CN111898020A - Knowledge learning system recommendation method, device and medium based on BERT and LSTM - Google Patents
Knowledge learning system recommendation method, device and medium based on BERT and LSTM Download PDFInfo
- Publication number
- CN111898020A CN111898020A CN202010560073.1A CN202010560073A CN111898020A CN 111898020 A CN111898020 A CN 111898020A CN 202010560073 A CN202010560073 A CN 202010560073A CN 111898020 A CN111898020 A CN 111898020A
- Authority
- CN
- China
- Prior art keywords
- learning
- knowledge
- target
- vector
- learning system
- 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 57
- 239000013598 vector Substances 0.000 claims abstract description 110
- 238000012549 training Methods 0.000 claims description 33
- 230000015654 memory Effects 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 21
- 230000008569 process Effects 0.000 claims description 19
- 238000003860 storage Methods 0.000 claims description 12
- 230000000750 progressive effect Effects 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 12
- 238000013528 artificial neural network Methods 0.000 description 9
- 238000005516 engineering process Methods 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 238000011156 evaluation Methods 0.000 description 7
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000003058 natural language processing Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000002457 bidirectional effect Effects 0.000 description 2
- 238000013461 design 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
- 230000006399 behavior Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000002250 progressing effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000126 substance Substances 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
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- 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
- G06N3/045—Combinations of networks
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Strategic Management (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The application discloses a method, equipment and a medium for recommending a knowledge learning system based on BERT and LSTM, wherein the method comprises the following steps: receiving a target knowledge point and a target mastering degree of user planning learning; acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector; inputting the input vector into a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user. The embodiment of the invention utilizes the model to carry out personalized recommendation according to the specific requirements of the learner, can provide a learning path which is more accordant with the capability of the learner, and enables the learner to acquire and master the forming capability of the knowledge points more quickly. In addition, the learner learns and feeds back in time according to the recommended learning path, and the recommendation model is continuously optimized.
Description
Technical Field
The application relates to the technical field of deep learning, in particular to a knowledge learning system recommendation method, device and medium based on BERT and LSTM.
Background
In recent years, the development of artificial intelligence technology is rapid, the commercialization speed of the technology is beyond expectations, and artificial intelligence brings subversive changes to the whole society and becomes an important development strategy for countries in the future. Particularly, the algorithm evolution taking deep learning as a core has the super-strong evolutionary capability, and under the support of big data, a large-scale neural network similar to a human brain structure is obtained through training and construction, so that various problems can be solved.
With the rapid development of internet technology, the traditional education industry also caters for a new mode of internet +, the development of education informatization also changes the learning mode of learners from traditional classroom learning to online learning, and the online learning is used as the classic application of internet + education, which deeply influences the existing education concept and education method. Especially from the internet to the mobile internet, a cross-space-time life, work and learning mode is created, and a knowledge acquisition mode is fundamentally changed.
However, in the face of a large explosion of knowledge information, learners can hardly learn some cross-domain knowledge by self judgment.
Disclosure of Invention
The embodiment of the specification provides a method, equipment and a medium for recommending a knowledge learning system based on BERT and LSTM, which are used for solving the following technical problems in the prior art:
how to make a learning scheme and a path for a learner according to the existing online education big data.
The embodiment of the specification adopts the following technical scheme:
the first aspect of the embodiment of the invention provides a knowledge learning system recommendation method based on BERT and LSTM, which comprises the following steps:
receiving a target knowledge point and a target mastering degree of user planning learning;
acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector;
inputting the input vector into a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user.
The embodiment of the invention utilizes the model to carry out personalized recommendation according to the specific requirements of the learner, can provide a learning path which is more accordant with the capability of the learner, and enables the learner to acquire and master the forming capability of the knowledge points more quickly.
In one example, the inputting the input vector to a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user includes:
sending the input vector corresponding to the historical learning data to a trained context vector encoder for processing, wherein the context vector encoder is realized based on BERT and is a part of the knowledge learning system recommendation model;
sending the input vector corresponding to the target knowledge point and the target mastering degree to a trained learning path predictor for processing, wherein the learning path predictor is realized based on LSTM and is another part of the knowledge learning system recommendation model;
determining a learned path output to the user through a processing result of the context vector encoder and a learned path predictor.
The embodiment of the invention respectively processes different parameters based on the characteristic that two neural networks of BERT and LSTM are not communicated, thereby improving the accuracy of prediction.
