A big data learning analysis method for panoramic scanning of learning situation

Abstract

The present invention discloses a big data learning analysis method for panoramic scanning of learning situation, including: collecting students' learning situation panoramic data and preprocessing them to obtain preprocessed learning situation panoramic data; vectorizing the preprocessed learning situation panoramic data to obtain a learning situation panoramic vector sequence; constructing a knowledge state evaluation model, inputting the learning situation panoramic vector sequence into the model, and outputting the students' knowledge level state; constructing a personalized learning resource recommendation model, and matching learning resources based on the similarity matching analysis between the students' knowledge level state and different learning resource data. The present invention constructs a learning situation panoramic vector sequence that characterizes the changes in students' knowledge points during the learning process, evaluates the students' current knowledge level state, and realizes personalized and accurate recommendation of learning resources based on the overlap between the weaknesses of the knowledge level state and the knowledge points contained in the learning resources.

Description

Technical Field

[0001] The present invention relates to the technical field of big data learning analysis, and in particular to a big data learning analysis method for panoramic scanning of learning situations. Background Art

[0002] In recent years, with the rapid development of information technology, the pace of digital transformation in the field of education has continued to accelerate, and the process of educational informatization has also been steadily advancing. Educational informatization has not only promoted the sharing of educational resources and the improvement of educational quality, but also doubled the amount of data in the educational process.

[0003] At present, with the large-scale popularization of digital tools such as online learning platforms, smart campuses, and learning management systems (LMS), teachers, students, parents, and education managers can easily and quickly obtain a large amount of educational data. These data cover multiple dimensions such as students' learning behavior, academic performance, learning paths, and participation, providing rich materials for educational research and teaching decisions. But on the other hand, the large amount and complex structure of these data make traditional educational research methods seem powerless when facing big data.

[0004] Korean patent KR20230060869A discloses a learning management system and a method for providing learning information therefor, the method comprising the following steps: storing student data, course data and experience data in a database, the student data including age, learning habits, understanding level, weaknesses and parent communication, the course data including subjects, courses, progress, course content, homework and achievement rate, the experience data including student tendencies, coach tendencies, events, problems, solutions and the impact of solutions; using a natural language processing model to pre-process student data, classroom data and experience data, including structured or unstructured text; generating a learning guidance model by associating the pre-processed student data with classroom data, and forming a personality counseling model by associating the pre-processed student data with experience data; inputting student data, outputting personalized learning information based on the learning guidance model, and outputting personalized personality information based on the personality counseling model. Although this patent also uses a big data model, the data involved is too complicated and there are too many interference items. It can only plan the future learning and growth direction of students in a general direction. It lacks a clear solution to check for current learning weaknesses, and it is not associated with real teaching resources, so it is not very practical.

[0005] Therefore, how to use big data technology to conduct in-depth analysis of these data, explore the potential patterns and information therein, and provide precise support for teaching has become an important issue that needs to be urgently addressed in this industry. Summary of the invention

[0006] In view of this, the present invention proposes a big data learning analysis method for panoramic scanning of learning situation. By scanning and analyzing the students' learning situation data throughout the whole process, hidden rules and information are excavated, which helps to deepen the understanding of students' learning behavior and learning process, and provide a theoretical basis for personalized learning and teaching.

[0007] In order to achieve the above-mentioned purpose, the present invention provides a big data learning analysis method for panoramic scanning of learning situation, comprising the following steps:

[0008] S1. Collect and preprocess the students' learning data to obtain the preprocessed learning data, where the learning data includes the students' test score data, course learning data, and exercise answer data;

[0009] S2, vectorizing the preprocessed learning situation panoramic data to obtain a learning situation panoramic vector sequence;

[0010] S3, construct a knowledge status evaluation model, input the learning situation panorama vector sequence into the model, and output the student's knowledge level status;

[0011] S4. Build a personalized learning resource recommendation model to match learning resources based on the similarity matching analysis between students’ knowledge level and different learning resource data, including:

[0012] S421. The students' knowledge level And multiple groups of learning resource data are input into the personalized learning resource recommendation model, where the learning resource data is:

[0013] ;

[0014] in:

[0015] Indicates Group learning resource data, Indicates the total number of learning resource data;

[0016] S422. Knowledge level status Perform similarity matching calculations on heterogeneous data with different groups of learning resource data:

