CN112328804A - Method, apparatus and storage medium for determining learning situation - Google Patents
Method, apparatus and storage medium for determining learning situation Download PDFInfo
- Publication number
- CN112328804A CN112328804A CN202011157699.4A CN202011157699A CN112328804A CN 112328804 A CN112328804 A CN 112328804A CN 202011157699 A CN202011157699 A CN 202011157699A CN 112328804 A CN112328804 A CN 112328804A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- graph
- user
- node
- learning
- 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 70
- 238000003062 neural network model Methods 0.000 claims abstract description 81
- 230000000306 recurrent effect Effects 0.000 claims abstract description 34
- 230000008569 process Effects 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims description 38
- 238000013528 artificial neural network Methods 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000007405 data analysis Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 description 14
- 238000012544 monitoring process Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 7
- 125000004122 cyclic group Chemical group 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000013500 data storage Methods 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 206010027175 memory impairment Diseases 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000003909 pattern recognition 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Educational Technology (AREA)
- Educational Administration (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Primary Health Care (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The application discloses a method, a device and a storage medium for determining learning condition. Wherein, the method comprises the following steps: receiving learning data information input by a user and generated in a learning process; calculating learning data information by using a pre-trained recurrent neural network model, and determining coding information corresponding to a knowledge point contained in the learning data information; and updating the preset knowledge graph according to the coding information by using a pre-trained graph neural network model, and determining the mastering condition of each knowledge node in the updated knowledge graph by the user, wherein the coding information of each knowledge point related to the user is stored in each knowledge node in the preset knowledge graph respectively.
Description
Technical Field
The present application relates to the field of adaptive learning technologies, and in particular, to a method, an apparatus, and a storage medium for determining a learning situation.
Background
With the development of online learning education, more and more learning platforms and systems are provided to help users to learn. In order to better guide the user to learn, the learning system needs to monitor the learning condition of the user. However, the existing general learning system uses a fixed feature function to monitor the learning condition of the user, so the monitoring mode is relatively fixed. In addition, the traditional learning condition monitoring mode only focuses on a single knowledge point, and cannot monitor the related knowledge points.
In view of the technical problems in the prior art that the learning condition monitoring mode for the user is relatively fixed and the learning conditions of a plurality of associated knowledge points cannot be monitored, no effective solution has been proposed at present.
Disclosure of Invention
The embodiment of the disclosure provides a method, a device and a storage medium for determining a learning condition, so as to solve at least the technical problems that in the prior art, a learning condition monitoring mode for a user is relatively fixed, and the learning condition of a plurality of associated knowledge points cannot be monitored.
According to an aspect of the embodiments of the present disclosure, there is provided a method of determining a learning situation, including: receiving learning data information input by a user and generated in a learning process; calculating learning data information by using a pre-trained recurrent neural network model, and determining coding information corresponding to a knowledge point contained in the learning data information; and updating the preset knowledge graph according to the coding information by using a pre-trained graph neural network model, and determining the mastering condition of each knowledge node in the updated knowledge graph by the user, wherein the coding information of each knowledge point related to the user is stored in each knowledge node in the preset knowledge graph respectively.
According to another aspect of the embodiments of the present disclosure, there is also provided a storage medium including a stored program, wherein the method of any one of the above is performed by a processor when the program is executed.
According to another aspect of the embodiments of the present disclosure, there is also provided an apparatus for determining a learning condition, including: the data receiving module is used for receiving learning data information input by a user and generated in the learning process; the coding information determining module is used for calculating the learning data information by utilizing a pre-trained recurrent neural network model and determining coding information corresponding to the knowledge points contained in the learning data information; and the learning condition determining module is used for updating the preset knowledge graph according to the coding information by using the pre-trained graph neural network model and determining the mastering condition of each knowledge node in the updated knowledge graph by the user, wherein the coding information of each knowledge point related to the user is stored in each knowledge node in the preset knowledge graph respectively.
According to another aspect of the embodiments of the present disclosure, there is also provided an apparatus for determining a learning condition, including: a processor; and a memory coupled to the processor for providing instructions to the processor for processing the following processing steps: receiving learning data information input by a user and generated in a learning process; calculating learning data information by using a pre-trained recurrent neural network model, and determining coding information corresponding to a knowledge point contained in the learning data information; and updating the preset knowledge graph according to the coding information by using a pre-trained graph neural network model, and determining the mastering condition of each knowledge node in the updated knowledge graph by the user, wherein the coding information of each knowledge point related to the user is stored in each knowledge node in the preset knowledge graph respectively.
In the embodiment of the present disclosure, the cyclic neural network model is used to determine the coding information corresponding to the knowledge points included in the learning data information, and then the graph neural network model is used to calculate the coding information, so as to update the knowledge graph formed by the knowledge nodes, and determine the mastering conditions of the user on the knowledge nodes in the updated knowledge graph. Therefore, compared with the prior art, the scheme can continuously update the knowledge graph according to the knowledge points contained in the learning data information, and further can analyze the mastering condition of the knowledge nodes by the user more flexibly. In addition, the knowledge graph is constructed by knowledge nodes, so that association exists among the knowledge points, the mastering conditions of a plurality of associated knowledge points can be further determined, and the learning condition can be monitored more accurately. The technical effect of flexibly and accurately determining the mastering condition of each knowledge node by the user is achieved. And the technical problems that the learning condition monitoring mode of the user is fixed and the learning conditions of a plurality of associated knowledge points cannot be monitored in the prior art are solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the disclosure and together with the description serve to explain the disclosure and not to limit the disclosure. In the drawings:
fig. 1 is a hardware block diagram of a computing device for implementing the method according to embodiment 1 of the present disclosure;
fig. 2 is a schematic flow chart of a method for determining a learning situation according to the first aspect of embodiment 1 of the present disclosure;
fig. 3A is a schematic diagram of a preset knowledge-graph according to a first aspect of embodiment 1 of the present disclosure;
FIG. 3B is a schematic diagram of an updated knowledge-graph according to the first aspect of embodiment 1 of the present disclosure;
fig. 4 is a schematic diagram of an apparatus for determining a learning situation according to embodiment 2 of the present disclosure; and
fig. 5 is a schematic diagram of an apparatus for determining a learning situation according to embodiment 3 of the present disclosure.