In one example, the inputting the input vector to a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user includes:
sending the input vector corresponding to the historical learning data to the trained context vector encoder, performing context-based semantic processing to obtain a knowledge point coding vector sequence,
and sending the input vector corresponding to the target knowledge point and the target mastery degree and the knowledge point coding vector sequence to the trained learning path predictor for processing, and determining a learning path output to the user.
The embodiment of the invention takes the output vector of the BERT neural network as the input vector of the LSTM neural network, and integrates and considers the target knowledge point, the target learning degree and the mastery degree of the knowledge point which is learned by the learner, so that the obtained learning path is more effective for the learner.
In one example, further comprising: training a knowledge learning system recommendation model through a data set, wherein the training step comprises the following steps:
training a context vector encoder in the knowledge learning system recommendation model;
and jointly training the trained context vector encoder and a learning path predictor in the knowledge learning system to obtain a trained knowledge learning system recommendation model.
In one example, the training the context vector coder and a learning path predictor in the knowledge learning system together to obtain a trained knowledge learning system recommendation model includes:
and training the learning path predictor by adjusting the parameters of the trained context vector encoder to obtain a trained knowledge learning system recommendation model.
In one example, the training a context vector encoder in the knowledge learning system recommendation model includes:
extracting a knowledge point vector and a mastery degree vector of the learner based on the collected learning resource data of the learner;
forming a learning path vector sequence based on the collected learning process data of the learner, the knowledge point vector and the mastery degree vector;
training the context vector editor with the sequence of learned path vectors.
In one example, the extracting a knowledge point vector and a mastery level vector of the learner comprises:
and evaluating the examination questions according to the knowledge points designed by the experts, and acquiring the mastery degree of the learners for learning the knowledge points to form a mastery degree vector.
In one example, further comprising:
designing a progressive knowledge point sequence according to expert experience;
and training the knowledge learning system recommendation model through the progressive knowledge point sequence to obtain a priori knowledge-based knowledge learning system recommendation model.
A second aspect of an embodiment of the present invention provides a BERT and LSTM-based knowledge learning system recommendation apparatus, including:
a processor; and
a memory communicatively coupled to the processor, the memory having stored thereon instructions executable by the processor to enable the processor to:
receiving a target knowledge point and a target mastering degree of user planning learning;
acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector;
inputting the input vector into a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user.
A third aspect of an embodiment of the present invention provides a BERT and LSTM-based knowledge learning system recommendation non-volatile computer storage medium storing computer-executable instructions configured to:
receiving a target knowledge point and a target mastering degree of user planning learning;
acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector;
inputting the input vector into a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user.
The embodiment of the specification adopts at least one technical scheme which can achieve the following beneficial effects:
a knowledge learning system recommendation model is designed based on BERT and LSTM technologies, the characteristics of online learning are fully considered, especially online learning can be repeatedly carried out, meanwhile, the knowledge learning path of excellent learners is combined, the conditions, the learning history and the actual capacity of learners of different levels are considered, the online examination and the evaluation of the designed online knowledge points are taken as the basis, and the knowledge learning system recommendation model is formed through training based on massive learning resource data and online learning process data of learners.
Compared with the traditional recommendation mode based on the similarity of learners, the knowledge learning recommendation method based on deep learning better conforms to the learning characteristics of online learning, and the learning habits of learners gradually progressing and knowledge from shallow to deep can be better met by adding the experience design of field experts in the training process;
the model is used for carrying out personalized recommendation according to the specific requirements of the learner, so that a learning path which is more consistent with the capability of the learner can be provided, and the learner can acquire and master the knowledge point forming capability more quickly. In addition, the learner learns and feeds back in time according to the recommended learning path, and the recommendation model is continuously optimized.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a schematic flow chart of a method provided by an embodiment of the present invention;
FIG. 2 is a schematic diagram of a knowledge learning system recommendation model provided by an embodiment of the present invention;
FIG. 3 is a schematic flow chart of a training method of a knowledge learning system recommendation model according to an embodiment of the present invention;
fig. 4 is a schematic diagram of an apparatus framework according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present disclosure more apparent, the technical solutions of the present disclosure will be clearly and completely described below with reference to the specific embodiments of the present disclosure and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments obtained by a person skilled in the art without making any inventive step based on the embodiments in the description belong to the protection scope of the present application.
A large number of digital learning resources are continuously emerged on the network, online learning, online examination and online evaluation become important learning means, a learning mode based on network resources is generally concerned by more and more learners, the mass learning resources are emerged, meanwhile, a large number of behavior data generated by the learners through online learning are obtained, richer education big data are formed to a certain extent, and the possibility is brought to more effective knowledge system selection and learning path planning of the learners.
