Exercise recommendation method and device based on course knowledge graph and knowledge tracking

Abstract

The invention discloses an exercise recommendation method and device based on a course knowledge graph and knowledge tracking, and the method comprises the steps: constructing a subject knowledge graph based on obtained teaching reference data according to the knowledge characteristics of a target subject; constructing a knowledge tracker based on a DKT model according to the answer historical sequence and the current knowledge state of the student; and according to the subject knowledge graph and the knowledge tracker, generating an exercise recommendation result. According to the embodiment of the invention, learning behaviors and preferences of students can be deeply mined, accurate and diversified exercise recommendation can be provided, meanwhile, rich relation information and semantic content contained in the subject knowledge graph are fully utilized, clear and definite explanation is attached to each recommended exercise, and the learning efficiency is improved. Therefore, more efficient and personalized learning experience can be realized in the field of online education, and the method can be widely applied to the technical field of computers.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for recommending exercises based on course knowledge graph and knowledge tracking. Background Art

[0002] With the rapid development of Internet technology, people's demand for efficient and flexible learning methods is growing, which has led to the rapid rise of online education, becoming a major trend in the global education field. However, the prosperity of online education has also brought about the problems of information overload and difficulty in meeting personalized needs. How to accurately provide personalized learning experience based on the characteristics and needs of each student has become a key issue that needs to be solved in the field of online education.

[0003] As a core component of student learning resources, exercises play an irreplaceable role in consolidating knowledge and testing learning effects. However, most online test question websites currently provide relatively single exercise resources, lacking sufficient diversity and pertinence, and are unable to meet the growing personalized learning needs of students. Summary of the invention

[0004] The main purpose of the embodiments of the present invention is to propose a method and device for recommending exercises based on course knowledge graphs and knowledge tracking, which can deeply explore students' learning behaviors and preferences, and can provide accurate and diverse exercise recommendations. At the same time, it makes full use of the rich relationship information and semantic content contained in the subject knowledge graph, and attaches clear and unambiguous explanations to each recommended exercise, thereby helping the online education field to achieve a more efficient and personalized learning experience.

[0005] To achieve the above purpose, one aspect of an embodiment of the present invention proposes a method for recommending exercises based on a course knowledge graph and knowledge tracking, comprising the following steps:

[0006] According to the knowledge characteristics of the target subject, build a subject knowledge map based on the obtained teaching reference materials;

[0007] According to the students' answer history sequence and current knowledge status, a knowledge tracker based on the DKT model is constructed;

[0008] Generate exercise recommendation results based on the subject knowledge graph and the knowledge tracker.

[0009] In some embodiments, the step of constructing a subject knowledge graph based on the acquired teaching reference materials according to the knowledge characteristics of the target subject includes the following steps:

[0010] Carry out entity definition processing, specifically: according to the knowledge characteristics of the target subject, entities are divided into two types, the first type is concepts and definitions, and the second type is theorems and methods;

[0011] Define relationships, specifically: according to the knowledge characteristics of the target subject, define nine relationships between knowledge points, and divide the nine relationships into three categories, including superior knowledge points, same-level knowledge points, and subordinate knowledge points; wherein the superior knowledge points include dependency relationships, inclusion relationships, and ownership relationships; the same-level knowledge points include synonymy relationships, antonym relationships, and apposition relationships; the subordinate knowledge points include dependent relationships, inclusion relationships, and ownership relationships;

[0012] Collect textbooks and supporting teaching reference materials of the target subject as the core data source, and supplement and improve the core data source by crawling relevant data from the network platform to build a corpus;

[0013] Combining the BiLSTM model with the CRF model, pre-training the word vectors is performed based on the corpus to extract entity knowledge points;

[0014] By combining the BiLSTM model with the attention mechanism, we extract relationships between entity knowledge points and obtain a subject knowledge graph represented in the form of triples.

[0015] In some embodiments, the step of constructing a knowledge tracker based on the DKT model according to the student's answer history sequence and current knowledge status includes the following steps:

[0016] By analyzing the students' historical answer sequences, we can model the students' knowledge mastery at any given point in time.