Detailed Description
In order to make those skilled in the art better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in the embodiments of the present disclosure. It is to be understood that the described embodiments are merely exemplary of some, and not all, of the present disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
It should be noted that the terms "first," "second," and the like in the description and claims of the present disclosure and in the above-described drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the disclosure described herein are capable of operation in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
According to the present embodiment, there is also provided an embodiment of a method of determining a learning condition, it being noted that the steps illustrated in the flowchart of the drawings may be performed in a computer system such as a set of computer-executable instructions, and that while a logical order is illustrated in the flowchart, in some cases the steps illustrated or described may be performed in an order different than here.
The method embodiments provided by the present embodiment may be executed in a server or similar computing device. Fig. 1 illustrates a block diagram of a hardware architecture of a computing device for implementing a method of determining learning conditions. As shown in fig. 1, the computing device may include one or more processors (which may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory for storing data, and a transmission device for communication functions. Besides, the method can also comprise the following steps: a display, an input/output interface (I/O interface), a Universal Serial Bus (USB) port (which may be included as one of the ports of the I/O interface), a network interface, a power source, and/or a camera. It will be understood by those skilled in the art that the structure shown in fig. 1 is only an illustration and is not intended to limit the structure of the electronic device. For example, the computing device may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
It should be noted that the one or more processors and/or other data processing circuitry described above may be referred to generally herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part in software, hardware, firmware, or any combination thereof. Further, the data processing circuitry may be a single, stand-alone processing module, or incorporated in whole or in part into any of the other elements in the computing device. As referred to in the disclosed embodiments, the data processing circuit acts as a processor control (e.g., selection of a variable resistance termination path connected to the interface).
The memory may be used to store software programs and modules of application software, such as program instructions/data storage devices corresponding to the method for determining learning condition in the embodiments of the present disclosure, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, implementing the method for determining learning condition of application software. The memory may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory located remotely from the processor, which may be connected to the computing device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device is used for receiving or transmitting data via a network. Specific examples of such networks may include wireless networks provided by communication providers of the computing devices. In one example, the transmission device includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission device may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.
The display may be, for example, a touch screen type Liquid Crystal Display (LCD) that may enable a user to interact with a user interface of the computing device.
It should be noted here that in some alternative embodiments, the computing device shown in fig. 1 described above may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that FIG. 1 is only one example of a particular specific example and is intended to illustrate the types of components that may be present in a computing device as described above.
In the operating environment described above, according to the first aspect of the present embodiment, there is provided a method of determining a learning situation, which can be applied to a server of a learning education platform system to operate, for example. Fig. 2 shows a flow diagram of the method, which, with reference to fig. 2, comprises:
s202: receiving learning data information input by a user and generated in a learning process;
s204: calculating learning data information by using a pre-trained recurrent neural network model, and determining coding information corresponding to a knowledge point contained in the learning data information; and
s206: and updating the preset knowledge graph according to the coding information by using a pre-trained graph neural network model, and determining the mastering condition of each knowledge node in the updated knowledge graph by the user, wherein the coding information of each knowledge point related to the user is stored in each knowledge node in the preset knowledge graph respectively.
As described in the background, with the development of online learning education, more and more learning platforms and systems have appeared to help users to learn. In order to better guide the user to learn, the learning system needs to monitor the learning condition of the user. However, the existing general learning system uses a fixed feature function to monitor the learning condition of the user, so the monitoring mode is relatively fixed. In addition, the traditional learning condition monitoring mode only focuses on a single knowledge point, and cannot monitor the related knowledge points.
In order to solve the technical problem in the prior art, in step S202, the server first receives learning data information generated in a learning process input by a user. In a specific application scenario, a user may learn at the learning education platform, and during the learning process, learning data information may be generated, such as: the learning data information may be answer information, learning notes, post-lesson work, etc., which the user may actively upload to the learning education platform, in which case the server of the education platform may receive. In addition, the server may also actively acquire the learning data information during the learning process of the user, and the generation and acquisition manner of the learning data information is not particularly limited herein.
Further, in step S204, the server calculates the learning data information using a previously trained recurrent neural network model, and specifies the encoded information corresponding to the knowledge point included in the learning data information. The education platform can preset a trained recurrent neural network model (RNN model), and under the condition that the server receives learning data information input by a user, the server can call the recurrent neural network model to calculate the learning data information and determine coding information corresponding to knowledge points contained in the learning data information.
In one particular example, the user performs a learning of a math lesson, for example, so that the learning data information entered by the user includes knowledge points related to "math". The server firstly determines knowledge points contained in the learning data information by using a recurrent neural network model, for example, the knowledge points are functions, and then determines coding information corresponding to the knowledge points, namely, corresponding coding information is generated according to the knowledge points. It should be noted that, the above is only an example of one knowledge point, but in the actual learning process, a plurality of knowledge points may be involved, so that a plurality of knowledge points may also be included in the learning data information, and finally, the encoded information corresponding to each knowledge point needs to be determined.