Under the circumstance, how to utilize the deep learning technology, realize the personalized recommendation of knowledge learning based on BERT and LSTM, find a learning mode more suitable for learners, and enable learners to acquire and master knowledge forming ability more quickly becomes a problem which needs to be solved urgently.
The Long Short-Term Memory network (LSTM) is a time-cycle neural network, which is specially designed to solve the Long-Term dependence problem of the general RNN (cyclic neural network), and all RNNs have a chain form of repeated neural network modules. LSTM is suitable for processing and predicting significant events of very long intervals and delays in a time series.
BERT (Bidirectional Encoder reproduction from Transformers), which is an encoding (english: Encoder) of a Bidirectional Transformer, is a natural language processing mode with revolutionary significance compared with a traditional natural language processing mode, has important application in the field of natural language processing, and brings inspiration to many existing computer logic frameworks and training methods. In particular, the ability to abstract continuous long sequence features is one of the most important language processing models at present.
The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
The embodiment of the application provides a knowledge learning system recommendation method and a corresponding scheme based on BERT and LSTM, which are characterized in that massive learning resource data and learner online learning process data are collected, content knowledge points in the learning resource data are extracted according to learning resource contents, online examination and designed online knowledge point evaluation are taken as the basis, a neural network is designed by utilizing BERT and LSTM technologies in combination with the learning process of a learner, a knowledge learning system recommendation model is formed through training, and an optimal learning path which is personalized for the learner and accords with knowledge point learning is found based on specific knowledge points and the current situation and the actual capability of the learner.
Fig. 1 is a schematic flow chart of a method according to an embodiment of the present invention. As shown, the method comprises:
s101, receiving a target knowledge point and a target mastering degree of user planning learning;
s102, acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector;
s103, inputting the input vector to a trained knowledge learning system recommendation model based on BERT and LSTM, so as to output a recommended learning path to the user.
According to an embodiment of the present invention, in step S103, the input vector is input to a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user, and the specific steps include:
sending the input vector corresponding to the historical learning data to a trained context vector encoder for processing, wherein the context vector encoder is realized based on BERT and is a part of the knowledge learning system recommendation model;
sending the input vector corresponding to the target knowledge point and the target mastering degree to a trained learning path predictor for processing, wherein the learning path predictor is realized based on LSTM and is another part of the knowledge learning system recommendation model;
determining a learned path output to the user through a processing result of the context vector encoder and a learned path predictor.
In some preferred embodiments of the present invention, fig. 2 is a schematic diagram of a knowledge learning system recommendation model provided in an embodiment of the present invention, and as shown in fig. 2, the working mode of a trained knowledge learning system recommendation model is as follows:
sending the input vector corresponding to the historical learning data to the trained context vector encoder, performing context-based semantic processing to obtain a knowledge point coding vector sequence,
and sending the input vector corresponding to the target knowledge point and the target mastery degree and the knowledge point coding vector sequence to the trained learning path predictor for processing, and determining a learning path output to the user. Fig. 3 is a schematic flow chart of a training method for recommending a model by a knowledge learning system according to an embodiment of the present invention, and a detailed description is given below of a model training process according to an embodiment of the present invention with reference to fig. 3.
The first step is as follows: the method comprises the steps of collecting massive online learning resource data and learning knowledge point process data of a learner, providing online learning services for the learner by an online learning platform, and simultaneously recording the learning process data of the learner.
The second step is that: and extracting knowledge points of the massive online learning resource data based on the massive online learning resource data to form knowledge point vectors, wherein the one-hot vectors are adopted by the knowledge point vectors extracted by the massive online learning resource data and are used for describing different knowledge points.
The third step: evaluating examination questions according to the knowledge points designed by the experts, and acquiring the mastery degree of the learners for learning the knowledge points to form a mastery degree vector; learners learn knowledge points through online learning, participate in online examination and online evaluation, and evaluate the mastering degree and comprehensive capacity of the learned knowledge points;
specifically, the knowledge point evaluation examination questions are the examination questions of the knowledge points designed by the experts and are continuously perfected; the knowledge point mastering degree is a one-dimensional vector with a fixed length and represents the level of the mastered knowledge point.