[0017] Based on the student's current knowledge status, predict the student's accuracy in answering questions in the future, thereby providing a scientific basis for the student's learning path planning;

[0018] Specifically, given a student's historical learning interaction sequence X = (x1, x2..., x t ), the DKT model is used to predict the student’s next interaction x t+1 The performance of t It is usually expressed as (q t , α t ), indicating that the student answered question q at time t t , the answer is α t ; The formula of the DKT model is as follows:

[0019] h t =tanh(W hx x t +W hh h t-1 +b h ),

[0020] yt =σ(W yh h t +b y ),

[0021] Among them, h t represents the student’s knowledge status at time t, y t It records the probability of the student answering each question correctly at time t; Whx represents the input weight matrix; W hh represents the cyclic weight matrix; h t-1 represents the student’s learning status at time t-1; b h represents the bias of the potential unit; σ represents the sigmoid function; W yh represents the readout weight matrix; b y Represents the deviation of the readout unit.

[0022] In some embodiments, generating exercise recommendation results according to the subject knowledge graph and the knowledge tracker includes the following steps:

[0023] According to the current exercises encountered by students and the knowledge points corresponding to the exercises, find the related exercises as the candidate exercise set E; for each exercise e in E i , all predict the probability p of the student answering this exercise correctly based on the knowledge tracker i ;

[0024] Set an expected difficulty δ and calculate the distance between the probability of answering each exercise correctly and the expected difficulty. The calculation formula for this process is:

[0025]

[0026] in, Indicates the difficulty of the question;

[0027] According to the distance value DISTANCE(e i ,δ) Sort the exercises from small to large, and take the top-k exercises as the recommended exercise list;

[0028] Based on the relationship between knowledge points in the knowledge graph and the exercise recommendation list, an exercise recommendation result is finally generated.

[0029] In some embodiments, the step of finally generating an exercise recommendation result based on the relationship between knowledge points in the knowledge graph and the exercise recommendation list includes the following steps:

[0030] For each knowledge point K, the knowledge points connected to it are divided into three categories, corresponding to the three types of relationships in the knowledge graph, including upper-level knowledge points, same-level knowledge points, and lower-level knowledge points;

[0031] When recommending exercises, we set two thresholds β and χ, where 0<β<χ<1. We use α to represent the degree of students’ mastery of a certain knowledge point. The specific recommendations are as follows:

[0032] If the student's mastery of knowledge point k is 0<α≤β, then recommend the superior knowledge point of knowledge point k to the student; if there is no direct superior knowledge point, then recommend the superior knowledge point of the same level knowledge point of knowledge point k; if there is no direct superior knowledge point, then recommend knowledge point k and its same level knowledge points;

[0033] If the student's mastery of knowledge point k is β<α≤χ, then the knowledge point k itself and the knowledge points of the same level are added to the recommended knowledge point list;

[0034] If the student's cognitive level of knowledge point k is χ<α≤1, the subordinate knowledge points of knowledge point k will be recommended to the student; if there are no subordinate knowledge points, the subordinate knowledge points of the same level as knowledge point k will continue to be recommended; if there are no subordinate knowledge points, the knowledge point k and its same level knowledge points will be recommended.

[0035] Another aspect of the embodiment of the present invention further provides an exercise recommendation device based on course knowledge graph and knowledge tracking, comprising:

[0036] The first module is used to construct a subject knowledge map based on the knowledge characteristics of the target subject and the acquired teaching reference materials;

[0037] The second module is used to build a knowledge tracker based on the DKT model according to the students' answer history sequence and current knowledge status;

[0038] The third module is used to generate exercise recommendation results based on the subject knowledge graph and the knowledge tracker.

[0039] To achieve the above objective, another aspect of an embodiment of the present invention provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above-mentioned method when executing the computer program.

[0040] To achieve the above objective, another aspect of an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned method is implemented.

[0041] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the above method.

[0042] The embodiments of the present invention include at least the following beneficial effects: The present invention provides a method and device for recommending exercises based on course knowledge graph and knowledge tracking. The scheme constructs a subject knowledge graph based on the knowledge characteristics of the target subject and the acquired teaching reference materials; constructs a knowledge tracker based on the DKT model according to the student's answer history sequence and current knowledge status; generates exercise recommendation results based on the subject knowledge graph and the knowledge tracker. The embodiments of the present invention can deeply explore students' learning behaviors and preferences, and can provide both accurate and diverse exercise recommendations. At the same time, it makes full use of the rich relationship information and semantic content contained in the subject knowledge graph, and attaches clear and explicit explanations to each recommended exercise, thereby helping the online education field to achieve a more efficient and personalized learning experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of an implementation environment provided by an embodiment of the present invention;