Finally, in step S206, the server updates the preset knowledge graph according to the coding information by using the pre-trained graph neural network model, wherein each knowledge node in the preset knowledge graph stores the coding information of each knowledge point related to the user. In one embodiment, the knowledge-graph is continuously more informative based on the learning process of the user, such as: from the beginning of learning by the user to the present, knowledge points such as "set", "sequence", "inequality", "equation", "quadratic equation", etc. have been learned, and therefore, referring to fig. 3A, five knowledge nodes (in addition, "math" nodes) such as "set", "sequence", "inequality", "equation", "quadratic equation", etc. may be included in the knowledge graph, and the knowledge graph in fig. 3A is only an exemplary explanation, and in an actual scenario, the nodes of the knowledge graph may be determined according to the learning progress of the user. Furthermore, in the knowledge-graph, each knowledge node stores the coding information of each knowledge point, namely: the knowledge points are stored on corresponding knowledge nodes in the knowledge graph in the form of coded information. Under the condition that the user inputs information including a function knowledge point at this time, the server may input the encoded information corresponding to the function knowledge point to the neural network model of the graph for calculation, so as to update the entire knowledge graph, as shown in fig. 3B, the updated knowledge graph includes six knowledge nodes (including a mathematic node) such as a set, a sequence, an inequality, an equation, a quadratic equation, and a function.
Moreover, the server needs to determine the user's grasp of each knowledge node in the updated knowledge graph. The grasping condition of the user on each knowledge node can be determined according to the learning data information uploaded by the user, for example: the learning data information uploaded by the user is answer information, so that the mastering condition of the knowledge point by the user can be determined according to the accuracy of the answer information. For example: the accuracy rate of the answer information uploaded by the user at this time for the function knowledge point is 80%, the user can be determined to know the function knowledge point by 80%, and other nodes are the same as the function knowledge point calculation mode, which is not repeated here. And analyzing and calculating the learning data information uploaded by the user each time by using the model, continuously and iteratively updating the knowledge graph, and finally determining the mastering condition of the user on each knowledge node in the knowledge graph.
Therefore, in this way, the cyclic neural network model is used for determining the coding information corresponding to the knowledge points contained in the learning data information, then the graph neural network model is used for calculating the coding information, the knowledge graph formed by the knowledge nodes is updated, and the mastering condition of the user on each knowledge node in the updated knowledge graph is determined. Therefore, compared with the prior art, the scheme can continuously update the knowledge graph according to the knowledge points contained in the learning data information, and further can analyze the mastering condition of the knowledge nodes by the user more flexibly. In addition, the knowledge graph is constructed by knowledge nodes, so that association exists among the knowledge points, the mastering conditions of a plurality of associated knowledge points can be further determined, and the learning condition can be monitored more accurately. The technical effect of flexibly and accurately determining the mastering condition of each knowledge node by the user is achieved. And the technical problems that the learning condition monitoring mode of the user is fixed and the learning conditions of a plurality of associated knowledge points cannot be monitored in the prior art are solved.
Optionally, the operation of updating the preset knowledge graph according to the encoded information by using a pre-trained graph neural network model includes: extracting relationship strength among all knowledge nodes in the preset knowledge graph from big data analysis by using a graph neural network model as an updating weight for updating the preset knowledge graph; generating a new knowledge node to be added to a preset knowledge graph, wherein the new knowledge node is used for storing coding information corresponding to the knowledge point contained in the learning data information; and adding the new knowledge node to the preset knowledge graph, and updating the preset knowledge graph according to the updating weight.
Specifically, in the operation of updating the preset knowledge graph according to the coding information by using the pre-trained graph neural network model, the server firstly extracts the relationship strength between each knowledge node in the preset knowledge graph from the big data analysis by using the graph neural network model as the updating weight for updating the preset knowledge graph. The big data can be applied to extract user information of all learning users and identify different user modes, and then all knowledge nodes in the knowledge graph are analyzed to determine the relationship strength among all knowledge nodes.
The relationship strength between knowledge points included in the learning process can be determined through big data analysis, and the relationship strength between different knowledge nodes may be different, for example: the relationship strength between the above equation and quadratic equation is strong, the relationship strength between the equation and the set is weak, and the like, the strength relationship between all knowledge points (not limited to knowledge nodes in the existing knowledge graph) included in the learning process can be determined through big data analysis, and then the strength relationship is used as the update weight for updating the preset knowledge graph. Further, the server generates a new knowledge node to be added to the preset knowledge graph, wherein the new knowledge node is used for storing the coding information corresponding to the knowledge point contained in the learning data information, that is, generating a new knowledge node, the new knowledge node needs to be added to the preset knowledge graph, and the new knowledge node is used for storing the coding information corresponding to the knowledge point contained in the learning data information, that is: and adding the coding information corresponding to the function into the new generation node. Finally, the server adds the new knowledge node to the preset knowledge graph, namely, the function knowledge node is added to the knowledge graph to update the knowledge graph. Because a knowledge node is added in the knowledge graph, the coding information corresponding to each knowledge node in the updated knowledge graph needs to be updated, wherein the updating mode can adopt a convolution method and combines the determined updating weight to update the knowledge graph.