The fourth step: and extracting the learning process of the knowledge points of the learner based on the collected mass learning process data of the learner to form a learning path vector sequence of (knowledge point vector + mastery degree vector). The learning path vector sequence is a knowledge point learning process sequence of online learning, and the mastery degree of the learning path vector sequence is obtained through knowledge point evaluation;
the fifth step: the knowledge point context vector encoder based on the BERT is trained, and knowledge point coding vectors are generated based on a learning path vector sequence of a learner mainly through methods such as a Masked partial vector sequence and a next sequence Prediction learning segmentation sequence relation. The main body of the knowledge point context vector encoder is based on a BERT structure, and learns the knowledge point encoding vector based on the context through an input knowledge point vector sequence; i.e. the knowledge points are reordered according to a certain logic.
Preferably, the learning path vector sequence of the excellent learner is selected as a positive example, and the learning path sequence which cannot reach a better mastery degree or is finally terminated is selected as a negative example, so that the context vector encoder is trained based on mass data.
And a sixth step: and forming a large network by the trained knowledge point context vector coder and the learning path predictor for training. The learning path predictor main body is LSTM, and forms a learning path prediction model through training and learning by inputting a target knowledge point planned to be learned by a learner, a degree of expecting to master the knowledge point and a condition (history learning record) of the knowledge point currently mastered by the learner, and outputs a learning path sequence of the knowledge point.
Preferably, in the training process of the knowledge learning system recommendation model, the knowledge point context vector encoder finely adjusts parameters, the learning path predictor comprehensively considers the knowledge points, the learning degree and the mastery degree of the knowledge points learned by the learner, and the network model of the learning path predictor memorizes the state of the currently output knowledge point sequence and considers the currently output recommendation sequence.
The seventh step: through the training process, training is carried out based on mass data, and finally a knowledge learning system recommendation model is formed.
Optionally, a knowledge point sequence with gradual progression and knowledge from shallow to deep is designed according to the field expert experience and used for model training to form a knowledge system learning recommendation model added with prior knowledge.
The embodiment of the invention designs the neural network based on the BERT and LSTM technologies, fully considers the knowledge learning path of excellent learners, simultaneously considers the self condition, the learning history and the actual ability of learners with different levels, forms a knowledge learning system recommendation model by training based on online examination and designed online knowledge point evaluation and combining the characteristics of online learning, and recommends the optimal learning path which meets the learner ability and the personalized requirement according to the specific requirement of the learner.
Eighth step: the learner inputs a target knowledge point and a target mastery degree to be learned in the plan, acquires historical learning data of the learner through the online learning platform, and encodes the historical learning data to form a vector to be input into a knowledge system learning recommendation model.
The ninth step: the knowledge system learning recommendation model outputs a recommended learning path according to the input vector of the learner.
The tenth step: and the learner learns according to the recommended learning path and feeds back the learning path for continuous optimization of the model.
Based on the same idea, some embodiments of the present application further provide a device and a non-volatile computer storage medium corresponding to the above method.
Fig. 4 is a schematic diagram of an apparatus framework provided in an embodiment of the present specification, and a knowledge learning system recommendation apparatus for BERT and LSTM includes:
at least one processor; and the number of the first and second groups,
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to:
receiving a target knowledge point and a target mastering degree of user planning learning;
acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector;
inputting the input vector into a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user.
Some embodiments of the present application provide a knowledge learning system recommendation non-volatile computer storage medium corresponding to one of BERT and LSTM of fig. 1, having stored thereon computer-executable instructions configured to:
receiving a target knowledge point and a target mastering degree of user planning learning;
acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector;
inputting the input vector into a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user.
The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the device and media embodiments, the description is relatively simple as it is substantially similar to the method embodiments, and reference may be made to some descriptions of the method embodiments for relevant points.
The device and the medium provided by the embodiment of the application correspond to the method one to one, so the device and the medium also have the similar beneficial technical effects as the corresponding method, and the beneficial technical effects of the method are explained in detail above, so the beneficial technical effects of the device and the medium are not repeated herein.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is directed to methods, apparatus (systems), and computer program products according to embodiments of the present invention
A flowchart and/or block diagram of an article. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A knowledge learning system recommendation method based on BERT and LSTM is characterized by comprising the following steps:
receiving a target knowledge point and a target mastering degree of user planning learning;
acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector;
inputting the input vector into a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user.
2. The method of claim 1, wherein inputting the input vector to a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user comprises:
sending the input vector corresponding to the historical learning data to a trained context vector encoder for processing, wherein the context vector encoder is realized based on BERT and is a part of the knowledge learning system recommendation model;
sending the input vector corresponding to the target knowledge point and the target mastering degree to a trained learning path predictor for processing, wherein the learning path predictor is realized based on LSTM and is another part of the knowledge learning system recommendation model;
determining a learned path output to the user through a processing result of the context vector encoder and a learned path predictor.
