Artificial intelligence education knowledge tracking system based on big data

Abstract

The invention belongs to the technical field of intelligent education, and particularly relates to an artificial intelligence education knowledge tracking system based on big data, which constructs a data set by collecting basic information of students, classifies the data set and stores the classified data set in a distributed file system, and converts the sub-data set into feature representation suitable for machine learning to obtain feature data; a deep learning framework is adopted to train a machine learning model, and a neural network model is constructed after the model is optimized; inputting the neural network model into a student management platform, generating a personalized learning path in combination with the data set, tracking a learning result, and displaying the learning result through a user interface; according to the method, big data analysis of learning behaviors of students is realized, personalized learning contents and paths are recommended for the students through a machine learning algorithm, learning conditions of the students can be autonomously tracked, statistics of the learning conditions can be carried out, and learning information can be timely and accurately acquired.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent education technology, and specifically relates to an artificial intelligence education knowledge tracking system based on big data. Background Art

[0002] Project background: With the rapid development of science and technology, artificial intelligence has become one of the hottest fields in today's society. Especially in the field of education, the application of artificial intelligence can provide students with a more personalized, intelligent and efficient learning experience.

[0003] As disclosed in the publication (announcement) number: CN111241165B, an artificial intelligence education system and data processing method based on big data is proposed, which uses big data analysis technology to conduct targeted data analysis and processing on online and offline education application scenarios as well as student objects and teacher objects, and characterizes the education needs of different application scenarios and objects from multiple dimensions based on the results of the data analysis and processing, and finally formulates a corresponding education scenario adjustment mode to adaptively transform the current specific education scenario, thereby improving the accuracy and efficiency of the operation of the education system.

[0004] However, there are still certain defects in the operation of artificial intelligence-based education systems, such as: it is unable to independently track students' learning situation and statistics on their learning situation, which makes it impossible to obtain learning information in a timely and accurate manner. Summary of the invention

[0005] The purpose of the present invention is to provide an artificial intelligence education knowledge tracking system based on big data, to realize big data analysis of students' learning behavior, and to recommend personalized learning content and paths to students through machine learning algorithms to solve the problems in the prior art raised in the above background technology.

[0006] To achieve the above purpose, the present invention adopts the following technical solution: an artificial intelligence education knowledge tracking system based on big data, comprising: The data collection module is used to collect students' basic information data, learning behavior data, and other factors that may affect their learning behavior, and use the big data processing framework to store and process the collected data to construct a data set; A data input module is used to receive the data set, classify the data set, store it in a distributed file system, and obtain a sub-data set; A feature extraction module, used to extract a sub-dataset in the distributed file system and convert the sub-dataset into a feature representation suitable for machine learning to obtain feature data; A model training module is used to train a machine learning model using a deep learning framework according to the feature data, and to build a neural network model after tuning the model; The model prediction module is used to input the neural network model into the student management platform, generate a personalized learning path in combination with the data set, track the learning outcomes, and display the learning outcomes through the user interface.

[0007] Preferably, the data acquisition module includes: A data collection unit, used to automatically or manually collect students' basic information and learning behavior data; A big data analysis unit, which is used to clean, transform, summarize and store the collected data in a database; The data processing and analysis unit is used to use machine learning and data mining techniques to screen out students' learning behaviors and factors that may affect their learning behaviors from the database.

[0008] Preferably, the basic information and learning behavior data are obtained through questionnaires, online form filling, test score records, student management platforms, and education platforms.

[0009] Preferably, the data input module comprises: The data reading unit is used to read data sets from input sources using Python; The data classification unit is used to classify the read data set using a naive Bayes classifier. After the classification is completed, the classification result is written into the distributed file system; The sub-dataset acquisition unit is used to retrieve the corresponding sub-dataset from the distributed file system according to the classification result and request.

[0010] Preferably, the data input module further includes: A sub-dataset loading unit, used to write code in Python to load the sub-dataset into memory; The API interface unit is used to provide an API interface to connect to the feature extraction module to obtain a sub-dataset.

