Semiconductor education enhancement system and method based on artificial intelligence

Abstract

The invention discloses a semiconductor education enhancement system and method based on artificial intelligence. The system comprises a data collection module, an artificial intelligence processing module, a personalized teaching recommendation module, a virtual laboratory module, an online feedback and evaluation module and a knowledge graph construction module. According to the invention, multiple modules are arranged to combine cyclic feedback learning with an artificial intelligence model, and through automatic data collection and analysis, a teacher can more effectively understand the learning condition of students, so that time is saved and teaching quality is improved; customized learning resources and activities can be provided according to characteristics and requirements of each student, so that the learning effect is improved; through virtual laboratory simulation, the demand on physical laboratory resources is reduced, and the education cost is reduced; through interactive and immersive learning experience, the participation degree and learning motivation of students are improved; furthermore, the collected data can be used for educational research to improve teaching methods and curriculum design.

Description

Technical Field

[0001] The present invention relates to the field of intelligent education technology, and in particular, to a semiconductor education enhancement system and method based on artificial intelligence. Background Art

[0002] With the continuous progress of technology, semiconductor technology plays an increasingly important role in the modern electronic information industry. As the foundation of information technology, the improvement of the performance and the reduction of the cost of semiconductor devices have always been the key factors driving the development of the entire industry. In recent years, with the rapid development of artificial intelligence (AI) technology, the application of AI in various links such as semiconductor design, manufacturing, and testing has gradually increased, bringing new opportunities and challenges to the semiconductor industry.

[0003] In the field of semiconductor education, due to the lag in the update of educational resources and the imbalance in the distribution of educational resources in the traditional education model, it has gradually been unable to meet the rapidly developing industry needs. Therefore, it is particularly important for the education model to keep pace with the times and use emerging technologies such as virtual reality, big data, and artificial intelligence to improve the quality and efficiency of education. At present, the application of AI technology in the field of education, especially in personalized learning, intelligent tutoring, intelligent recommendation of teaching resources, etc., has begun to show its unique advantages. Combining semiconductor education with artificial intelligence can not only greatly improve the teaching quality and efficiency, reduce the imbalance in the distribution of educational resources, but also truly achieve individualized teaching, reasonably arrange the learning progress and learning content according to the knowledge level, cognitive ability, and memory ability of each student.

[0004] At the same time, the field of semiconductor education also faces some challenges. For example, the complexity and update speed of semiconductor technology require that educational content must be continuously updated to keep up with the development of the industry. In addition, the practicality and technicality of the semiconductor industry require that there must be sufficient experimental and practical links in the education process to cultivate students' practical operation ability. After applying artificial intelligence in the semiconductor field, relying on the powerful information search and processing ability of artificial intelligence, it can ensure the real-time update of teaching content and greatly improve the lag of educational resources. Moreover, combining artificial intelligence, digital twin, and virtual training, digitally replicating the actual production line into the teaching system, and supplemented by computers, virtual reality, or augmented reality technology, can enable students to experience the actual production line training experience without leaving home.

[0005] However, the currently applied semiconductor education systems and methods do not adequately consider the comprehensiveness and internal relevance of semiconductor process knowledge and still cannot meet the needs of efficiently cultivating technical personnel. Summary of the Invention

[0006] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a semiconductor education enhancement system and method based on artificial intelligence.

[0007] To achieve the foregoing invention purpose, the technical solutions adopted by the present invention include: In the first aspect, the present invention provides a semiconductor education enhancement system based on artificial intelligence, which includes: A data collection module for collecting learning materials and students' learning data, where the learning data is related to the students' learning process of semiconductor process knowledge; An artificial intelligence processing module for analyzing the students' learning progress and understanding degree of the semiconductor process knowledge based on the learning data; A personalized teaching recommendation module for generating personalized recommendations for the students based on the analysis results of the artificial intelligence processing module, where the personalized recommendations include selected learning materials and corresponding experimental projects; A virtual laboratory module for providing a virtual laboratory that simulates the semiconductor manufacturing process for the students using virtual reality technology, and conducting simulation experiments based on the experimental projects and outputting experimental results to the students; An online feedback and evaluation module for collecting the students' learning feedback on the learning materials and experimental projects and updating the students' learning data; A knowledge graph construction module for constructing a knowledge graph in the semiconductor field, and the knowledge graph is used to assist the students in understanding the learning data and experimental projects.