[0044] Figure 2 is a flow chart of overall steps provided by an embodiment of the present invention;

[0045] Figure 3 It is a flow chart of a specific implementation process provided by an embodiment of the present invention;

[0046] Figure 4 It is a flow chart for constructing a subject knowledge graph provided by an embodiment of the present invention;

[0047] Figure 5 is a flowchart of the steps of recommending exercises provided by an embodiment of the present invention;

[0048] Figure 6 is another flow chart of a specific implementation process provided by an embodiment of the present invention;

[0049] Figure 7 It is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present invention. They are only examples of devices and methods consistent with some aspects of the embodiments of the present invention as detailed in the attached claims.

[0051] It is understood that the terms "first", "second", etc. used in the present invention may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present invention, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0052] The terms "at least one", "multiple", "each", "any", etc. used in the present invention, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein are only for the purpose of describing the embodiments of the present invention and are not intended to limit the present invention.

[0054] The method and device for recommending exercises based on course knowledge graph and knowledge tracking provided by the embodiment of the present invention relate to the field of computer technology. The method for recommending exercises based on course knowledge graph and knowledge tracking provided by the embodiment of the present invention can be applied to a terminal, can be applied to a server, or can be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application that implements the method for recommending exercises based on course knowledge graph and knowledge tracking, etc., but is not limited to the above forms.

[0055] The present invention can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present invention can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present invention can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0056] like Figure 1 FIG. 1 is a schematic diagram of an implementation environment provided by an embodiment of the present invention. Figure 1 The implementation environment includes at least one terminal 102 and a server 101. The terminal 102 and the server 101 can be connected to a network wirelessly or wired to complete data transmission and exchange.

[0057] Server 101 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers. It can also be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0058] In addition, the server 101 can also be a node server in the blockchain network. Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, and encryption algorithm.

[0059] The terminal 102 may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc. The terminal 102 may also be a vehicle-mounted terminal of various device types described above, but is not limited thereto. The terminal 102 and the server 101 may be directly or indirectly connected via wired or wireless communication, which is not limited in this embodiment of the present invention.

[0060] Based on the example Figure 1In the implementation environment shown, an embodiment of the present invention provides an exercise recommendation method based on a course knowledge graph and knowledge tracking. The following is explained using the example of the exercise recommendation method based on a course knowledge graph and knowledge tracking being applied to a server 101. It can be understood that the method can also be applied to a terminal 102.

[0061] Reference Figure 2 , Figure 2 This is a flowchart of a method for recommending exercises based on course knowledge graph and knowledge tracking applied to a server according to an embodiment of the present invention. The execution subject of this method can be any of the aforementioned computer devices (including servers or terminals). Figure 2 , the method may include the following steps:

[0062] According to the knowledge characteristics of the target subject, build a subject knowledge map based on the obtained teaching reference materials;

[0063] According to the students' answer history sequence and current knowledge status, a knowledge tracker based on the DKT model is constructed;

[0064] Generate exercise recommendation results based on the subject knowledge graph and the knowledge tracker.

[0065] In some embodiments, the step of constructing a subject knowledge graph based on the acquired teaching reference materials according to the knowledge characteristics of the target subject includes the following steps:

[0066] Carry out entity definition processing, specifically: according to the knowledge characteristics of the target subject, entities are divided into two types, the first type is concepts and definitions, and the second type is theorems and methods;

[0067] Define relationships, specifically: according to the knowledge characteristics of the target subject, define nine relationships between knowledge points, and divide the nine relationships into three categories, including superior knowledge points, same-level knowledge points, and subordinate knowledge points; wherein the superior knowledge points include dependency relationships, inclusion relationships, and ownership relationships; the same-level knowledge points include synonymy relationships, antonym relationships, and apposition relationships; the subordinate knowledge points include dependent relationships, inclusion relationships, and ownership relationships;

[0068] Collect textbooks and supporting teaching reference materials of the target subject as the core data source, and supplement and improve the core data source by crawling relevant data from the network platform to build a corpus;

[0069] Combining the BiLSTM model with the CRF model, pre-training the word vectors is performed based on the corpus to extract entity knowledge points;

[0070] By combining the BiLSTM model with the attention mechanism, we extract relationships between entity knowledge points and obtain a subject knowledge graph represented in the form of triples.