[0017] Extract the knowledge micro-skill sequence in the learning resource data, and convert the knowledge micro-skill sequence into a matrix form to form a knowledge matrix of the learning resource data, wherein the knowledge matrix is ​​in the form of G rows and H columns, G represents the number of knowledge skills, and H represents the number of knowledge micro-skills in each knowledge skill. The element value of the g-th row and h-th column in the knowledge matrix represents whether the learning resource data explains the h-th knowledge micro-skill in the g-th knowledge skill. If yes, the element value is 0, otherwise, the element value is 1;

[0018] Calculate knowledge level status and learning resource data The similarity between:

[0019] ;

[0020] in:

[0021] Indicates the knowledge level and learning resource data The similarity between Represents learning resource data Knowledge matrix;

[0022] Representing the knowledge matrix The degree matrix of Indicates the knowledge level The degree matrix of

[0023] represents the L1 norm;

[0024] S423, Selection and Knowledge Level Status The learning resource data with the highest similarity is used as the recommendation result to recommend learning resources to students.

[0025] Preferably, the step S1 comprises:

[0026] S11. Collect students’ full-view data on their academic performance, including their test scores, course learning data, and exercise answer data:

[0027] ;

[0028] ;

[0029] ;

[0030] ;

[0031] in:

[0032] It represents the collected panoramic data of students’ learning situation. They represent the students’ test score data, course learning data, and exercise answer data respectively;

[0033] Indicates student correspondence The test score data, represents the total number of exams taken by the student, Represents the student's score data for the nth test, , performance data Include exam title , total score and test scores ;

[0034] Indicates student correspondence Learning data for each course, represents the total number of courses the student has taken, Indicates that students Learning data for each course, , learning data Include course title 、Course knowledge points And the duration of study ;

[0035] Indicates student correspondence The answer data of the exercises, represents the total number of exercises completed by the students. Indicates that students The answer data of the exercises, , answer data Including exercise knowledge points , Difficulty of Exercises And the answer ;

[0036] S12. Test score data , course learning data And exercise answer data Preprocess them respectively to obtain the preprocessed learning situation panoramic data x.

[0037] Preferably, the step S12 comprises:

[0038] S121. Test score data Preprocessing:

[0039] For the exam name Perform word segmentation processing to obtain the word segmentation phrase sequence of the exam name:

[0040] ;

[0041] in:

[0042] Indicates the name of the exam of participle phrases, Indicates the name of the exam The total number of participle phrases;

[0043] Calculate the student's score rate for each test:

[0044] ;

[0045] in:

[0046] represents the score rate of students in the nth test;

[0047] S122. Course learning data Preprocessing:

[0048] Course Name Perform word segmentation processing to obtain the word segmentation phrase sequence of the course name:

[0049] ;

[0050] in:

[0051] Indicates the course name of participle phrases, Indicates the course name The total number of participle phrases;

[0052] Course knowledge points Perform word segmentation processing to obtain the word segmentation phrase sequence of the course knowledge points:

[0053] ;

[0054] in:

[0055] Indicates course knowledge points of participle phrases, Indicates course knowledge points The total number of participle phrases;

[0056] Study time To standardize:

[0057] ;

[0058] in:

[0059] Indicates The duration of the course, Indicates the duration of study The standardized processing results;

[0060] S123. Data on answering exercises Preprocessing:

[0061] Knowledge points for exercises Perform word segmentation processing to obtain the word segmentation phrase sequence of the exercise knowledge point:

[0062] ;

[0063] in:

[0064] Indicates knowledge points of exercises of participle phrases, Indicates knowledge points of exercises The total number of participle phrases;

[0065] Difficulty of exercises Perform normalization processing to obtain normalized results , the maximum value of the difficulty of the exercise is 1 and the minimum value is 0;

[0066] Encode the answer and get the encoding result , 1 means the student answered correctly, 0 means the student answered incorrectly;

[0067] S124, constructing the pre-processed learning situation panoramic data x:

[0068] ;

[0069] ;

[0070] ;

[0071] ;

[0072] ;

[0073] ;

[0074] ;

[0075] in:

[0076] Represents test score data , course learning data and exercise answer data The preprocessing results;

[0077] Respectively represent the performance data , learning data and answer data The preprocessing results.