Therefore, by the method, the relation strength among the knowledge nodes can be analyzed by using big data in the process of updating the knowledge graph, namely, the user information of all learning users is used in the process of updating the knowledge graph, then the relation strength among the knowledge nodes obtained by analysis is used as an updating weight, and finally the knowledge graph is updated according to the updating weight. Due to the adoption of big data analysis, the updated knowledge graph is more accurate, and the logical relationship between the nodes is more reasonable.
Optionally, the updating weights include node updating weights of all knowledge nodes in a preset knowledge graph and edge updating weights of all edges, and the operation of updating the preset knowledge graph according to the updating weights includes: recalculating the coding information stored by each knowledge node in the knowledge graph added with the new knowledge node according to the node update weight and the edge update weight; and respectively storing the recalculated result to each knowledge node in the knowledge graph added with the new knowledge node.
Specifically, the update weight includes a node update weight of each knowledge node in a preset knowledge graph and an edge update weight of each edge, for example: referring to fig. 3A, each knowledge node (knowledge node such as a set, an equation, etc.) is associated with an update weight, and a connection line (edge) between each knowledge node is also associated with an update weight, for example: the edges between the nodes of the equation and the quadratic equation are corresponding to update weights, the edges between the nodes of the mathematic equation and the equation are also corresponding to update weights, and other nodes and edges are corresponding to update weights, which is not described herein again. And the server distributes the weight of the edge in a mode of pattern recognition when determining the relation strength among all knowledge nodes in the knowledge graph.
In the operation of updating the preset knowledge graph according to the updating weight, the server firstly recalculates the coding information stored by each knowledge node in the knowledge graph to which the new knowledge node is added according to the node updating weight and the edge updating weight. Since the knowledge graph is added with "function" knowledge nodes, the node codes corresponding to each node in the updated knowledge graph need to be recalculated, wherein the node codes can be calculated according to the node update weights and the edge update weights, for example, the code information of the node is calculated according to the weight corresponding to each node and the update weight corresponding to the node connection point edge. Therefore, by the method, in the operation of the coding information corresponding to each knowledge node in the updated knowledge graph, the calculation can be performed according to the updating weight of each node and each edge, and the coding information stored by each knowledge node can accurately represent the information of the stored knowledge.
Optionally, the operation of determining the grasping condition of the user on each knowledge node in the updated knowledge graph by using a pre-trained graph neural network model includes: respectively carrying out decoding operation on the coding information stored by each knowledge node in the updated knowledge graph by using a preset decoding network; inputting the result obtained by the decoding operation into a preset fully-connected neural network, and outputting the mastering condition of each knowledge node in the updated knowledge graph by a user; and determining the grasping condition of the user on each knowledge node in the updated knowledge graph according to the historical grasping condition of the user, the current learning progress information and the output grasping condition.
Specifically, the technical solution of the present embodiment further includes a decoding network (i.e., a decoder). In the operation of determining the grasping condition of each knowledge node in the updated knowledge graph by using a pre-trained graph neural network model, the server firstly performs the decoding operation on the coding information stored by each knowledge node in the updated knowledge graph by using a preset decoding network. Wherein the encoded information corresponding to each node can be decoded into a data form that can be calculated by the system, for example, through a decoding operation. Further, the server inputs the result obtained by the decoding operation into a preset fully-connected neural network, and outputs the mastering condition of the user on each knowledge node in the updated knowledge graph. The preset fully-connected neural network can output according to actual requirements, for example, a percentage value of the mastery condition of each node is output. In the using process, the decoded result can be input into the fully-connected neural network, and finally the grasping condition is output.
Further, in order to ensure the accuracy of the learning and mastering conditions fed back to the user, the server needs to determine the mastering conditions of the user on each knowledge node in the updated knowledge graph according to the historical mastering conditions, the current learning progress information and the output mastering conditions of the user. That is, in the operation of determining the final grasping situation, the grasping situations of the knowledge nodes in the updated knowledge graph by the user can be determined jointly in combination with the historical grasping situations (for example, the grasping situation of each knowledge point in the knowledge graph updated each time), the current learning progress (for example, the grasping situations of the learned knowledge points and the knowledge points which are not learned), and the grasping situations output this time. Therefore, the grasping condition determined in the mode is more comprehensive, and the learning progress condition of each knowledge node can be reflected by combining historical data and learning progress.
Optionally, the present solution further includes: determining a learning path corresponding to the user according to the grasping condition of the user on each knowledge node in the updated knowledge graph; and pushing the learning path to the terminal device of the user.
Specifically, the learning path may also be determined according to the grasping condition of the user on each knowledge node in the knowledge graph, and in a specific example, the learning path may be determined according to the grasping condition of each knowledge node, for example: if the mastering condition of the quadratic equation is poor, the knowledge related to the quadratic equation is placed at the front end of the learning path, so that the user can learn the quadratic equation knowledge preferentially. Furthermore, the scheme can also send the learning path to the terminal equipment of the user, so that the user can learn according to the learning path formulated by the system, and the learning path is determined according to the grasping condition of the user through the system, so that the learning efficiency of the user can be improved.
Optionally, the calculating the learning data information by using a pre-trained recurrent neural network model, and determining the coding information corresponding to the knowledge point included in the learning data information, includes: extracting characteristic information corresponding to a knowledge point contained in the learning data information by using a pre-trained recurrent neural network model; and coding the characteristic information by using a preset coding network to determine the coding information of the knowledge points.