[0071] In some embodiments, the step of constructing a knowledge tracker based on the DKT model according to the student's answer history sequence and current knowledge status includes the following steps:

[0072] By analyzing the students' historical answer sequences, we can model the students' knowledge mastery at any given point in time.

[0073] Based on the student's current knowledge status, predict the student's accuracy in answering questions in the future, thereby providing a scientific basis for the student's learning path planning;

[0074] Specifically, given a student's historical learning interaction sequence X = (x1, x2..., x t ), the DKT model is used to predict the student’s next interaction x t+1 The performance of t It is usually expressed as (q t , α t ), indicating that the student answered question q at time t t , the answer is α t ; The formula of the DKT model is as follows:

[0075] h t =tanh(W hx x t +W hh h t-1 +b h ),

[0076] y t =σ(W yh h t +b y ),

[0077] Among them, h t represents the student’s knowledge status at time t, y t W records the probability that the student answers each question correctly at time t; hx Represents the input weight matrix; W hh represents the cyclic weight matrix; h t-1 represents the student’s learning status at time t-1; b h represents the bias of the potential unit; σ represents the sigmoid function; W yh represents the readout weight matrix; b y Represents the deviation of the readout unit.

[0078] In some embodiments, generating exercise recommendation results according to the subject knowledge graph and the knowledge tracker includes the following steps:

[0079] According to the current exercises encountered by students and the knowledge points corresponding to the exercises, find the related exercises as the candidate exercise set E; for each exercise e in E i , all predict the probability p of the student answering this exercise correctly based on the knowledge tracker i ;

[0080] Set an expected difficulty δ and calculate the distance between the probability of answering each exercise correctly and the expected difficulty. The calculation formula for this process is:

[0081]

[0082] in, Indicates the difficulty of the question;

[0083] According to the distance value DISTANCE(e i ,δ) Sort the exercises from small to large, and take the top-k exercises as the recommended exercise list:

[0084] Based on the relationship between knowledge points in the knowledge graph and the exercise recommendation list, an exercise recommendation result is finally generated.

[0085] In some embodiments, the step of finally generating an exercise recommendation result based on the relationship between knowledge points in the knowledge graph and the exercise recommendation list includes the following steps:

[0086] For each knowledge point K, the knowledge points connected to it are divided into three categories, corresponding to the three types of relationships in the knowledge graph, including upper-level knowledge points, same-level knowledge points, and lower-level knowledge points;

[0087] When recommending exercises, we set two thresholds β and χ, where 0<β<χ<1; α is used to represent the degree of students’ mastery of a certain knowledge point. The specific recommendations are as follows:

[0088] If the student's mastery of knowledge point k is 0<α≤β, then recommend the superior knowledge point of knowledge point k to the student; if there is no direct superior knowledge point, then recommend the superior knowledge point of the same level knowledge point of knowledge point k; if there is no direct superior knowledge point, then recommend knowledge point k and its same level knowledge points;

[0089] If the student's mastery of knowledge point k is β<α≤χ, then the knowledge point k itself and the knowledge points of the same level are added to the recommended knowledge point list;

[0090] If the student's cognitive level of knowledge point k is χ<α≤1, the subordinate knowledge points of knowledge point k will be recommended to the student; if there are no subordinate knowledge points, the subordinate knowledge points of the same level as knowledge point k will continue to be recommended; if there are no subordinate knowledge points, the knowledge point k and its same level knowledge points will be recommended.

[0091] The following describes the specific implementation process of the method of the present invention in detail by taking a specific application scenario as an example:

[0092] In view of the problems existing in the prior art, the present invention proposes an explainable exercise recommendation strategy based on subject knowledge graph and knowledge tracking technology. This strategy deeply explores students' learning behaviors and preferences, aiming to provide both accurate and diverse exercise recommendations. At the same time, it makes full use of the rich relationship information and semantic content contained in the subject knowledge graph, and provides clear and explicit explanations for each recommended exercise, thereby helping the online education field to achieve a more efficient and personalized learning experience.