[0078] Preferably, the step S2 comprises:

[0079] S21, performing one-hot encoding on all the segmented word groups in the preprocessed learning situation panoramic data x to obtain the word vectors corresponding to the segmented word groups;

[0080] S22. Generate test score vector :

[0081] ;

[0082] in:

[0083] represents an exponential function with a natural constant as the base, * represents the convolution operator, Represents the test score convolution matrix;

[0084] Indicates the test score data preprocessing results All word segmentation phrases in are replaced by word vector data sequences;

[0085] S23. Generate course learning vector :

[0086] ;

[0087] in:

[0088] represents the course learning convolution matrix, Represents the learning data embedding matrix;

[0089] Represents the results of course learning data preprocessing All word segments in are replaced by word vector data sequences, Represents a data sequence The individual data;

[0090] S24. Obtaining students’ answers time of each exercise, generating the exercise answer vector :

[0091] ;

[0092] ;

[0093] ;

[0094] ;

[0095] in:

[0096] Represents the convolution matrix of the exercise answer;

[0097] Represents a sequence of word segments The corresponding word vector sequence;

[0098] Indicates that the student answered Exercise time, Indicates the time when the exercise is forgotten, represents the parameter for regulating the forgetting time of an exercise;

[0099] Indicates the results of preprocessing the exercise answer data The vectorized representation result of ;

[0100] Indicates the result of answer data preprocessing Vectorized representation of ;

[0101] represents the L2 norm;

[0102] S25. Constructing a panoramic vector sequence of learning situation .

[0103] Preferably, step S3 comprises:

[0104] S31. Constructing a knowledge status evaluation model:

[0105] The knowledge state evaluation model includes an input layer, a knowledge level state initialization generation layer, an attention information extraction layer, a knowledge level state enhancement layer and an output layer;

[0106] The input layer is used to receive a panoramic learning situation vector sequence;

[0107] The knowledge level state initialization generation layer is used to convert the learning situation panorama vector sequence into the student's knowledge level initial state, and the knowledge level initial state is in the form of a matrix with G rows and H columns, G represents the number of knowledge skills, and H represents the number of knowledge micro-skills in each knowledge skill. The element value of the g-th row and h-th column in the knowledge level initial state matrix represents the student's mastery of the h-th knowledge micro-skill in the g-th knowledge skill;

[0108] The attention information extraction layer is an attention module structure in the Transformer model, which is used to obtain the attention information of the learning situation panoramic vector sequence;

[0109] The knowledge level state enhancement layer is used to enhance the initial state of the knowledge level based on the attention information to obtain the student's knowledge level state;

[0110] The output layer is used to output the knowledge level status obtained by evaluation;

[0111] S32. Use the knowledge status assessment model to assess the student's knowledge level.

[0112] Preferably, the step S32 includes:

[0113] S321, input layer receives the panoramic vector sequence of learning situation ;

[0114] S322, the knowledge level state initialization generation layer converts the learning situation panoramic vector sequence Convert to the student's initial state of knowledge :

[0115] ;

[0116] in:

[0117] represents the knowledge and skills transposed matrix, represents transpose, represents the activation function;

[0118] S323, the attention information extraction layer obtains the learning situation panoramic vector sequence Attention information:

[0119] ;

[0120] ;

[0121] in:

[0122] Representing the panoramic vector sequence of learning situation attention information;

[0123] represents the attention convolution matrix, Representation vector attention information;

[0124] S324, the knowledge level state enhancement layer performs information enhancement on the initial state of the knowledge level based on the attention information:

[0125] ;

[0126] in:

[0127] Indicates the student's knowledge level. Indicates the initial state of knowledge level The degree matrix of

[0128] S325, the output layer outputs the knowledge level status obtained by evaluation .

[0129] Preferably, the step S4 comprises:

[0130] S41. Build a personalized learning resource recommendation model:

[0131] The personalized learning resource recommendation model includes an input layer, a similarity matching layer and an output layer;

[0132] The input layer is used to receive knowledge level status and learning resource data;

[0133] The similarity matching layer is used to perform similarity matching calculations on the knowledge level status and different learning resource data under heterogeneous data, and select the learning resource data with the highest similarity to the knowledge level status as the recommendation result;

[0134] The output layer is used to output the recommendation results;

[0135] S42. Use the personalized learning resource recommendation model to match learning resources.

[0136] The present invention further provides an electronic device, comprising:

[0137] A memory storing at least one instruction;

[0138] Communication interface, enabling electronic equipment to communicate; and

[0139] The processor executes the instructions stored in the memory to implement the above-mentioned big data learning and analysis method for panoramic scanning of the learning situation.

