning process;
b5, analyzing the online learning data recorded and stored in the B4, specifically comprising online learning process analysis and online learning result analysis, and evaluating online learning according to the obtained analysis data;
b6, updating and adjusting the pushed content in the B3 in real time based on the online learning evaluation result obtained in the B5;
and B7, after the adjustment is completed, automatically performing intelligent learning path planning and accurate pushing and personalized recommendation of online learning resources to the learner.
5. The method for analyzing and recommending knowledge-graph-based network learning resources as claimed in claim 1, wherein the learning path recommendation method based on the centrality of knowledge points mentioned in S4 is characterized in that the centrality of knowledge points is dynamically adjusted, when the centrality of knowledge points is evaluated, the degree value and the contribution value of a network node of knowledge points are considered at the same time, and based on the degree value and the contribution value, a computation model of the centrality of knowledge points is proposed, and its computation formula is:
C i ＝AZ i Q (1)
in the formula, C i Representing the centrality of the node; a is an evaluation coefficient matrix which represents the contribution degree of the knowledge point itself and each order of neighbor nodes to the importance of the knowledge point i; z is a linear or branched member i Evaluating an index matrix for the value and the contribution value of the knowledge point i, wherein the index matrix comprises the value and the contribution value of the knowledge point i and the neighbor nodes of each order; q is an index weight matrix, and for the calculation method of the degree value and the contribution value centrality, the calculation method of the contribution amount of the same-order neighbor node in the centrality calculated according to the contribution value to the node i centrality is reserved, namely:
in the formula (I), the compound is shown in the specification,
the total contribution of the m-order neighbor node set of the node i to the centrality of the node i,
concentrating the value of a node j for an m-order neighbor node of a node i; therefore, the degree value and contribution value evaluation index matrix Z in the formula (1) i Can be expressed as:
in the formula, the first row of elements are values, the second row of elements are contribution values, the magnitude of data is different due to different value ranges of the values and the contribution values and different meanings and measurement units of each index, and an index matrix Z is evaluated on the values and the contribution values i Carrying out normalization processing; to Z i The value of (1) and the contribution value adopt a normalization formula:
in the formula (I), the compound is shown in the specification,
for normalized index value, from
Calculating an evaluation index matrix Z of the normalization value and the contribution value i ', order
C i ＝AZ i Q, whereby the centrality of node i can be expressed as C i ＝AZ i ' Q, i.e.:
the knowledge point and the centrality of the m-order neighbor node under the constraints of the values and the contribution values are comprehensively considered, so that an accurate result can be obtained; in summary, the algorithm of knowledge point centrality based on the value and the contribution value can be summarized as the following steps:
c1, extracting a1 to m-order neighbor node set of knowledge points i according to the topological structure of the directed acyclic graph:
Λ,
c2, calculating the value and contribution value of each order neighbor of the knowledge point i, and determining the evaluation index matrix Z of the value and contribution value of the knowledge point i i ；
C3 for Z i Normalizing the value and the contribution value, and calculating an evaluation index matrix Z of the normalized value and the contribution value i ′；
C4, calculating the centrality of each node according to the formula (5), and outputting C i 。
6. The method for analyzing and recommending network learning resources based on knowledge graph according to claim 1, wherein the learning path recommendation method based on dynamic heterogeneous information network mentioned in S4 specifically includes the following steps:
d1, extracting sub-graphs from the knowledge graph to construct a heterogeneous information network related to the learning resources of the learner;
d2, fully mining the complex relation hidden by semantic association in the heterogeneous information network in D1 by adopting a random walk strategy guided by a symmetric element path;
d3, introducing a fusion strategy based on an attention mechanism, and organically fusing preference features generated by different weight element paths;
d4, solving the expandability problem by using the attribute information through a joint optimization matrix decomposition model and a fusion function;