[0093] The present invention mainly targets the problem of personalized exercise recommendation in online education, which can be summarized into three parts: subject knowledge graph construction, knowledge tracker and exercise recommender. First, by constructing a subject knowledge graph, the system can display the knowledge points and their relationships within the subject in a structured manner, providing rich semantic information and theoretical basis for subsequent recommendation tasks. Next, the knowledge tracker (based on the DKT model) is used to analyze the students' answer history, accurately predict the students' knowledge mastery status and the accuracy of the exercises, and provide a basis for personalized recommendations. Finally, the exercise recommender intelligently generates a list of Top-k recommended exercises based on the prediction results, and uses the relationship between knowledge points in the knowledge graph to provide a reasonable explanation for the recommendation results.

[0094] Compared with the prior art, the present invention not only improves the accuracy and efficiency of recommendations, but also enhances the interpretability of recommendations, providing strong support for smart education and online education platforms.

[0095] refer to Figure 3 and Figure 6 , these three parts are described below:

[0096] 1. Construction of subject knowledge graph

[0097] refer to Figure 4 , constructing a subject knowledge graph is the key foundation for implementing explainable exercise recommendations in the embodiment of the present invention. Here, the embodiment of the present invention takes junior high school mathematics as an example as the subject exploration background of the embodiment of the present invention, and uses the knowledge graph to deeply characterize the subject knowledge.

[0098] 1. Entity definition

[0099] According to the knowledge characteristics of junior high school mathematics, the embodiment of the present invention divides entities into two types, namely, concepts and definitions (such as functions, quadratic equations, etc.) and theorems and methods (such as matching method, tangent length theorem, etc.).

[0100] 2. Relationship Definition

[0101] Similarly, the embodiment of the present invention defines nine relationships between knowledge points based on the knowledge characteristics of junior high school mathematics, and divides them into three categories.

[0102] 1) Advanced knowledge points: contain (e.g., a triangle contains a right triangle), depend (e.g., a trapezoid depends on a parallelogram), possess (e.g., a right triangle has a right angle);

[0103] 2) Knowledge points of the same level: synonyms (e.g., factorization and factorization are synonyms), antonyms (e.g., acute angles and obtuse angles are opposites), and co-positions (e.g., an isosceles right triangle and a right triangle are co-positions);

[0104] 3) Subordinate knowledge points: being contained (e.g. an isosceles triangle is contained in a triangle), being dependent (e.g. equality is dependent on an isosceles triangle), being owned (e.g. the axis of symmetry is owned by an equilateral triangle).

[0105] 3. Data Acquisition

[0106] The first step of the embodiment of the present invention is to widely collect junior high school mathematics textbooks and their supporting teaching reference materials, such as teaching plans, as core data sources, and then systematically summarize and carefully classify the knowledge points in these materials. At the same time, the embodiment of the present invention also makes necessary supplements and improvements to the existing knowledge system by crawling rich data from online platforms such as Baidu Encyclopedia to ensure its comprehensiveness and accuracy.

[0107] 4. Entity extraction and relationship extraction

[0108] Next, the embodiment of the present invention uses the word2vec algorithm to pre-train word vectors on a carefully constructed corpus, which lays a solid foundation for subsequent entity and relationship extraction. In terms of entity extraction, the embodiment of the present invention cleverly combines the BiLSTM model and the CRF model to achieve accurate recognition and extraction of knowledge points. In the relationship extraction link, the embodiment of the present invention adopts a powerful combination of the BiLSTM model and the attention mechanism, so that the complex and subtle associations between knowledge points can be efficiently excavated. Through this series of precise processing procedures, the embodiment of the present invention finally obtains a knowledge graph represented in the form of triples, which clearly shows the internal logic and structural relationship between each knowledge point in the junior high school mathematics subject.

[0109] 2. Knowledge Tracker Based on DKT Model

[0110] The core of the knowledge tracking task is to deeply analyze and model students' learning data in order to track the dynamic changes in students' knowledge level. Its core goal is to build a model that can accurately depict the evolution of students' knowledge status over time, and then predict their future learning outcomes and needs, so as to provide students with tailored learning support and suggestions. Specifically, the knowledge tracking task includes two aspects: first, by analyzing the historical sequence of students' answers, accurately model the students' knowledge mastery at any given point in time; second, based on the students' current knowledge status, predict the accuracy of the students in future answers, and provide a scientific basis for the planning of their learning path.