[0011] Preferably, the feature extraction module comprises: A data extraction unit, used to extract the required sub-dataset from the distributed file system through an API interface; Data preprocessing unit, used to remove terminal duplicates of sub-datasets and fill missing values, eliminate potential outliers and unbalanced data distribution; The feature representation unit is used to train the preprocessed data using a machine learning algorithm to obtain feature representation data.

[0012] Preferably, the model training module includes: Feature processing unit, used to pre-process the original feature representation data, including data cleaning, normalization, and standardization operations; The model building unit is used to use TensorFlow and its provided APIs and tools to build neural network models.

[0013] Preferably, the model training module further includes: Model compilation unit, used to compile the neural network model after it is built; The training unit is used to train the compiled neural network model. After the model training is completed, it is evaluated and further optimized if the model performs poorly.

[0014] Preferably, the model prediction module comprises: A deployment unit, used to convert the trained neural network model into a format that can be deployed on the student management platform; The unit is used to load the API and input it into the model of the student management platform. After the model is loaded, the student's personal information and course information are used to generate a personalized learning path; A prediction unit that uses a random forest algorithm to predict a student's performance on a specific course, which will then be used to adjust the learning path; Preferably, the model prediction module further includes: a display unit, which is used to display learning outcomes using visualization tools in the student management platform.

[0015] Technical effects and advantages of the present invention: Compared with the prior art, the artificial intelligence education knowledge tracking system based on big data proposed by the present invention has the following advantages: The present invention constructs a data set by collecting basic information of students, classifies the data set and stores it in a distributed file system, converts sub-data sets into feature representations suitable for machine learning to obtain feature data; uses a deep learning framework to train the machine learning model, and constructs a neural network model after tuning the model; inputs the neural network model into a student management platform, generates a personalized learning path in combination with the data set, tracks the learning outcomes, and displays the learning outcomes through a user interface; implements big data analysis of student learning behaviors, recommends personalized learning content and paths to students through machine learning algorithms, can autonomously track students' learning situations and statistics on the learning situations, thereby enabling timely and accurate acquisition of learning information. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a module diagram of an artificial intelligence education knowledge tracking system based on big data of the present invention; Figure 2This is a flow chart of an artificial intelligence education knowledge tracking method based on big data of the present invention; Figure 3 This is the topological diagram of the deep learning framework of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0018] The present invention provides an artificial intelligence education knowledge tracking system based on big data. Figure 1 As shown, it includes: a data acquisition module, a data input module, a feature extraction module, a model training module and a model prediction module.

[0019] Exemplarily, the data collection module is used to collect students' basic information data, learning behavior data, and other factors that may affect learning behavior, and use a big data processing framework to store and process the collected data to construct a data set; Specifically, the data acquisition module includes: a data collection unit, which is used to automatically or manually collect students' basic information and learning behavior data; a big data analysis unit, which is used to clean, convert, summarize and store the collected data in a database; a data processing and analysis unit, which is used to use machine learning and data mining techniques to filter out students' learning behaviors and factors that may affect learning behaviors from the database.

[0020] Furthermore, basic information and learning behavior data are obtained through questionnaires, online form filling, test score records, student management platforms, and education platforms.

[0021] For the above functions, first of all, a data collection system needs to be designed to automatically or manually collect students' basic information and learning behavior data. These data can be obtained through questionnaires, online form filling, test score records, etc. In order to ensure the accuracy and completeness of the data, a special data review team can be set up to carefully review and verify the collected data.

[0022] Next, we need to build an efficient big data analysis platform, which can use distributed computing frameworks such as Hadoop and Spark to store and process student data. Through these frameworks, the collected data can be cleaned, converted, and summarized to extract valuable information and store it in the database for subsequent data analysis and application.

[0023] In the data processing and analysis stage, it is necessary to use technologies such as machine learning and data mining to filter out useful patterns and rules from massive amounts of data, and further understand students' learning behaviors and factors that may affect their learning behaviors. In this way, more accurate learning paths and methods can be obtained, providing students with more personalized learning services.

[0024] Finally, these data results need to be organized into interactive data visualization tools, such as charts, heat maps, maps, etc., so that teachers, parents, students, etc. can intuitively understand and grasp the learning situation and behavior changes. At the same time, data interfaces can also be opened through APIs, etc., so that third-party developers can develop new applications and services based on the data set.