Specifically, in the operation of calculating the learning data information by using the pre-trained recurrent neural network model and determining the coding information corresponding to the knowledge points included in the learning data information, the server first extracts feature information corresponding to the knowledge points included in the learning data information by using the pre-trained recurrent neural network model, that is, extracts feature information of the knowledge points from the learning data information. Further, in order to make the feature information recognizable by the neural network, the server further needs to perform encoding processing on the feature information by using a preset encoding network to determine the encoding information of the knowledge points, for example: and encoding the characteristic information by using an encoding network corresponding to the decoding network to obtain the encoded information. Thus, the obtained coding information can be input to the graph neural network for calculation.
Optionally, the method further comprises training the neural network model according to the following steps: acquiring training coding information for training, wherein the training coding information is determined by calculating learning data information for training by a recurrent neural network model; inputting the training coding information into a graph neural network model for calculation; and carrying out optimization training on the graph neural network model by using a gradient descent mode according to the calculation result.
Specifically, the technical solution of this embodiment further includes a step of training the neural network model of the graph. Specifically, first, training code information for training is acquired, wherein the training code information is determined by calculating learning data information for training by a recurrent neural network model. It should be added here that the present solution further includes training the recurrent neural network model, specifically training the recurrent neural network with sequence data through the learning condition of the individual user. Variables of the recurrent neural network are then frozen (i.e., secured)The trained recurrent neural network is kept unchanged), and then a graph neural network model is trained by a convolution method. The storage structure of the graph neural network is as follows: assuming that the knowledge graph has N nodes and adjacent edges E are arranged between every two nodes, the whole knowledge graph is composed of { V }i，Eij：i<n，j<n }, is used as a reference. In one embodiment, the graph neural network model training process is as follows:
1. user input data S for knowledge points K at a timek。
2. We use a recurrent neural network encoder pair SkAfter processing, the code B is obtainedk。
3. We will turn BkAs an input value for the graph neural network model, the following is calculated:
where f (-) is a non-linear equation, we use LeakyReLU for implementation, W(k)And
respectively representing the updated weight matrix of the vertex and the edge, alpha is the forgetfulness of a proportional parameter determining system, and | V | represents the base number of the current vertex. The obtained result is stored in the corresponding node Vk。
4. We use gradient descent to train the graph neural network so that its stored vertex codes can maximally represent the information of the stored knowledge.
5. We use a de-coder to translate vertex-coded information, as calculated:
In addition, after the training of the graph neural network model is completed, the cycle neural network and the graph neural network are frozen (i.e., the parameters of the trained model are kept unchanged), and the fully-connected neural network is trained.
It is particularly supplementary to note that knowledge-graph updates also use large data. All user information is extracted by big data and different user modes are identified, the side weight is distributed in a mode identification mode when the relation of knowledge graph relations is established, and the high-efficiency cold start and the completely differentiated knowledge storage and extraction process are achieved by combining the historical records of the users.
The technical scheme of the embodiment combines a Recurrent Neural Network (RNN), a convolutional neural network (GNN), a knowledge graph and big data. The method is a novel knowledge storage system based on the artificial neural network and a method for extracting information in a differentiation mode. Compared with the prior art, the learning system of the scheme adopts the integration of a plurality of sets of self-adaptive artificial neural networks and big data, and can provide an optimized learning path for each user according to the learning conditions and historical data of the individual user and all users. The data storage system based on the Graph Attention Neural Network (Graph Attention Neural Network) realized by the method not only can link different knowledge nodes, but also can determine the weight of the link according to different knowledge point relationships.
Further, referring to fig. 1, according to a second aspect of the present embodiment, there is provided a storage medium. The storage medium comprises a stored program, wherein the method of any of the above is performed by a processor when the program is run.
Therefore, according to this embodiment, the cyclic neural network model is used to determine the coding information corresponding to the knowledge points included in the learning data information, and then the graph neural network model is used to calculate the coding information, so as to update the knowledge graph formed by the knowledge nodes and determine the grasping condition of the user on each knowledge node in the updated knowledge graph. Therefore, compared with the prior art, the scheme can continuously update the knowledge graph according to the knowledge points contained in the learning data information, and further can analyze the mastering condition of the knowledge nodes by the user more flexibly. In addition, the knowledge graph is constructed by knowledge nodes, so that association exists among the knowledge points, the mastering conditions of a plurality of associated knowledge points can be further determined, and the learning condition can be monitored more accurately. The technical effect of flexibly and accurately determining the mastering condition of each knowledge node by the user is achieved. And the technical problems that the learning condition monitoring mode of the user is fixed and the learning conditions of a plurality of associated knowledge points cannot be monitored in the prior art are solved.
It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.
Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
Example 2
Fig. 4 shows an apparatus 400 for determining a learning situation according to the present embodiment, the apparatus 400 corresponding to the method according to the first aspect of embodiment 1. Referring to fig. 4, the apparatus 400 includes: a data receiving module 410, configured to receive learning data information generated in a learning process input by a user; the coding information determining module 420 is configured to calculate learning data information by using a pre-trained recurrent neural network model, and determine coding information corresponding to a knowledge point included in the learning data information; and a learning condition determining module 430, configured to update a preset knowledge graph according to the coding information by using a pre-trained graph neural network model, and determine a grasping condition of a user on each knowledge node in the updated knowledge graph, where each knowledge node in the preset knowledge graph stores coding information of each knowledge point related to the user.
Optionally, the learning condition determining module 430 includes: the weight determining submodule is used for extracting the relationship strength among all knowledge nodes in the preset knowledge graph from big data analysis by using the graph neural network model to serve as the updating weight for updating the preset knowledge graph; the node adding submodule is used for generating a new knowledge node to be added to a preset knowledge graph, wherein the new knowledge node is used for storing coding information corresponding to the knowledge point contained in the learning data information; and the knowledge graph updating submodule is used for adding the new knowledge node to the preset knowledge graph and updating the preset knowledge graph according to the updating weight.