[0140] The present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned big data learning and analysis method for panoramic scanning of the learning situation.

[0141] Compared with the prior art, the technical solution of the present invention has at least the following beneficial effects:

[0142] The present invention obtains students' test score data, course learning data and exercise answer data, and adopts different preprocessing methods to perform data preprocessing, so as to realize vectorized representation of students' mastery of knowledge points in the learning process and extract data features, further construct a learning situation panoramic vector sequence representing the changes of students' knowledge points in the learning process, and use the knowledge state evaluation model to convert the learning situation panoramic vector sequence into the initial state of knowledge level to obtain the students' initial mastery of different knowledge skills and knowledge micro-skills, and perform attention enhancement processing on the initial state of knowledge level according to the attention difference, improve the key weight of key learning situation in the initial state of knowledge level, and obtain the knowledge level state representing the panoramic changes of students' learning; in addition, extract knowledge micro-skill sequences from different learning resource data, convert the knowledge micro-skill sequences into matrix form, and constitute a knowledge matrix of learning resource data; match the knowledge matrix with the students' current knowledge level state, the higher the similarity between the two, the higher the overlap between the weaknesses of the knowledge level state and the knowledge points covered by the learning resources, which is conducive to realizing personalized and accurate recommendation of learning resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0143] Figure 1 This is a flow chart of a big data learning and analysis method for panoramic scanning of learning conditions in Example 1 of the present invention;

[0144] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0145] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0146] The embodiment of the present application provides a big data learning and analysis method for panoramic scanning of the learning situation. The execution subject of the big data learning and analysis method for panoramic scanning of the learning situation includes but is not limited to at least one of the electronic devices such as the server and the terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the big data learning and analysis method for panoramic scanning of the learning situation can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0147] Embodiment 1:

[0148] S1. Collect and preprocess the students' panoramic learning data to obtain the preprocessed panoramic learning data:

[0149] S11. Collect students’ full-view data on their academic performance, including their test scores, course learning data, and exercise answer data:

[0150] ;

[0151] ;

[0152] ;

[0153] ;

[0154] in:

[0155] It represents the collected panoramic data of students’ learning situation. They represent the students’ test score data, course learning data, and exercise answer data respectively;

[0156] Indicates student correspondence The test score data, represents the total number of exams taken by the student, Represents the student's score data for the nth test, , performance data Include exam title , total score and test scores ;

[0157] Indicates student correspondence Learning data for each course, represents the total number of courses the student has taken, Indicates that students Learning data for each course, , learning data Include course title 、Course knowledge points And the duration of study ;

[0158] Indicates student correspondence The answer data of the exercises, represents the total number of exercises completed by the students. Indicates that students The answer data of the exercises, , answer data Including exercise knowledge points , Difficulty of Exercises And the answer ;

[0159] S12. Test score data , course learning data And exercise answer data Preprocess them respectively to obtain the preprocessed learning situation panoramic data x.

[0160] S121. Test score data Preprocessing:

[0161] For the exam name Perform word segmentation processing to obtain the word segmentation phrase sequence of the exam name:

[0162] ;

[0163] in:

[0164] Indicates the name of the exam of participle phrases, Indicates the name of the exam The total number of segmentation phrases; in this embodiment, the segmentation tools used are all jieba segmentation tools;

[0165] Calculate the student's score rate for each test:

[0166] ;

[0167] in:

[0168] represents the score rate of students in the nth test;

[0169] S122. Course learning data Preprocessing:

[0170] Course Name Perform word segmentation processing to obtain the word segmentation phrase sequence of the course name:

[0171] ;

[0172] in:

[0173] Indicates the course name of participle phrases, Indicates the course name The total number of participle phrases;

[0174] Course knowledge points Perform word segmentation processing to obtain the word segmentation phrase sequence of the course knowledge points:

[0175] ;

[0176] in:

[0177] Indicates course knowledge points of participle phrases, Indicates course knowledge points The total number of participle phrases;

[0178] Study time To standardize:

[0179] ;

[0180] in:

[0181] Indicates The duration of the course, Indicates the duration of study The standardized processing results;

[0182] S123. Data on answering exercises Preprocessing:

[0183] Knowledge points for exercises Perform word segmentation processing to obtain the word segmentation phrase sequence of the exercise knowledge point:

[0184] ;

[0185] in:

[0186] Indicates knowledge points of exercises of participle phrases, Indicates knowledge points of exercises The total number of participle phrases;

[0187] Difficulty of exercises Perform normalization processing to obtain normalized results , the maximum value of the difficulty of the exercise is 1 and the minimum value is 0;

[0188] Encode the answer and get the encoding result , 1 means the student answered correctly, 0 means the student answered incorrectly;

[0189] S124, constructing the pre-processed learning situation panoramic data x:

[0190] ;

[0191] ;

[0192] ;

[0193] ;

[0194] ;

[0195] ;

[0196] ;

[0197] in:

[0198] Represents test score data , course learning data and exercise answer data The preprocessing results;

[0199] Respectively represent the performance data , learning data and answer data The preprocessing results.

[0200] S2. Vectorize the preprocessed learning situation panoramic data to obtain a learning situation panoramic vector sequence:

[0201] S21, performing one-hot encoding on all the segmented word groups in the preprocessed learning situation panoramic data x to obtain the word vectors corresponding to the segmented word groups;

[0202] S22. Generate test score vector :

[0203] ;

[0204] in:

[0205] represents an exponential function with a natural constant as the base, * represents the convolution operator, Represents the test score convolution matrix;

[0206] Represents the test score data preprocessing results All word segmentation phrases in are replaced by word vector data sequences;

[0207] S23. Generate course learning vector :

[0208] ;

[0209] in:

[0210] represents the course learning convolution matrix, Represents the learning data embedding matrix;

[0211] Represents the results of course learning data preprocessing All word segments in are replaced by word vector data sequences, Represents a data sequence The individual data;

[0212] S24. Obtaining students’ answers time of each exercise, generating the exercise answer vector :

[0213] ;

[0214] ;

[0215] ;

[0216] ;

[0217] in:

[0218] Represents the convolution matrix of the exercise answer;

[0219] Represents a sequence of word segments The corresponding word vector sequence;

[0220] Indicates that the student answered Exercise time, Indicates the time when the exercise is forgotten, represents the parameter for regulating the forgetting time of an exercise;

[0221] Indicates the results of preprocessing the exercise answer data The vectorized representation result of ;

[0222] Indicates the result of answer data preprocessing Vectorized representation of ;

[0223] represents the L2 norm;

[0224] S25. Constructing a panoramic vector sequence of learning situation .

[0225] S3. Construct a knowledge status evaluation model, input the learning situation panorama vector sequence into the model, and output the student’s knowledge level status:

[0226] S31. Constructing a knowledge status evaluation model:

[0227] The knowledge state evaluation model includes an input layer, a knowledge level state initialization generation layer, an attention information extraction layer, a knowledge level state enhancement layer and an output layer; the knowledge state evaluation model takes the learning situation panoramic vector sequence as input and the evaluated knowledge level state as output, wherein the graph enhanced neural network based on the attention mechanism is the implementation method of the knowledge state evaluation model;

[0228] The input layer is used to receive a panoramic learning situation vector sequence;

[0229] The knowledge level state initialization generation layer is used to convert the learning situation panorama vector sequence into the student's knowledge level initial state, and the knowledge level initial state is in the form of a matrix with G rows and H columns, G represents the number of knowledge skills, and H represents the number of knowledge micro-skills in each knowledge skill. The element value of the g-th row and h-th column in the knowledge level initial state matrix represents the student's mastery of the h-th knowledge micro-skill in the g-th knowledge skill;

[0230] The attention information extraction layer is an attention module structure in the Transformer model, which is used to obtain the attention information of the learning situation panoramic vector sequence;

[0231] The knowledge level state enhancement layer is used to enhance the initial state of the knowledge level based on the attention information to obtain the student's knowledge level state;

[0232] The output layer is used to output the knowledge level status obtained by evaluation;

[0233] S32. Use the knowledge status evaluation model to evaluate the student's knowledge level:

[0234] S321, input layer receives the panoramic vector sequence of learning situation ;

[0235] S322, the knowledge level state initialization generation layer converts the learning situation panoramic vector sequence Convert to the student's initial state of knowledge :

[0236] ;

[0237] in:

[0238] represents the knowledge and skills transposed matrix, represents transpose, represents an activation function. In this embodiment, the activation function used is the ReLU function;

[0239] S323, the attention information extraction layer obtains the learning situation panoramic vector sequence Attention information:

[0240] ;

[0241] ;

[0242] in:

[0243] Representing the panoramic vector sequence of learning situation attention information;

[0244] represents the attention convolution matrix, Representation vector attention information;

[0245] S324, the knowledge level state enhancement layer performs information enhancement on the initial state of the knowledge level based on the attention information:

[0246] ;

[0247] in:

[0248] Indicates the student's knowledge level. Indicates the initial state of knowledge level The degree matrix of

[0249] S325, the output layer outputs the knowledge level status obtained by evaluation .