[0008] In the second aspect, the present invention also provides a semiconductor education method based on the above semiconductor education enhancement system, which includes the following steps: Step 1: Collect learning materials and students' learning data through the data collection module; Step 2: Analyze the students' learning progress and understanding degree of the semiconductor process knowledge based on the learning data through the artificial intelligence processing module; Step 3: Generate personalized recommendations for the students based on the analysis results of the artificial intelligence processing module, where the personalized recommendations include selected learning materials and corresponding experimental projects; Step 4: Provide a virtual laboratory that simulates the semiconductor manufacturing process for the students using virtual reality technology, and conduct simulation experiments based on the experimental projects and output experimental results to the students; Step 5: Collect the students' learning feedback on the learning materials and experimental projects and update the students' learning data; Step 6: Construct a knowledge graph in the semiconductor field to assist the students in understanding the learning data and experimental projects.

[0009] In a third aspect, the present invention also provides a readable storage medium storing a computer program which, when run, executes the steps of the above semiconductor education method.

[0010] Based on the above technical solutions, compared with the prior art, the beneficial effects of the present invention at least include: The semiconductor education enhancement system provided by the present invention is provided with multiple modules, combining cyclic feedback learning and artificial intelligence models. Through automated data collection and analysis, teachers can more effectively understand the learning situation of students, thus saving time and improving teaching quality; it can provide customized learning resources and activities according to the characteristics and needs of each student, thereby improving learning effects; through virtual laboratory simulations, the demand for physical laboratory resources is reduced, and educational costs are lowered; through interactive and immersive learning experiences, student participation and learning motivation are increased; in addition, the data collected can be used for educational research to improve teaching methods and curriculum design.

[0011] The above description is only an overview of the technical solution of the present invention. In order to enable those skilled in the art to more clearly understand the technical means of the present application and implement it in accordance with the content of the specification, the following is described by way of preferred embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a schematic diagram of associated factors of a lithography process provided by a typical embodiment of the present invention; Figure 2 is a schematic diagram of different etching results of an etching process provided by another typical embodiment of the present invention; Figure 3 is a schematic diagram of associated factors of an etching process provided by another typical embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0013] In view of the deficiencies in the prior art, the inventors of this case have, through long-term research and a large number of practices, been able to propose the technical solution of the present invention. The following will further explain the technical solution, its implementation process, principles, etc.

[0014] In the following description, many specific details are set forth in order to provide a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited by the specific embodiments disclosed below.

[0015] An embodiment of the present invention provides a semiconductor education enhancement system based on artificial intelligence, aiming to improve the efficiency and quality of semiconductor education through artificial intelligence technology, achieve personalized teaching, and reduce the consumption of educational resources. The system forms a comprehensive education platform by integrating six modules, namely a data collection module, an artificial intelligence processing module, a personalized teaching recommendation module, a virtual laboratory module, an online feedback and evaluation module, and a knowledge graph construction module, to support students, teachers, and educational institutions, and it includes: A data collection module for collecting learning materials and students' learning data, where the learning data is related to the students' learning process of semiconductor process knowledge; An artificial intelligence processing module for analyzing the students' learning progress and understanding degree of the semiconductor process knowledge based on the learning data; A personalized teaching recommendation module for generating personalized recommendations for the students based on the analysis results of the artificial intelligence processing module, where the personalized recommendations include selected learning materials and corresponding experimental projects; A virtual laboratory module for providing a virtual laboratory that simulates the semiconductor manufacturing process for the students using virtual reality technology, and conducting simulation experiments based on the experimental projects and outputting experimental results to the students; An online feedback and evaluation module for collecting the students' learning feedback on the learning materials and experimental projects and updating the students' learning data; A knowledge graph construction module for constructing a knowledge graph in the semiconductor field, where the knowledge graph is used to assist the students in understanding the learning data and experimental projects.

[0016] Regarding the artificial intelligence processing module, in some embodiments, the artificial intelligence processing module includes a recurrent neural network and an LSTM model. The recurrent neural network is used to track the students' learning progress, and the LSTM model is used to track the students' understanding degree.

[0017] In some embodiments, the algorithms adopted by the artificial intelligence processing module include any one or a combination of two or more of decision tree, random forest, neural network, and convolutional neural network.

[0018] In some embodiments, the analysis content of the artificial intelligence processing module includes the students' classroom performance, homework grades, and experimental operation data.