Optionally, the update weight includes a node update weight of each knowledge node in a preset knowledge graph and an edge update weight of each edge, and the knowledge graph update sub-module includes: the calculation unit is used for recalculating the coding information stored by each knowledge node in the knowledge graph to which the new knowledge node is added according to the node update weight and the edge update weight; and the updating unit is used for respectively storing the recalculated results to each knowledge node in the knowledge graph to which the new knowledge node is added.
Optionally, the learning condition determining module 430 includes: the decoding submodule is used for respectively performing decoding operation on the coding information stored by each knowledge node in the updated knowledge graph by using a preset decoding network; the first output submodule is used for inputting a result obtained by the decoding operation into a preset fully-connected neural network and outputting the mastering condition of each knowledge node in the updated knowledge graph by a user; and the second output submodule is used for determining the grasping condition of the user on each knowledge node in the updated knowledge graph according to the historical grasping condition of the user, the current learning progress information and the output grasping condition.
Optionally, the apparatus 400 further comprises: the learning path determining module is used for determining a learning path corresponding to the user according to the mastering condition of the user on each knowledge node in the updated knowledge graph; and the pushing submodule is used for pushing the learning path to the terminal equipment of the user.
Optionally, the encoding information determining module 420 includes: the characteristic extraction submodule is used for extracting characteristic information corresponding to the knowledge points contained in the learning data information by utilizing a pre-trained recurrent neural network model; and the coding submodule is used for coding the characteristic information by using a preset coding network and determining the coding information of the knowledge points.
Optionally, the apparatus 400 further comprises a training module for training the neural network model according to the following steps: acquiring training coding information for training, wherein the training coding information is determined by calculating learning data information for training by a recurrent neural network model; inputting the training coding information into a graph neural network model for calculation; and carrying out optimization training on the graph neural network model by using a gradient descent mode according to the calculation result.
Therefore, according to this embodiment, the cyclic neural network model is used to determine the coding information corresponding to the knowledge points included in the learning data information, and then the graph neural network model is used to calculate the coding information, so as to update the knowledge graph formed by the knowledge nodes and determine the grasping condition of the user on each knowledge node in the updated knowledge graph. Therefore, compared with the prior art, the scheme can continuously update the knowledge graph according to the knowledge points contained in the learning data information, and further can analyze the mastering condition of the knowledge nodes by the user more flexibly. In addition, the knowledge graph is constructed by knowledge nodes, so that association exists among the knowledge points, the mastering conditions of a plurality of associated knowledge points can be further determined, and the learning condition can be monitored more accurately. The technical effect of flexibly and accurately determining the mastering condition of each knowledge node by the user is achieved. And the technical problems that the learning condition monitoring mode of the user is fixed and the learning conditions of a plurality of associated knowledge points cannot be monitored in the prior art are solved.
Example 3
Fig. 5 shows an apparatus 500 for determining a learning situation according to the present embodiment, the apparatus 500 corresponding to the method according to the first aspect of embodiment 1. Referring to fig. 5, the apparatus 500 includes: a processor 510; and a memory 520 coupled to processor 510 for providing processor 510 with instructions to process the following process steps: receiving learning data information input by a user and generated in a learning process; calculating learning data information by using a pre-trained recurrent neural network model, and determining coding information corresponding to a knowledge point contained in the learning data information; and updating the preset knowledge graph according to the coding information by using a pre-trained graph neural network model, and determining the mastering condition of each knowledge node in the updated knowledge graph by the user, wherein the coding information of each knowledge point related to the user is stored in each knowledge node in the preset knowledge graph respectively.
Optionally, the operation of updating the preset knowledge graph according to the encoded information by using a pre-trained graph neural network model includes: extracting relationship strength among all knowledge nodes in the preset knowledge graph from big data analysis by using a graph neural network model as an updating weight for updating the preset knowledge graph; generating a new knowledge node to be added to a preset knowledge graph, wherein the new knowledge node is used for storing coding information corresponding to the knowledge point contained in the learning data information; and adding the new knowledge node to the preset knowledge graph, and updating the preset knowledge graph according to the updating weight.
Optionally, the updating weights include node updating weights of all knowledge nodes in a preset knowledge graph and edge updating weights of all edges, and the operation of updating the preset knowledge graph according to the updating weights includes: recalculating the coding information stored by each knowledge node in the knowledge graph added with the new knowledge node according to the node update weight and the edge update weight; and respectively storing the recalculated result to each knowledge node in the knowledge graph added with the new knowledge node.
Optionally, the operation of determining the grasping condition of the user on each knowledge node in the updated knowledge graph by using a pre-trained graph neural network model includes: respectively carrying out decoding operation on the coding information stored by each knowledge node in the updated knowledge graph by using a preset decoding network; inputting the result obtained by the decoding operation into a preset fully-connected neural network, and outputting the mastering condition of each knowledge node in the updated knowledge graph by a user; and determining the grasping condition of the user on each knowledge node in the updated knowledge graph according to the historical grasping condition of the user, the current learning progress information and the output grasping condition.
Optionally, the memory 520 is further configured to provide the processor 510 with instructions to process the following process steps: determining a learning path corresponding to the user according to the mastering condition of the user on each knowledge node in the updated knowledge graph; and pushing the learning path to the terminal device of the user.