[0250] S4. Build a personalized learning resource recommendation model to match learning resources based on the similarity matching analysis between students’ knowledge level and different learning resource data:

[0251] S41. Build a personalized learning resource recommendation model:

[0252] The personalized learning resource recommendation model includes an input layer, a similarity matching layer and an output layer; the personalized learning resource recommendation model takes the evaluated knowledge level status and learning resource data as input, takes the most suitable matching learning resource data as output, and adopts similarity matching analysis to recommend learning resources suitable for students' knowledge level status;

[0253] The input layer is used to receive knowledge level status and learning resource data;

[0254] The similarity matching layer is used to perform similarity matching calculations on the knowledge level status and different learning resource data under heterogeneous data, and select the learning resource data with the highest similarity to the knowledge level status as the recommendation result;

[0255] The output layer is used to output the recommendation results;

[0256] S42. Use the personalized learning resource recommendation model to match learning resources.

[0257] S421. The students' knowledge level And multiple groups of learning resource data are input into the personalized learning resource recommendation model, where the learning resource data is:

[0258] ;

[0259] in:

[0260] Indicates Group learning resource data, Indicates the total number of learning resource data;

[0261] S422. Knowledge level status Perform similarity matching calculations on heterogeneous data with different groups of learning resource data:

[0262] Extract the knowledge micro-skill sequence in the learning resource data, and convert the knowledge micro-skill sequence into a matrix form to form a knowledge matrix of the learning resource data, wherein the knowledge matrix is ​​in the form of G rows and H columns, G represents the number of knowledge skills, and H represents the number of knowledge micro-skills in each knowledge skill. The element value of the g-th row and h-th column in the knowledge matrix represents whether the learning resource data explains the h-th knowledge micro-skill in the g-th knowledge skill. If yes, the element value is 0, otherwise, the element value is 1;

[0263] Calculate knowledge level status and learning resource data The similarity between:

[0264] ;

[0265] in:

[0266] Indicates the knowledge level and learning resource data The higher the similarity, the higher the overlap between the weak knowledge and skills and knowledge micro-skills of the knowledge level and the knowledge points explained by the learning resource data, and the more targeted the learning resources are; Represents learning resource data Knowledge matrix;

[0267] Representing the knowledge matrix The degree matrix of Indicates the knowledge level The degree matrix of

[0268] represents the L1 norm;

[0269] S423, Selection and Knowledge Level Status The learning resource data with the highest similarity is used as the recommendation result to recommend learning resources to students.

[0270] Embodiment 2:

[0271] This embodiment provides an electronic device that can implement a big data learning analysis method for panoramic scanning of the learning situation.

[0272] The electronic device may include a memory, a processor, a communication interface and a bus, and may also include a computer program stored in the memory and executable on the processor, such as a program.

[0273] Wherein, the memory includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (for example: SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory may be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory may also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the memory may also include both an internal storage unit of the electronic device and an external storage device. The memory can be used not only to store application software and various types of data installed in the electronic device, such as program code, etc., but also to temporarily store data that has been output or is to be output.

[0274] In some embodiments, the processor may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and a combination of various control chips. The processor is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect the various components of the entire electronic device. It runs or executes programs or modules stored in the memory (for implementing the big data learning analysis program for panoramic scanning of learning conditions in Example 1, etc.), and calls the data stored in the memory to perform various functions of the electronic device and process data.

[0275] The communication interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device and other electronic devices, and to achieve connection and communication between internal components of the electronic device.

[0276] The bus may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory and at least one processor, etc.

[0277] For example, although not shown, the electronic device may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, etc. The electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0278] Preferably, the electronic device may further include a user interface, which may be a display, an input unit (such as a keyboard), and preferably, the user interface may also be a standard wired interface or a wireless interface. Preferably, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device, etc. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device and to display a visual user interface.