d5, carrying out experimental analysis on the processing flows proposed by the steps D1-D4 on a real large-scale data set, and further constructing a learning path recommendation model based on a dynamic heterogeneous information network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210633790.1A CN114861069A (en) | 2022-06-07 | 2022-06-07 | Knowledge graph-based network learning resource analysis and personalized recommendation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210633790.1A CN114861069A (en) | 2022-06-07 | 2022-06-07 | Knowledge graph-based network learning resource analysis and personalized recommendation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114861069A true CN114861069A (en) | 2022-08-05 |
Family
ID=82624962
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210633790.1A Pending CN114861069A (en) | 2022-06-07 | 2022-06-07 | Knowledge graph-based network learning resource analysis and personalized recommendation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114861069A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116720007A (en) * | 2023-08-11 | 2023-09-08 | 河北工业大学 | Online learning resource recommendation method based on multidimensional learner state and joint rewards |
| CN116797052A (en) * | 2023-08-25 | 2023-09-22 | 之江实验室 | Resource recommendation method, device, system and storage medium based on programming learning |
| CN116860978A (en) * | 2023-08-31 | 2023-10-10 | 南京云创大数据科技股份有限公司 | Primary school Chinese personalized learning system based on knowledge graph and large model |
-
2022
- 2022-06-07 CN CN202210633790.1A patent/CN114861069A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116720007A (en) * | 2023-08-11 | 2023-09-08 | 河北工业大学 | Online learning resource recommendation method based on multidimensional learner state and joint rewards |
| CN116720007B (en) * | 2023-08-11 | 2023-11-28 | 河北工业大学 | Online learning resource recommendation method based on multidimensional learner state and joint rewards |
| CN116797052A (en) * | 2023-08-25 | 2023-09-22 | 之江实验室 | Resource recommendation method, device, system and storage medium based on programming learning |
| CN116860978A (en) * | 2023-08-31 | 2023-10-10 | 南京云创大数据科技股份有限公司 | Primary school Chinese personalized learning system based on knowledge graph and large model |
| CN116860978B (en) * | 2023-08-31 | 2023-11-21 | 南京云创大数据科技股份有限公司 | Primary school Chinese personalized learning system based on knowledge graph and large model |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114861069A (en) | Knowledge graph-based network learning resource analysis and personalized recommendation method | |
| Schvaneveldt et al. | Network structures in proximity data | |
| Liang et al. | Heterogeneous multi-attribute nonadditivity fusion for behavioral three-way decisions in interval type-2 fuzzy environment | |
| CN107644062A (en) | The knowledge content Weight Analysis System and method of a kind of knowledge based collection of illustrative plates | |
| CN105893483A (en) | Construction method of general framework of big data mining process model | |
| CN109614495A (en) | A kind of associated companies method for digging of combination knowledge mapping and text information | |
| CN111914162B (en) | Method for guiding personalized learning scheme based on knowledge graph | |
| Hannes et al. | Mental maps and travel behaviour: Meanings and models | |
| CN108172047A (en) | A kind of network on-line study individualized resource real-time recommendation method | |
| Hassan et al. | Enhancing learning objects recommendation using multi-criteria recommender systems | |
| CN113538162A (en) | Planting strategy generation method and device, electronic equipment and storage medium | |
| Biloslavo et al. | Scenario planning for climate strategies development by integrating group Delphi, AHP and dynamic fuzzy cognitive maps | |
| Zhong et al. | Design of a personalized recommendation system for learning resources based on collaborative filtering | |
| CN110263183A (en) | A kind of smart home DSS and decision support method | |
| Miao | Industry 4.0: technology spillover impact on digital manufacturing industry | |
| Ye et al. | TiDEC: A two-layered integrated decision cycle for population evolution | |
| Kim et al. | A study of an agricultural ontology model for an intelligent service in a vertical farm | |
| Zhang et al. | Optimized artificial bee colony algorithm for web service composition problem | |
| Sun et al. | Education teaching evaluation method aided by adaptive genetic programming and robust scheduling | |
| Gao et al. | Optimized integration of traditional folk culture based on DSOM-FCM | |
| CN109005049A (en) | Service combining method based on Bigraph consistency algorithm under a kind of internet environment | |
| Kotova et al. | Information technologies in the management of technical systems-development of the engineering education | |
| Kumar et al. | Investigation on effect of solar energy generation on electricity price forecasting | |
| Li et al. | Personalized learning paths Recommendation based on Learner Personas | |
| CN108804788A (en) | A kind of Web service evolution method based on data cell model |
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 |