3. The method of claim 2, wherein inputting the input vector to a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user comprises:
sending the input vector corresponding to the historical learning data to the trained context vector encoder, performing context-based semantic processing to obtain a knowledge point coding vector sequence,
and sending the input vector corresponding to the target knowledge point and the target mastery degree and the knowledge point coding vector sequence to the trained learning path predictor for processing, and determining a learning path output to the user.
4. The method of claim 1, further comprising: training a knowledge learning system recommendation model through a data set, wherein the training step comprises the following steps:
training a context vector encoder in the knowledge learning system recommendation model;
and jointly training the trained context vector encoder and a learning path predictor in the knowledge learning system to obtain a trained knowledge learning system recommendation model.
5. The method of claim 4, wherein the co-training the trained context vector encoder with a learning path predictor in the knowledge learning system to obtain a trained knowledge learning system recommendation model comprises:
and training the learning path predictor by adjusting the parameters of the trained context vector encoder to obtain a trained knowledge learning system recommendation model.
6. The method of claim 4, wherein training a context vector encoder in the knowledge learning system recommendation model comprises:
extracting a knowledge point vector and a mastery degree vector of the learner based on the collected learning resource data of the learner;
forming a learning path vector sequence based on the collected learning process data of the learner, the knowledge point vector and the mastery degree vector;
training the context vector editor with the sequence of learned path vectors.
7. The method of claim 6, wherein extracting a knowledge point vector and a mastery level vector for a learner comprises:
and evaluating the examination questions according to the knowledge points designed by the experts, and acquiring the mastery degree of the learners for learning the knowledge points to form a mastery degree vector.
8. The method of claim 4, further comprising:
designing a progressive knowledge point sequence according to expert experience;
and training the knowledge learning system recommendation model through the progressive knowledge point sequence to obtain a priori knowledge-based knowledge learning system recommendation model.
9. A BERT and LSTM-based knowledge learning system recommendation device, comprising:
a processor; and
a memory communicatively coupled to the processor, the memory having stored thereon instructions executable by the processor to enable the processor to:
receiving a target knowledge point and a target mastering degree of user planning learning;
acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector;
inputting the input vector into a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user.
10. A BERT and LSTM-based knowledge learning system recommendation non-volatile computer storage medium having stored thereon computer-executable instructions configured to:
receiving a target knowledge point and a target mastering degree of user planning learning;
acquiring historical learning data of the user, and encoding the target knowledge point, the target mastering degree and the historical learning data to form an input vector;
inputting the input vector into a trained BERT and LSTM-based knowledge learning system recommendation model to output a recommended learning path to the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010560073.1A CN111898020A (en) | 2020-06-18 | 2020-06-18 | Knowledge learning system recommendation method, device and medium based on BERT and LSTM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010560073.1A CN111898020A (en) | 2020-06-18 | 2020-06-18 | Knowledge learning system recommendation method, device and medium based on BERT and LSTM |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111898020A true CN111898020A (en) | 2020-11-06 |
Family
ID=73207764
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010560073.1A Pending CN111898020A (en) | 2020-06-18 | 2020-06-18 | Knowledge learning system recommendation method, device and medium based on BERT and LSTM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111898020A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112734142A (en) * | 2021-04-02 | 2021-04-30 | 平安科技（深圳）有限公司 | Resource learning path planning method and device based on deep learning |
| CN113190747A (en) * | 2021-04-29 | 2021-07-30 | 西安理工大学 | Personalized resource recommendation method based on learning style and cognitive level |
| CN114117033A (en) * | 2022-01-25 | 2022-03-01 | 山东建筑大学 | Knowledge tracking method and system |
| WO2022121705A1 (en) * | 2020-12-10 | 2022-06-16 | 株式会社日立制作所 | Information processing method, apparatus and device |