[0279] It should be understood that this embodiment is for illustration only and the scope of the patent application is not limited by this structure.

[0280] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0281] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0282] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A big data learning analysis method for panoramic scanning of learning situation, characterized in that: The following steps are involved: S1. Collect and preprocess the students' learning data to obtain the preprocessed learning data, where the learning data includes the students' test score data, course learning data, and exercise answer data; S2, vectorizing the preprocessed learning situation panoramic data to obtain a learning situation panoramic vector sequence; S3, construct a knowledge status evaluation model, input the learning situation panorama vector sequence into the model, and output the student's knowledge level status; S4. Build a personalized learning resource recommendation model to match learning resources based on the similarity matching analysis between students’ knowledge level and different learning resource data, including: S421. The students' knowledge level And multiple groups of learning resource data are input into the personalized learning resource recommendation model, where the learning resource data is: ； in: Indicates Group learning resource data, Indicates the total number of learning resource data; S422. Knowledge level status Perform similarity matching calculations on heterogeneous data with different groups of learning resource data: Extract the knowledge micro-skill sequence in the learning resource data, and convert the knowledge micro-skill sequence into a matrix form to form a knowledge matrix of the learning resource data, wherein the knowledge matrix is ​​in the form of G rows and H columns, G represents the number of knowledge skills, and H represents the number of knowledge micro-skills in each knowledge skill. The element value of the g-th row and h-th column in the knowledge matrix represents whether the learning resource data explains the h-th knowledge micro-skill in the g-th knowledge skill. If yes, the element value is 0, otherwise, the element value is 1; Calculate knowledge level status and learning resource data The similarity between: ； in: Indicates the knowledge level and learning resource data The similarity between Represents learning resource data Knowledge matrix; Representing the knowledge matrix The degree matrix of Indicates the knowledge level The degree matrix of represents the L1 norm; S423, Selection and Knowledge Level Status The learning resource data with the highest similarity is used as the recommendation result to recommend learning resources to students.

2. The big data learning and analysis method for panoramic scanning of learning situation according to claim 1 is characterized in that: The step S1 comprises: S11. Collect students’ full-view data on their academic performance, including their test scores, course learning data, and exercise answer data: ； ； ； ； in: It represents the collected panoramic data of students’ learning situation. They represent the students’ test score data, course learning data, and exercise answer data respectively; Indicates student correspondence The test score data, represents the total number of exams taken by the student. Represents the student's score data for the nth test, , performance data Include exam title , total score and test scores ; Indicates student correspondence Learning data for each course, represents the total number of courses the student has taken, Indicates that students Learning data for each course, , learning data Include course title 、Course knowledge points And the duration of study ; Indicates student correspondence The answer data of the exercises, represents the total number of exercises completed by the students. Indicates that students The answer data of the exercises, , answer data Including exercise knowledge points , Difficulty of Exercises And the answer ; S12. Test score data , course learning data And exercise answer data Preprocess them respectively to obtain the preprocessed learning situation panoramic data x.

3. The big data learning and analysis method for panoramic scanning of learning situation according to claim 2 is characterized in that: The step S12 comprises: S121. Test score data Preprocessing: For the exam name Perform word segmentation processing to obtain the word segmentation phrase sequence of the exam name: ； in: Indicates the name of the exam of participle phrases, Indicates the name of the exam The total number of participle phrases; Calculate the student's score rate for each test: ； in: represents the score rate of students in the nth test; S122. Course learning data Preprocessing: Course Name Perform word segmentation processing to obtain the word segmentation phrase sequence of the course name: ； in: Indicates the course name of participle phrases, Indicates the course name The total number of participle phrases; Course knowledge points Perform word segmentation processing to obtain the word segmentation phrase sequence of the course knowledge points: ； in: Indicates course knowledge points of participle phrases, Indicates course knowledge points The total number of participle phrases; Study time To standardize: ； in: Indicates The duration of the course, Indicates the duration of study The standardized processing results; S123. Data on answering exercises Preprocessing: Knowledge points for exercises Perform word segmentation processing to obtain the word segmentation phrase sequence of the exercise knowledge point: ； in: Indicates knowledge points of exercises of participle phrases, Indicates knowledge points of exercises The total number of participle phrases; Difficulty of exercises Perform normalization processing to obtain normalized results , the maximum value of the difficulty of the exercise is 1 and the minimum value is 0; Encode the answer and get the encoding result , 1 means the student answered correctly, 0 means the student answered incorrectly; S124, constructing the pre-processed learning situation panoramic data x: ； ； ； ； ； ； ； in: Represents test score data , course learning data and exercise answer data The preprocessing results; Respectively represent the performance data , learning data and answer data The preprocessing results.