[0019] In some embodiments, the semiconductor process includes spin coating, lithography, development, dry etching, heat treatment, coating, oxidation, doping, wet etching, cleaning, and corresponding process parameters and process effects; The artificial intelligence processing module also analyzes the connections between different processes based on the semiconductor process to evaluate the student's understanding of the connections between processes. For example, the influence of the photoresist morphology after lithography on the sidewall steepness of dry etching; the influence of the formed barrier layer during oxidation on the implantation depth of the ion implantation process in doping; the magnitude of the contact resistance with the substrate after different heat treatment processes for different combinations of thin film types and thicknesses in the multi-layer thin film coating process; the influence of the thickness of the hard mask prepared by the coating process on the wet etching process, etc.

[0020] Taking the lithography and etching processes as an example, see Figure 1 As shown, the parameters of the process in the lithography process will affect each other. The thickness and uniformity of the photoresist, the light intensity and time of exposure affect the exposure effect, and the exposure effect will further affect the development process. Therefore, all parameters must be considered comprehensively to ensure the overall process effect. Only when all conditions are suitable can an ideal lithography effect be obtained. This requires precise control of all lithography process parameters. The following Figure 1 lists that there will be 9 combination methods considering two parameters of exposure time and development time. Except for the combination of suitable exposure time and development time, any combination of exposure and development conditions may result in unqualified lithography effects. For example, insufficient or excessive exposure, and insufficient or excessive development will cause defects or distortion in the lithography pattern. In this module, students can deeply understand the process model under different parameters through this system during the learning process. This helps to understand the specific influence of parameter changes on the lithography effect, so as to better master each process.

[0021] The specific process is as follows: The artificial intelligence processing module generalizes different combinations of exposure and development through the decision tree algorithm, collects the process effects corresponding to different exposure and development times through web crawlers and manual input, performs machine learning on the pictures of the process effects through the convolutional neural network, and feeds back to the decision tree algorithm, enabling the decision tree algorithm to accurately predict the process effects achieved by the exposure and development combinations.

[0022] During teaching, the understanding degree of students on the different process effects caused by the combination of exposure conditions and development conditions can be evaluated according to the answers of students or the results of simulation experiments, so as to enhance the students' full understanding of the internal connection between the exposure process and the development process.

[0023] In addition, regarding the etching process, many phenomena will occur in the etching process, such as Figure 2Only (1) shows a suitable etching case among those shown; in the remaining cases, there are problems with the etching process. (2) The etching of the "bird's beak" microgrooves is related to the ions reflected from the sidewall surface. The ions reflected multiple times by the sidewall near the groove corners exacerbate the etching of the groove corners, resulting in the microgroove phenomenon. The kinetic energy of the reactive ions can be increased by increasing the RF power to weaken the deflection effect of the charged sidewalls on the reactive ions; or the collision of the reactive ions can be enhanced by increasing the chamber pressure to reduce the mean free path of the reactive ions and weaken their physical bombardment of the groove bottom corners. (3) Over-etching is usually caused by an unreasonable design of the process etching time, leading to over-etching of the material. Generally, it can be achieved by optimizing the process plan or designing a "self-stopping layer". The combination of dry etching and wet etching can be used to reduce over-etching; or it can be achieved by growing a "self-stopping layer" with an etching rate relatively slower than the expected etched material. (4) Insufficient etching mask selection ratio is usually due to the unreasonable design of the thickness of the expected etched material and the etching mask, resulting in the photoresist being completely consumed before the etched material. The etching can be optimized by increasing the thickness of the etching mask; or the etching selection ratio can be optimized by changing the etching parameters. The optimized parameters include the chamber pressure, etching gas ratio, etching gas flow rate, etching ICP, RF power, chamber temperature, etc. to improve the etching selection ratio. (5) An inappropriate etching angle is usually directly related to the angle of the etching mask. The etching angle can be increased by optimizing the steepness of the etching mask; or the etching selection ratio can be optimized by changing the etching parameters. Usually, the etching angle can be optimized by increasing the RF power to make the etching reactive ions move downward more easily, making the etching more directional and enhancing the anisotropic etching. However, too high RF power can also cause unnecessary etching damage to the sample. (6) The main reason for the generation of the etching micro-mask during etching is the shedding of the etching mask or the non-volatile by-products generated during the etching reaction that cannot be removed in time and are deposited at the bottom of the etching groove. As the etching continues, a "weed growth" phenomenon occurs at the bottom of the groove. The micro-mask effect can be reduced by introducing a gas that reacts with the by-products.

[0024] As can be seen from the above examples, there are many factors affecting the final etching result, and their correlation degrees are also different. Therefore, the random forest algorithm can be used to predict the etching result. Taking inductively coupled plasma (ICP) etching as an example, the specific correlation is as follows Figure 3 shown.