[0025] In summary, through the above four steps, we can effectively collect, manage and process students’ basic information, learning behavior data and data on factors that may affect their learning behavior, and use these data to construct useful data sets to further study and optimize educational and teaching work.

[0026] Exemplarily, the data input module is used to receive the data set, and store the data set in a distributed file system after classification, and obtain a sub-data set; Specifically, the data input module includes: The data reading unit is used to read data sets from input sources using Python; The data classification unit is used to classify the read data set using a naive Bayes classifier. After the classification is completed, the classification result is written into the distributed file system; The sub-dataset acquisition unit is used to retrieve the corresponding sub-dataset from the distributed file system according to the classification results and requests. This function can be implemented using a big data processing framework such as Apache Spark, Apache Hadoop MapReduce, or Google Cloud Dataflow.

[0027] The sub-dataset loading unit is used to write code using Python to load the sub-dataset into the memory; use Python or other programming languages ​​to write code to load the sub-dataset into the memory. Libraries such as NumPy and Pandas can be used to simplify this process.

[0028] The API interface unit is used to provide an API interface to connect to the feature extraction module to obtain sub-datasets. To facilitate the use of front-end applications, APIs are provided to obtain sub-datasets. These APIs should allow users to specify the required classification parameters (such as time range, geographic location, etc.) and filtering conditions.

[0029] With this solution, large data sets can be managed efficiently, and sub-data sets can be quickly accessed and queried when needed. This will help improve the efficiency and accuracy of data analysis.

[0030] Exemplarily, the feature extraction module is used to extract a sub-dataset in the distributed file system, and convert the sub-dataset into a feature representation suitable for machine learning to obtain feature data; Specifically, the feature extraction module includes: a data extraction unit, which is used to extract the required sub-dataset from the distributed file system through an API interface; a data preprocessing unit, which is used to remove terminal duplicates of the sub-dataset and fill in missing values, eliminate potential outliers and unbalanced data distribution; a feature representation unit, which is used to train the preprocessed data using a machine learning algorithm to obtain feature representation data.

[0031] In this embodiment, first, the required sub-dataset needs to be extracted from the distributed file system. This can be achieved by writing SQL queries or performing operations in the file system API. For example, if the distributed file system is HDFS, you can use the Apache Hadoop command line tool hdfs dfs -get -f <path-to-directory>to extract the directory contents.

[0032] After successfully extracting the sub-dataset, it needs to be preprocessed to facilitate the subsequent machine learning model training. The preprocessing process includes steps such as removing duplicates, filling missing values, standardization, and binning. These steps will help eliminate potential outliers and unbalanced data distribution.

[0033] Next, the preprocessed data will be trained using various machine learning algorithms to obtain feature representations. Here, Python and its widely available libraries such as Pandas, NumPy, Scikit-learn, and TensorFlow are used. First, a base model (such as linear regression or support vector machine) is selected and its parameters are adjusted to optimize its performance. Then, different model types will be tried and their performance will be compared. Finally, the best performing model is selected and used as the feature representation.

[0034] Once feature representations have been obtained, they can be used for tasks such as classification, clustering, dimensionality reduction, and visualization. They can also be combined with traditional statistical methods to improve the performance of the model.

[0035] Exemplarily, the model training module is used to train the machine learning model using a deep learning framework according to the feature data, and to construct a neural network model after tuning the model; Specifically, the model training module includes: a feature processing unit, which is used to pre-process the original feature representation data, including data cleaning, normalization, and standardization operations; a model building unit, which is used to use TensorFlow and its provided APIs and tools to build a neural network model. A model compilation unit is used to compile the neural network model after it is built; a training unit is used to train the compiled neural network model, and after the model training is completed, it is evaluated and further optimized if the model performs poorly.

[0036] In this embodiment, data preprocessing: First, the original feature data needs to be preprocessed, including data cleaning, normalization, standardization and other operations to ensure the quality of the data and the stability of the model.