4. The big data learning and analysis method for panoramic scanning of learning situation according to claim 3 is characterized in that: The step S2 comprises: S21, performing one-hot encoding on all the segmented word groups in the preprocessed learning situation panoramic data x to obtain the word vectors corresponding to the segmented word groups; S22. Generate test score vector : ； in: represents an exponential function with a natural constant as the base, * represents the convolution operator, Represents the test score convolution matrix; Represents the test score data preprocessing results All word segmentation phrases in are replaced by word vector data sequences; S23. Generate course learning vector : ； in: represents the course learning convolution matrix, Represents the learning data embedding matrix; Represents the results of course learning data preprocessing All word segments in are replaced by word vector data sequences, Represents a data sequence The individual data; S24. Obtaining students’ answers time of each exercise, generating the exercise answer vector : ； ； ； ； in: Represents the convolution matrix of the exercise answer; Represents a sequence of word segments The corresponding word vector sequence; Indicates that the student answered Time for exercises, Indicates the time when the exercise is forgotten, represents the parameter for regulating the forgetting time of an exercise; Indicates the results of preprocessing the exercise answer data The vectorized representation result of ; Indicates the result of answer data preprocessing Vectorized representation of ; represents the L2 norm; S25. Constructing a panoramic vector sequence of learning situation .

5. The big data learning and analysis method for panoramic scanning of learning situation according to claim 1 is characterized in that: The step S3 comprises: S31. Constructing a knowledge status evaluation model: The knowledge state evaluation model includes an input layer, a knowledge level state initialization generation layer, an attention information extraction layer, a knowledge level state enhancement layer and an output layer; The input layer is used to receive a panoramic learning situation vector sequence; The knowledge level state initialization generation layer is used to convert the learning situation panorama vector sequence into the student's knowledge level initial state, and the knowledge level initial state is in the form of a matrix with G rows and H columns, G represents the number of knowledge skills, and H represents the number of knowledge micro-skills in each knowledge skill. The element value of the g-th row and h-th column in the knowledge level initial state matrix represents the student's mastery of the h-th knowledge micro-skill in the g-th knowledge skill; The attention information extraction layer is an attention module structure in the Transformer model, which is used to obtain the attention information of the learning situation panoramic vector sequence; The knowledge level state enhancement layer is used to enhance the initial state of the knowledge level based on the attention information to obtain the student's knowledge level state; The output layer is used to output the knowledge level status obtained by evaluation; S32. Use the knowledge status assessment model to assess the student's knowledge level.

6. The big data learning and analysis method for panoramic scanning of learning situation according to claim 5 is characterized in that: The step S32 comprises: S321, input layer receives the panoramic vector sequence of learning situation ; S322, the knowledge level state initialization generation layer converts the learning situation panoramic vector sequence Convert to the student's initial state of knowledge : ； in: represents the knowledge and skills transposed matrix, represents transpose, represents the activation function; S323, the attention information extraction layer obtains the learning situation panoramic vector sequence Attention information: ； ； in: Representing the panoramic vector sequence of learning situation attention information; represents the attention convolution matrix, Representation vector attention information; S324, the knowledge level state enhancement layer performs information enhancement on the initial state of the knowledge level based on the attention information: ； in: Indicates the student's knowledge level. Indicates the initial state of knowledge level The degree matrix of S325, the output layer outputs the knowledge level status obtained by evaluation .

7. The big data learning and analysis method for panoramic scanning of learning situation according to claim 1 is characterized in that: The step S4 comprises: S41. Build a personalized learning resource recommendation model: The personalized learning resource recommendation model includes an input layer, a similarity matching layer and an output layer; The input layer is used to receive knowledge level status and learning resource data; The similarity matching layer is used to perform similarity matching calculations on the knowledge level status and different learning resource data under heterogeneous data, and select the learning resource data with the highest similarity to the knowledge level status as the recommendation result; The output layer is used to output the recommendation results; S42. Use the personalized learning resource recommendation model to match learning resources.

Images