[0111] Specifically, given a student’s historical learning interaction sequence X = (x1, x2…, x t ), the DKT model is used to predict the student’s next interaction x t+1 Here, x t It is usually expressed as (q t , α t ), indicating that the student answered question q at time t t , the answer is α t (0 represents an incorrect answer, 1 represents a correct answer). The formula of the DKT model can be expressed as follows:

[0112] h t =tanh(W hx x t +W hh h t-1 +b h ),

[0113] y t =σ(W yh h t +b y ),

[0114] Among them, h t represents the student’s knowledge status at time t, y t What is recorded is the probability that the student answers each question correctly at time t.

[0115] (III) Exercise Recommender

[0116] refer to Figure 5 Through the knowledge tracking model DKT, the knowledge state of each student at a certain moment can be obtained. The embodiment of the present invention can use this knowledge state to evaluate the student's mastery of each knowledge point (expressed in the range of [0, 1]). First, according to the current exercises encountered by the student and the knowledge points corresponding to the exercises, find the related exercises as the candidate exercise set E. For each exercise e in E i, the probability p of the student answering this question correctly can be predicted based on the knowledge tracker i Then, in order to ensure that the difficulty of the recommended exercises is appropriate for students, an expected difficulty δ is set, and the distance between the probability of answering each exercise correctly and this expected difficulty is calculated:

[0117]

[0118] in, Indicates the difficulty of this question. According to the distance value DISTANCE(e i ,δ) Sort the exercises from small to large, and take the top-k exercises as the recommendation list. Finally, the embodiment of the present invention generates recommendations and gives reasonable explanations based on the relationship between knowledge points in the knowledge graph and the exercise recommendation list. Specifically, for each knowledge point K, the knowledge points connected to it are divided into three categories, corresponding to the three types of relationships in the knowledge graph, namely, superior knowledge points (inclusion, dependence, possession), same-level knowledge points (synonymous, antonymous, homonymous), and subordinate knowledge points (included, dependent, possessed). When making exercise recommendations, two thresholds β and χ are set, where 0<β<χ<1. α is used to represent the student's mastery of a certain knowledge point.

[0119] (1) If the student's mastery of knowledge point k is 0<α≤β, it means that the student's mastery of this knowledge point is relatively low, which may be caused by the failure to master the superior knowledge point. Therefore, the superior knowledge point of knowledge point k is recommended to the student. If there is no direct superior knowledge point, the superior knowledge point of the same level knowledge point of knowledge point k is recommended. If there is no direct superior knowledge point, the superior knowledge point of the same level knowledge point of knowledge point k is recommended. If there is no superior knowledge point, the knowledge point k and its same level knowledge points are recommended.

[0120] (2) If the student's mastery of knowledge point k is β<α≤χ, it means that the student has a moderate mastery of this knowledge point and needs to do some more related exercises to further consolidate it. In this case, knowledge point k itself and knowledge points of the same level are added to the recommended knowledge point list;

[0121] (3) If the student's cognitive level of knowledge point k is χ<α≤1, it means that the student has mastered this knowledge point to a high degree and can proceed to learn subsequent knowledge points, so the subordinate knowledge points of knowledge point k are recommended to the student. If there are no subordinate knowledge points, the subordinate knowledge points of the same level of knowledge point k are recommended. If there are no subordinate knowledge points, the knowledge point k and its same level knowledge points are recommended.

[0122] In summary, the present invention discloses an interpretable exercise recommendation strategy based on subject knowledge graph and knowledge tracing. The specific process is summarized as follows: First, the embodiment of the present invention selects a certain subject field as the research background, and constructs a subject knowledge graph through a systematic knowledge point induction and information extraction process. This graph not only contains rich semantic information and the relationship between knowledge points, but also lays a solid foundation for subsequent knowledge tracing and exercise recommendation tasks, so that the embodiment of the present invention can more deeply mine and analyze the knowledge system. Then, the embodiment of the present invention adopts the DKT (Deep Knowledge Tracing) model to accurately predict the correct rate of exercises and the knowledge mastery status of students at a specific time point from the exercise text and the student's answer history record. This step provides a key basis for the subsequent personalized exercise recommendation. Finally, under the dual support of subject knowledge graph and knowledge tracing technology, the embodiment of the present invention designs a set of interpretable exercise recommendation strategies. This strategy intelligently generates a Top-k recommended exercise list based on the students' mastery of each knowledge point and their expected exercise difficulty predicted by the DKT model. At the same time, with the help of the organizational relationship of knowledge points in the knowledge graph, the embodiment of the present invention can provide a reasonable and easy-to-understand explanation for each recommendation result, thereby significantly enhancing the interpretability and practicality of the model while improving the accuracy of the recommendation system. In summary, as an interpretable exercise recommendation strategy that integrates subject knowledge graphs and knowledge tracking technology, the present invention has shown broad application prospects and value in multiple fields such as smart education and online education platforms.