[0037] Model building: Next, choose a suitable deep learning framework (such as TensorFlow, PyTorch, etc.) and use the API and tools it provides to build a neural network model. Common model structures include convolutional neural network (CNN), recurrent neural network (RNN), autoencoder (AE), etc.

[0038] Model compilation: After the model is built, it needs to be compiled so that the computer can run the model. This usually involves specifying parameters such as loss function, optimizer, and evaluation metrics.

[0039] Model training: Now we can start training the model. During this process, we will continuously adjust some hyperparameters such as learning rate, batch size, regularization parameters, etc. according to the performance of the model in order to obtain better performance.

[0040] Model evaluation and optimization: After the model training is completed, the model needs to be evaluated to see how it performs on unsupervised data. If the model performs poorly, it needs to be further optimized, which may require adjusting the model structure, modifying some hyperparameters, or changing the training strategy.

[0041] Model deployment: When the model performs well enough, it can be deployed to real-world applications such as online products, mobile applications, or embedded devices.

[0042] Continuous optimization: Finally, it is necessary to regularly monitor the performance of the model and perform fine-tuning and model updates based on actual conditions to ensure that the model always remains in a good working condition.

[0043] Through the above steps, we can successfully use the deep learning framework to train the machine learning model based on the feature data, and build a neural network model after tuning the model. This will help us better understand and predict complex phenomena and make more informed decisions.

[0044] Exemplarily, the model prediction module is used to input the neural network model into the student management platform, generate a personalized learning path in combination with the data set and track the learning outcomes, and display the learning outcomes through the user interface.

[0045] Specifically, the model prediction module includes: a deployment unit, which is used to convert the trained neural network model into a format that can be deployed on the student management platform; an embedding unit, which is used to load the API and input it into the model of the student management platform. After the model is loaded, the student's personal information and course information are used to generate a personalized learning path; a prediction unit, which is used to use a random forest algorithm to predict the student's performance in a specific course, and then these prediction results will be used to adjust the learning path; a display unit, which is used to use visualization tools in the student management platform to display the learning outcomes.

[0046] Going further, first, the trained neural network model needs to be converted into a format that can be deployed on the student management platform, such as SavedModel or TorchScript file of TensorFlow or PyTorch.

[0047] Next, we will use the APIs of these models to load them and input them into the model of the student management platform. After the model is loaded, the student's personal information and course information will be used to generate a personalized learning path.

[0048] Machine learning algorithms such as random forests, gradient boosted trees, or support vector machines can be used to predict how well a student will perform on a particular course. These predictions are then used to adjust the learning path to ensure that each student achieves the best possible learning outcomes.

[0049] In the student management platform, dashboards or other visualization tools can be used to display learning outcomes, such as grade distribution graphs, progress lines, or learning curves. These dashboards can help students and administrators better understand their learning progress.

[0050] Finally, students’ new achievements can be added to learning paths and dashboards periodically so that students and administrators can continually track their progress and make adjustments as needed.

[0051] Through the above steps, personalized learning paths and learning outcomes tracking can be achieved in the student management platform, and a visual dashboard can also be provided to help students and administrators better understand and grasp the learning progress.

[0052] Specifically, Figure 3 As shown, the deep learning framework includes: (1) Embedding layer Determine the basic learning unit vector corresponding to the learning process data ; Wherein, the basic learning unit vector Including knowledge point-exercise feature vector and the learner's ability to answer questions feature vector , Knowledge Point - Exercise Feature Vector Includes: Knowledge point number feature vector , Exercise number feature vector And the knowledge point difficulty feature vector ; Learner's ability to answer questions feature vector Includes: knowledge point corresponding to the question answer result vector , the time spent answering the corresponding knowledge point questions And the number of prompts . The knowledge point-exercise feature vector and the learner's ability to answer questions feature vector Input to LSTM neural network Processed in to get the basic learning unit embedding vector .

[0053]

[0054]

[0055]

[0056] in, , is the weight, and is the bias term.

[0057] The difficulty of the test is , Indicates the total number of times a question has been answered. Indicates the score for answering a question correctly.

[0058] (2) Learning layer ① The basic learning unit vector corresponding to the learning process data in (1) , the basic learning unit vector at the previous moment , the time interval vector between two learning , and the relevant knowledge state vector at the previous moment Connect to get the learning vector .