Optionally, the calculating the learning data information by using a pre-trained recurrent neural network model, and determining the coding information corresponding to the knowledge point included in the learning data information, includes: extracting characteristic information corresponding to a knowledge point contained in the learning data information by using a pre-trained recurrent neural network model; and coding the characteristic information by using a preset coding network to determine the coding information of the knowledge points.
Optionally, training the neural network model according to the following steps: acquiring training coding information for training, wherein the training coding information is determined by calculating learning data information for training by a recurrent neural network model; inputting the training coding information into a graph neural network model for calculation; and carrying out optimization training on the graph neural network model by using a gradient descent mode according to the calculation result.
Therefore, according to this embodiment, the cyclic neural network model is used to determine the coding information corresponding to the knowledge points included in the learning data information, and then the graph neural network model is used to calculate the coding information, so as to update the knowledge graph formed by the knowledge nodes and determine the grasping condition of the user on each knowledge node in the updated knowledge graph. Therefore, compared with the prior art, the scheme can continuously update the knowledge graph according to the knowledge points contained in the learning data information, and further can analyze the mastering condition of the knowledge nodes by the user more flexibly. In addition, the knowledge graph is constructed by knowledge nodes, so that association exists among the knowledge points, the mastering conditions of a plurality of associated knowledge points can be further determined, and the learning condition can be monitored more accurately. The technical effect of flexibly and accurately determining the mastering condition of each knowledge node by the user is achieved. And the technical problems that the learning condition monitoring mode of the user is fixed and the learning conditions of a plurality of associated knowledge points cannot be monitored in the prior art are solved.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (10)
1. A method of determining learning conditions, comprising:
receiving learning data information input by a user and generated in a learning process;
calculating the learning data information by utilizing a pre-trained recurrent neural network model, and determining coding information corresponding to a knowledge point contained in the learning data information; and
and updating a preset knowledge graph according to the coding information by using a pre-trained graph neural network model, and determining the mastering condition of the user on each knowledge node in the updated knowledge graph, wherein the coding information of each knowledge point related to the user is stored in each knowledge node in the preset knowledge graph respectively.
2. The method of claim 1, wherein the operation of updating the predetermined knowledge-graph according to the encoded information by using a pre-trained graph neural network model comprises:
extracting relationship strength among all knowledge nodes in the preset knowledge graph from big data analysis by using the graph neural network model as an updating weight for updating the preset knowledge graph;
generating a new knowledge node to be added to the preset knowledge graph, wherein the new knowledge node is used for storing coding information corresponding to the knowledge point contained in the learning data information; and
and adding the new knowledge node to the preset knowledge graph, and updating the preset knowledge graph according to the updating weight.
3. The method according to claim 2, wherein the update weights include a node update weight of each knowledge node and an edge update weight of each edge in the preset knowledge graph, and the operation of updating the preset knowledge graph according to the update weights includes:
recalculating the coding information stored by each knowledge node in the knowledge graph to which the new knowledge node is added according to the node update weight and the edge update weight; and
and respectively storing the recalculated result to each knowledge node in the knowledge graph to which the new knowledge node is added.
4. The method of claim 3, wherein the operation of determining the user's mastery of each knowledge node in the updated knowledge-graph using a pre-trained graph neural network model comprises:
respectively carrying out decoding operation on the updated coding information stored by each knowledge node in the knowledge graph by using a preset decoding network;
inputting the result obtained by the decoding operation into a preset fully-connected neural network, and outputting the mastering condition of the user on each knowledge node in the updated knowledge graph; and
and determining the grasping condition of the user on each knowledge node in the updated knowledge graph according to the historical grasping condition of the user, the current learning progress information and the output grasping condition.
5. The method of claim 1, further comprising:
determining a learning path corresponding to the user according to the mastery condition of the user on each knowledge node in the updated knowledge graph; and
and pushing the learning path to the terminal equipment of the user.
6. The method according to claim 1, wherein the calculating the learning data information by using a pre-trained recurrent neural network model to determine the coding information corresponding to the knowledge points included in the learning data information comprises:
extracting characteristic information corresponding to the knowledge points contained in the learning data information by utilizing a pre-trained recurrent neural network model; and
and coding the characteristic information by using a preset coding network to determine the coding information of the knowledge points.
7. The method of claim 1, further comprising training the graph neural network model according to the steps of:
acquiring training coding information for training, wherein the training coding information is determined by calculating learning data information for training by the recurrent neural network model;
inputting the training coding information into the graph neural network model for calculation; and
and performing optimization training on the graph neural network model by using a gradient descent mode according to a calculation result.
8. A storage medium comprising a stored program, wherein the method of any one of claims 1 to 7 is performed by a processor when the program is run.
9. An apparatus for determining learning conditions, comprising:
the data receiving module is used for receiving learning data information input by a user and generated in the learning process;
the coding information determining module is used for calculating the learning data information by utilizing a pre-trained recurrent neural network model and determining coding information corresponding to the knowledge points contained in the learning data information; and
and the learning condition determining module is used for updating a preset knowledge graph according to the coding information by using a pre-trained graph neural network model and determining the grasping condition of the user on each knowledge node in the updated knowledge graph, wherein the coding information of each knowledge node related to the user is stored in each knowledge node in the preset knowledge graph respectively.