[0123] Another aspect of the embodiment of the present invention further provides an exercise recommendation device based on course knowledge graph and knowledge tracking, comprising:

[0124] The first module is used to construct a subject knowledge map based on the knowledge characteristics of the target subject and the acquired teaching reference materials;

[0125] The second module is used to build a knowledge tracker based on the DKT model according to the students' answer history sequence and current knowledge status;

[0126] The third module is used to generate exercise recommendation results based on the subject knowledge graph and the knowledge tracker.

[0127] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0128] The embodiment of the present invention further provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned exercise recommendation method based on course knowledge graph and knowledge tracking when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0129] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0130] See also Figure 7 , Figure 7 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0131] The processor 701 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present invention.

[0132] The memory 702 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 702 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 702, and the processor 701 calls and executes the exercise recommendation method based on the course knowledge graph and knowledge tracking of the embodiment of the present invention;

[0133] Input / output interface 703, used to implement information input and output;

[0134] Communication interface 704, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0135] A bus 705 that transmits information between the various components of the device (e.g., the processor 701, the memory 702, the input / output interface 703, and the communication interface 704);

[0136] The processor 701 , the memory 702 , the input / output interface 703 and the communication interface 704 are connected to each other in communication within the device via a bus 705 .

[0137] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned exercise recommendation method based on course knowledge graph and knowledge tracking.

[0138] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0139] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0140] It should be noted that in various specific embodiments of the present invention, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present invention needs to obtain the user's sensitive personal information, it will obtain the user's separate permission or consent through a pop-up window or jump to a confirmation page, and after clearly obtaining the user's separate permission or consent, it will obtain the necessary user-related data for the normal operation of the embodiment of the present invention.

[0141] The embodiments described in the embodiments of the present invention are intended to more clearly illustrate the technical solutions of the embodiments of the present invention, and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. Those skilled in the art can appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0142] Those skilled in the art will appreciate that the technical solutions shown in the figures do not limit the embodiments of the present invention and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0143] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0144] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0145] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0146] It should be understood that in the present invention, "at least one (item)" refers to one or more, and "plurality" refers to two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can represent: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0147] In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0148] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0149] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0150] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including multiple instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store programs.

[0151] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the embodiments of the present invention is not limited thereby. Any modification, equivalent substitution and improvement made by those skilled in the art without departing from the scope and essence of the embodiments of the present invention shall be within the scope of the rights of the embodiments of the present invention.

Claims

1. A method for recommending exercises based on course knowledge graph and knowledge tracking, characterized in that: The following steps are involved: According to the knowledge characteristics of the target subject, build a subject knowledge map based on the obtained teaching reference materials; According to the students' answer history sequence and current knowledge status, a knowledge tracker based on the DKT model is constructed; Generate exercise recommendation results based on the subject knowledge graph and the knowledge tracker.

2. According to claim 1, a method for recommending exercises based on course knowledge graph and knowledge tracking is characterized in that: According to the knowledge characteristics of the target subject, the subject knowledge graph is constructed based on the acquired teaching reference materials, including the following steps: Carry out entity definition processing, specifically: according to the knowledge characteristics of the target subject, entities are divided into two types, the first type is concepts and definitions, and the second type is theorems and methods; Define relationships, specifically: according to the knowledge characteristics of the target subject, define nine relationships between knowledge points, and divide the nine relationships into three categories, including superior knowledge points, same-level knowledge points, and subordinate knowledge points; wherein the superior knowledge points include dependency relationships, inclusion relationships, and ownership relationships; the same-level knowledge points include synonymy relationships, antonym relationships, and apposition relationships; the subordinate knowledge points include dependent relationships, inclusion relationships, and ownership relationships; Collect textbooks and supporting teaching reference materials of the target subject as the core data source, and supplement and improve the core data source by crawling relevant data from the network platform to build a corpus; Combining the BiLSTM model with the CRF model, pre-training of word vectors is performed based on the corpus to extract entity knowledge points: By combining the BiLSTM model with the attention mechanism, we extract relationships between entity knowledge points and obtain a subject knowledge graph represented in the form of triples.