[0059]

[0060]

[0061] in, Represents the semantic representation vector of the knowledge points contained in the current exercise, Represents the knowledge state output vector at the previous moment.

[0062] ②The learning vector go through The activation function is used to obtain the learning gain vector

[0063] ③The learning vector go through The activation function is used to obtain the learning gate vector ,The main function of the learning gate vector is to control the students’ learning absorption ability.

[0064]

[0065] ④ By multiplying the learning gate vector that controls the learning absorption capacity with the learning gain vector, the actual learning gain vector is obtained

[0066] By transforming the actual learning gain vector Vector of knowledge points in exercises Multiplying them together, we can get the relevant learning gain vector of the relevant knowledge points

[0067] (3) Forgetting layer According to the Ebbinghaus forgetting curve theory, the number of remembered knowledge points decays exponentially over time. Although forgetting is inevitable, we can consolidate memory and prevent forgetting by regularly reviewing and repeating learning. In order to model complex forgetting defects, we designed a forgetting layer that connects the output state of the previous moment, the learning gain, and the interval time. The activation function obtains the forget vector

[0068] By adding the forget vector The knowledge state at the previous moment Multiplying together can eliminate the effect of forgetting. Therefore, we can calculate the current state of knowledge .

[0069] (4) Prediction layer The prediction layer uses the current state of knowledge Predict the student's next exercise question Performance on

[0070] Prediction Layer The output range of is (0, 1). We set the output threshold of the prediction layer to 0.5. , it indicates The answer is correct, otherwise, it means the answer is wrong.

[0071] (5) Objective function

[0072] Setting the forecast value With actual value The cross entropy logarithmic loss function is used as the objective function to optimize the model. Represents all parameters, represents the regularization hyperparameter. The objective function is minimized using the Adam optimizer on the mini-batch.

[0073] In the above function, Weight, is the bias term ( ).

[0074] On the other hand, this embodiment also proposes an artificial intelligence education knowledge tracking method based on big data, such as Figure 2 As shown, the following steps are included: Collect students' basic information data, learning behavior data, and other factors that may affect their learning behavior, and use the big data processing framework to store and process the collected data to construct a data set; Receiving the data set, and storing the data set in a distributed file system after classifying the data set, and obtaining a sub-data set; Extracting a sub-dataset in the distributed file system, and converting the sub-dataset into a feature representation suitable for machine learning to obtain feature data; Using a deep learning framework to train a machine learning model based on the feature data, and constructing a neural network model after tuning the model; The neural network model is input into the student management platform, and a personalized learning path is generated in combination with the data set, and the learning outcomes are tracked, and the learning outcomes are displayed through the user interface.

[0075] In this embodiment, big data analysis of student learning behavior is implemented. Through machine learning algorithms, personalized learning content and paths are recommended to students. Students' learning status and statistics of the learning status can be tracked independently, so that learning information can be obtained in a timely and accurate manner.

[0076] In addition, the present invention also provides a terminal device. The big data-based artificial intelligence education knowledge tracking method involved in this embodiment is mainly applied to the terminal device, which can be a PC, a portable computer, a mobile terminal, or other device with display and processing functions.

[0077] Specifically, the terminal device may include a processor (such as a CPU), a communication bus, a user interface, a network interface, and a memory. The communication bus is used to realize the connection and communication between these components; the user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard); the network interface may optionally include a standard wired interface, a wireless interface (such as a WI-FI interface); the memory may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory, and the memory may optionally be a storage device independent of the aforementioned processor.

[0078] Among them, the memory stores a readable storage medium, and the readable storage medium stores an artificial intelligence education knowledge tracking program. The processor can call the artificial intelligence education knowledge tracking program stored in the memory and execute the artificial intelligence education knowledge tracking method based on big data provided by an embodiment of the present invention.