| CN114647721A (en) * | 2022-05-23 | 2022-06-21 | 风林科技(深圳)有限公司 | Educational intelligent robot control method, device and medium |
| CN117573985A (en) * | 2024-01-16 | 2024-02-20 | 四川航天职业技术学院(四川航天高级技工学校) | Information pushing method and system applied to intelligent online education system |
| WO2024120504A1 (en) * | 2022-12-09 | 2024-06-13 | 华为技术有限公司 | Data processing method and related device |
| CN114491254B (en) * | 2022-01-24 | 2024-06-25 | 湖南大学 | Knowledge point recommendation method, knowledge point recommendation device and storage medium |
-
2020
- 2020-06-18 CN CN202010560073.1A patent/CN111898020A/en active Pending
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022121705A1 (en) * | 2020-12-10 | 2022-06-16 | 株式会社日立制作所 | Information processing method, apparatus and device |
| CN112734142A (en) * | 2021-04-02 | 2021-04-30 | 平安科技（深圳）有限公司 | Resource learning path planning method and device based on deep learning |
| CN112734142B (en) * | 2021-04-02 | 2021-07-02 | 平安科技（深圳）有限公司 | Resource learning path planning method and device based on deep learning |
| CN113190747A (en) * | 2021-04-29 | 2021-07-30 | 西安理工大学 | Personalized resource recommendation method based on learning style and cognitive level |
| CN114491254B (en) * | 2022-01-24 | 2024-06-25 | 湖南大学 | Knowledge point recommendation method, knowledge point recommendation device and storage medium |
| CN114117033A (en) * | 2022-01-25 | 2022-03-01 | 山东建筑大学 | Knowledge tracking method and system |
| CN114647721A (en) * | 2022-05-23 | 2022-06-21 | 风林科技(深圳)有限公司 | Educational intelligent robot control method, device and medium |
| CN114647721B (en) * | 2022-05-23 | 2022-09-16 | 风林科技(深圳)有限公司 | Educational intelligent robot control method, device and medium |
| WO2024120504A1 (en) * | 2022-12-09 | 2024-06-13 | 华为技术有限公司 | Data processing method and related device |
| CN117573985A (en) * | 2024-01-16 | 2024-02-20 | 四川航天职业技术学院(四川航天高级技工学校) | Information pushing method and system applied to intelligent online education system |
| CN117573985B (en) * | 2024-01-16 | 2024-04-05 | 四川航天职业技术学院(四川航天高级技工学校) | Information pushing method and system applied to intelligent online education system |
| CN118095106B (en) * | 2024-04-23 | 2024-06-28 | 珠海翔翼航空技术有限公司 | Method, system and equipment for predicting residual fuel value of airplane |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111898020A (en) | Knowledge learning system recommendation method, device and medium based on BERT and LSTM | |
| Nagatani et al. | Augmenting knowledge tracing by considering forgetting behavior | |
| Wang et al. | Deep knowledge tracing on programming exercises | |
| CN111126552B (en) | Intelligent learning content pushing method and system | |
| CN111897934B (en) | Question-answer pair generation method and device | |
| CN107544960A (en) | A kind of inference method activated based on Variable-Bindings and relation | |
| Benedetto et al. | On the application of transformers for estimating the difficulty of multiple-choice questions from text | |
| CN117252047B (en) | Teaching information processing method and system based on digital twinning | |
| WO2017041066A1 (en) | Flexible, personalized student success modeling for institutions with complex term structures and competency-based education | |
| CN112528010A (en) | Knowledge recommendation method and device, computer equipment and readable storage medium | |
| CN115525744A (en) | Dialog recommendation system based on prompt learning method | |
| KR20220098698A (en) | Learning content recommendation system that predicts the user's correct answer probability using collaborative filtering based on latent factors and operation method thereof | |
| CN114971066A (en) | Knowledge tracking method and system integrating forgetting factor and learning ability | |
| CN114398556A (en) | Learning content recommendation method, device, equipment and storage medium | |
| Tang et al. | Predictive modelling of student behaviour using granular large-scale action data | |
| CN117252739A (en) | Method, system, electronic equipment and storage medium for evaluating paper | |
| CN117312628A (en) | Method and device for recommending exercise course, storage medium and electronic equipment | |
| CN114579869B (en) | Model training method and related product | |
| CN110008880A (en) | A kind of model compression method and device | |
| CN115617959A (en) | Question answering method and device | |
| Jang et al. | Machine Learning-Based Programming Analysis Model Proposal: Based on User Behavioral Analysis | |
| CN112256858A (en) | Double-convolution knowledge tracking method and system fusing question mode and answer result | |
| Friedland | Reproducibility and Experimental Design for Machine Learning on Audio and Multimedia Data | |
| CN114238613B (en) | Method and device for determining mastery degree of knowledge points and electronic equipment | |
| CN116502713B (en) | Knowledge tracking method for enhancing topic similarity embedding based on weighted element path |
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 |