3. The method for recommending exercises based on course knowledge graph and knowledge tracking according to claim 1, characterized in that: The method of constructing a knowledge tracker based on the DKT model according to the student's answer history sequence and current knowledge status includes the following steps: By analyzing the students' historical answer sequences, we can model the students' knowledge mastery at any given point in time. Based on the student's current knowledge status, predict the student's accuracy in answering questions in the future, thereby providing a scientific basis for the student's learning path planning; Specifically: Given a student's historical learning interaction sequence X = (x1, x2, ..., x t ), the DKT model is used to predict the student’s next interaction x t+1 The performance of t It is usually expressed as (q t , α t ), indicating that the student answered question q at time t t , the answer is α t ; The formula of the DKT model is as follows: h t =tanh(W hx x t +W hh h t-1 +b h ), y t =σ(W yh h t +b y ), Among them, h t represents the student’s knowledge status at time t, y t W records the probability that the student answers each question correctly at time t; hx Represents the input weight matrix; W hh represents the cyclic weight matrix; h t-1 represents the student’s learning status at time t-1; b h represents the bias of the potential unit; σ represents the sigmoid function; W yh represents the readout weight matrix; b y Represents the deviation of the readout unit.

4. The method for recommending exercises based on course knowledge graph and knowledge tracking according to claim 1, characterized in that: Generating exercise recommendation results according to the subject knowledge graph and the knowledge tracker includes the following steps: According to the current exercises encountered by students and the knowledge points corresponding to the exercises, find the related exercises as the candidate exercise set E; for each exercise e in E i , all predict the probability p of the student answering this exercise correctly based on the knowledge tracker i ; Set an expected difficulty δ and calculate the distance between the probability of answering each exercise correctly and the expected difficulty. The calculation formula for this process is: in, Indicates the difficulty of the question; According to the distance value DISTANCE(e i ,δ) Sort the exercises from small to large, and take the top-k exercises as the recommended exercise list; Based on the relationship between knowledge points in the knowledge graph and the exercise recommendation list, an exercise recommendation result is finally generated.

5. The method for recommending exercises based on course knowledge graph and knowledge tracking according to claim 4, characterized in that: The method of finally generating an exercise recommendation result based on the relationship between knowledge points in the knowledge graph and the exercise recommendation list includes the following steps: For each knowledge point K, the knowledge points connected to it are divided into three categories, corresponding to the three types of relationships in the knowledge graph, including upper-level knowledge points, same-level knowledge points, and lower-level knowledge points; When recommending exercises, we set two thresholds β and χ, where 0<β<χ<1; α is used to represent the degree of students’ mastery of a certain knowledge point. The specific recommendations are as follows: If the student's mastery of knowledge point k is 0<α≤β, then recommend the superior knowledge point of knowledge point k to the student; if there is no direct superior knowledge point, then recommend the superior knowledge point of the same level knowledge point of knowledge point k; if there is no direct superior knowledge point, then recommend knowledge point k and its same level knowledge points; If the student's mastery of knowledge point k is β<α≤χ, then the knowledge point k itself and the knowledge points of the same level are added to the recommended knowledge point list; If the student's cognitive level of knowledge point k is χ<α≤1, the subordinate knowledge points of knowledge point k will be recommended to the student; if there are no subordinate knowledge points, the subordinate knowledge points of the same level as knowledge point k will continue to be recommended; if there are no subordinate knowledge points, the knowledge point k and its knowledge points of the same level will be recommended.

6. A device for recommending exercises based on course knowledge graph and knowledge tracking, characterized in that: include: The first module is used to construct a subject knowledge map based on the knowledge characteristics of the target subject and the acquired teaching reference materials; The second module is used to build a knowledge tracker based on the DKT model according to the students' answer history sequence and current knowledge status; The third module is used to generate exercise recommendation results based on the subject knowledge graph and the knowledge tracker.

7. An electronic device, characterized in that: including a processor and a memory; The memory is used to store programs; The processor executes the program to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The storage medium stores a program, and the program is executed by a processor to implement the method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images