10. An apparatus for determining learning conditions, comprising:
a processor; and
a memory coupled to the processor for providing instructions to the processor for processing the following processing steps:
receiving learning data information input by a user and generated in a learning process;
calculating the learning data information by utilizing a pre-trained recurrent neural network model, and determining coding information corresponding to a knowledge point contained in the learning data information; and
and updating a preset knowledge graph according to the coding information by using a pre-trained graph neural network model, and determining the mastering condition of the user on each knowledge node in the updated knowledge graph, wherein the coding information of each knowledge point related to the user is stored in each knowledge node in the preset knowledge graph respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011157699.4A CN112328804A (en) | 2020-10-26 | 2020-10-26 | Method, apparatus and storage medium for determining learning situation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011157699.4A CN112328804A (en) | 2020-10-26 | 2020-10-26 | Method, apparatus and storage medium for determining learning situation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112328804A true CN112328804A (en) | 2021-02-05 |
Family
ID=74311022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011157699.4A Pending CN112328804A (en) | 2020-10-26 | 2020-10-26 | Method, apparatus and storage medium for determining learning situation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112328804A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160133162A1 (en) * | 2014-11-10 | 2016-05-12 | International Business Machines Corporation | Student specific learning graph |
| CN108615423A (en) * | 2018-06-21 | 2018-10-02 | 中山大学新华学院 | Instructional management system (IMS) on a kind of line based on deep learning |
| KR20190004429A (en) * | 2017-07-04 | 2019-01-14 | 주식회사 알고리고 | Method and apparatus for determining training of unknown data related to neural networks |
| CN110263179A (en) * | 2019-06-12 | 2019-09-20 | 湖南酷得网络科技有限公司 | Learning path method for pushing, device, computer equipment and storage medium |
| CN110544414A (en) * | 2019-07-31 | 2019-12-06 | 安徽淘云科技有限公司 | knowledge graph processing method and device |
| CN110852071A (en) * | 2019-11-08 | 2020-02-28 | 科大讯飞股份有限公司 | Knowledge point detection method, device, equipment and readable storage medium |
| CN111582694A (en) * | 2020-04-29 | 2020-08-25 | 腾讯科技（深圳）有限公司 | Learning evaluation method and device |
| CN111753098A (en) * | 2020-06-23 | 2020-10-09 | 陕西师范大学 | Teaching method and system based on cross-media dynamic knowledge graph |
-
2020
- 2020-10-26 CN CN202011157699.4A patent/CN112328804A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160133162A1 (en) * | 2014-11-10 | 2016-05-12 | International Business Machines Corporation | Student specific learning graph |
| KR20190004429A (en) * | 2017-07-04 | 2019-01-14 | 주식회사 알고리고 | Method and apparatus for determining training of unknown data related to neural networks |
| CN108615423A (en) * | 2018-06-21 | 2018-10-02 | 中山大学新华学院 | Instructional management system (IMS) on a kind of line based on deep learning |
| CN110263179A (en) * | 2019-06-12 | 2019-09-20 | 湖南酷得网络科技有限公司 | Learning path method for pushing, device, computer equipment and storage medium |
| CN110544414A (en) * | 2019-07-31 | 2019-12-06 | 安徽淘云科技有限公司 | knowledge graph processing method and device |
| CN110852071A (en) * | 2019-11-08 | 2020-02-28 | 科大讯飞股份有限公司 | Knowledge point detection method, device, equipment and readable storage medium |
| CN111582694A (en) * | 2020-04-29 | 2020-08-25 | 腾讯科技（深圳）有限公司 | Learning evaluation method and device |
| CN111753098A (en) * | 2020-06-23 | 2020-10-09 | 陕西师范大学 | Teaching method and system based on cross-media dynamic knowledge graph |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109815339B (en) | Knowledge extraction method and device based on TextCNN, computer equipment and storage medium | |
| CN110599838A (en) | Mathematics automatic question setting method and device | |
| CN108090218A (en) | Conversational system generation method and device based on deeply study | |
| CN111178537B (en) | Feature extraction model training method and device | |
| CN115936180A (en) | Photovoltaic power generation power prediction method and device and computer equipment | |
| CN113094284A (en) | Application fault detection method and device | |
| CN115205736A (en) | Video data identification method and device, electronic equipment and storage medium | |
| CN111523798B (en) | Automatic modeling method, device, system and electronic equipment thereof | |
| CN116435995A (en) | Time series processing method, computer readable storage medium and electronic device | |
| CN112836807A (en) | Data processing method and device based on neural network | |
| CN112328804A (en) | Method, apparatus and storage medium for determining learning situation | |
| CN115082800B (en) | Image segmentation method | |
| CN115424725A (en) | Data analysis method and device, storage medium and processor | |
| CN114971053A (en) | Training method and device for online prediction model of network line loss rate of low-voltage transformer area | |
| CN114154415A (en) | Equipment life prediction method and device | |
| CN115203412A (en) | Emotion viewpoint information analysis method and device, storage medium and electronic equipment | |
| CN112749150B (en) | Error labeling data identification method, device and medium | |
| CN113868415A (en) | Knowledge base generation method and device, storage medium and electronic equipment | |
| CN113313615A (en) | Method and device for quantitatively grading and grading enterprise judicial risks | |
| CN110826582B (en) | Image feature training method, device and system | |
| CN116049448B (en) | Knowledge-graph-based construction and identification method of electric energy quality disturbance identification model | |
| CN111538914A (en) | Address information processing method and device | |
| KR102282126B1 (en) | Method for factorization through constant decomposition and a device of providing the same | |
| CN117993504A (en) | Data set construction method, language model determination method and information processing method | |
| CN113919604A (en) | Time series data prediction method, device, storage medium and processor |
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 |