[0079] It is understood that a readable storage medium can be a tangible device that can hold and store instructions used by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples of computer-readable storage media (a non-exhaustive list) include: portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), static random access memories (SRAM), portable compact disk read-only memories (CD-ROM), digital versatile disks (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or raised structures in grooves on which instructions are stored, and any suitable combination of the above. The computer-readable storage medium used here is not to be interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0080] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0081] The computer program instructions for performing the operation of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed completely on a user's computer, partially on a user's computer, executed as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be customized by utilizing the state information of a computer-readable program instruction, and the electronic circuit may execute a computer-readable program instruction, thereby realizing various aspects of the present disclosure.

[0082] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An artificial intelligence education knowledge tracking system based on big data, characterized in that: include: The data collection module is used to collect students' basic information data, learning behavior data, and other factors that may affect their learning behavior, and use the big data processing framework to store and process the collected data to construct a data set; A data input module is used to receive the data set, classify the data set, store it in a distributed file system, and obtain a sub-data set; A feature extraction module, used to extract a sub-dataset in the distributed file system and convert the sub-dataset into a feature representation suitable for machine learning to obtain feature data; A model training module is used to train a machine learning model using a deep learning framework according to the feature data, and to build a neural network model after tuning the model; The model prediction module is used to input the neural network model into the student management platform, generate a personalized learning path in combination with the data set, track the learning outcomes, and display the learning outcomes through the user interface.

2. According to the big data-based artificial intelligence education knowledge tracking system of claim 1, it is characterized in that: The data acquisition module comprises: A data collection unit, used to automatically or manually collect students' basic information and learning behavior data; A big data analysis unit, which is used to clean, transform, summarize and store the collected data in a database; The data processing and analysis unit is used to use machine learning and data mining techniques to screen out students' learning behaviors and factors that may affect their learning behaviors from the database.

3. According to claim 2, an artificial intelligence education knowledge tracking system based on big data is characterized in that: The basic information and learning behavior data are obtained through questionnaires, online form filling, test score records, student management platforms, and education platforms.

4. According to the big data-based artificial intelligence education knowledge tracking system of claim 1, it is characterized in that: The data input module comprises: The data reading unit is used to read data sets from input sources using Python; The data classification unit is used to classify the read data set using a naive Bayes classifier. After the classification is completed, the classification result is written into the distributed file system; The sub-dataset acquisition unit is used to retrieve the corresponding sub-dataset from the distributed file system according to the classification result and request.

5. According to claim 4, an artificial intelligence education knowledge tracking system based on big data is characterized in that: The data input module further includes: A sub-dataset loading unit, used to write code in Python to load the sub-dataset into memory; The API interface unit is used to provide an API interface to connect to the feature extraction module to obtain a sub-dataset.

6. According to claim 5, an artificial intelligence education knowledge tracking system based on big data is characterized in that: The feature extraction module comprises: A data extraction unit, used to extract the required sub-dataset from the distributed file system through an API interface; Data preprocessing unit, used to remove terminal duplicates of sub-datasets and fill missing values, eliminate potential outliers and unbalanced data distribution; The feature representation unit is used to train the preprocessed data using a machine learning algorithm to obtain feature representation data.

7. The big data-based artificial intelligence education knowledge tracking system according to claim 1 is characterized in that: The model training module includes: Feature processing unit, used to pre-process the original feature representation data, including data cleaning, normalization, and standardization operations; The model building unit is used to use TensorFlow and its provided APIs and tools to build neural network models.

8. The big data-based artificial intelligence education knowledge tracking system according to claim 7 is characterized in that: The model training module further includes: Model compilation unit, used to compile the neural network model after it is built; The training unit is used to train the compiled neural network model. After the model training is completed, it is evaluated and further optimized if the model performs poorly.

9. The artificial intelligence education knowledge tracking system based on big data according to claim 5 is characterized in that: The model prediction module comprises: A deployment unit, used to convert the trained neural network model into a format that can be deployed on the student management platform; The unit is used to load the API and input it into the model of the student management platform. After the model is loaded, the student's personal information and course information are used to generate a personalized learning path; A prediction unit that uses a random forest algorithm to predict a student's performance on a specific course, and then uses these predictions to adjust the learning path.

10. The big data-based artificial intelligence education knowledge tracking system according to claim 9 is characterized in that: The model prediction module also includes: a display unit, which is used to use visualization tools in the student management platform to display learning results.

Images