[0025] In the ICP etching process, different etching parameters correspond to different decision trees. The parameter levels under different decision trees have different effects on the etching rate, selectivity, and etching profile. First, calculate the effects of each parameter in different decision trees on the etching result, and then summarize the effects generated in each decision tree. Calculate the final etching result according to different influence weights. The weights of each factor are calculated through actual etching experiments.

[0026] When teaching, construct an association tree based on the different etching conditions obtained by the above decision tree algorithm and the impacts caused by different associated processes to train students, including specific demonstrations, answering questions, and simulation experiments, etc., and collect the performance of students to evaluate their understanding of the etching process conditions and the internal connections of various associated processes.

[0027] As a typical example, the algorithms used by the artificial intelligence processing module include, but are not limited to, algorithms such as decision trees, random forests, neural networks, and convolutional neural networks. Input the process parameters and corresponding process effects of processes such as spin coating, photolithography, development, dry etching, coating, oxidation, doping, wet etching, and cleaning in semiconductor processes into the artificial intelligence processing module. This module can perform deep learning on the input content through algorithms such as decision trees, random forests, neural networks, and convolutional neural networks, and can simulate the process parameters and process results under different process requirements according to the learning results and store them in the database. The system arranges teaching content for students according to the content in the database and collects the learning achievements of students at the same time. Use a recurrent neural network to track the learning content and learning achievements of students, and use an LSTM model to track the process of the dynamic change of students' knowledge proficiency over time.

[0028] Regarding the personalized teaching recommendation module, in some implementation schemes, the algorithms adopted by the personalized teaching recommendation module include any one or a combination of two or more of clustering algorithms, classification algorithms, regression algorithms, and ensemble learning.

[0029] In some implementation schemes, the algorithms adopted by the personalized teaching recommendation module further include a feedback loop algorithm, and the feedback loop algorithm is used to analyze the error-prone knowledge points of the students in the previous learning and recommend the relevant content of the error-prone knowledge points in the subsequent learning.

[0030] In some implementation schemes, the scoring system of the personalized teaching recommendation module includes teacher scoring, peer scoring, and self-scoring by students.

[0031] As a typical example, the personalized teaching recommendation module collects and makes intelligent recommendations for students' learning interests through algorithms such as clustering algorithms (such as k-means, hierarchical clustering), classification algorithms (such as nearest neighbor, naive Bayes, decision tree), regression algorithms (such as linear regression, logistic regression), and ensemble learning (such as gbdt, xgboost, lightgbm); through the feedback loop algorithm, it recommends that students repeatedly learn about their own error-prone points and the key and difficult points in the teaching process.

[0032] In some embodiments, the knowledge graph constructed by the knowledge graph construction module includes the connections between different processes and the students' understanding of the connections between different processes.

[0033] In the above technical solution, the online feedback and evaluation module allows teachers to monitor the students' learning progress and understanding level, the knowledge graph construction module is used to assist teaching, and the knowledge graph constructed by the knowledge graph construction module is used to display the working principles and manufacturing processes of semiconductor devices to help students build a knowledge system. In addition, the virtual laboratory simulation module allows students to perform experimental operations in a virtual environment without actual physical equipment. The learning resources include video lectures, reading materials, online exercises, and virtual laboratory experiments.

[0034] The second aspect of the embodiments of the present invention also provides a semiconductor education method based on the semiconductor education system provided in any of the above embodiments, which includes the following steps: Step 1: Collect learning materials and students' learning data through the data collection module; Step 2: Analyze the students' learning progress and understanding level of the semiconductor process knowledge based on the learning data through the artificial intelligence processing module; Step 3: Generate personalized recommendations for the students based on the analysis results of the artificial intelligence processing module, and the personalized recommendations include the selected learning materials and corresponding experimental projects; Step 4: Use virtual reality technology to provide the students with a virtual laboratory for simulating the semiconductor manufacturing process, and perform simulation experiments based on the experimental projects and output the experimental results to the students; Step 5: Collect the students' learning feedback on the learning materials and experimental projects, and update the students' learning data; Step 6: Construct a knowledge graph in the semiconductor field to assist the students in understanding the learning data and experimental projects.

[0035] The third aspect of the embodiments of the present invention also provides a readable storage medium, in which a computer program is stored, and when the computer program is run, it executes the steps of the above semiconductor education method.

[0036] The technical solutions of the present invention will be further described in detail through several embodiments below. However, the selected embodiments are only used to illustrate the present invention and do not limit the scope of the present invention.

[0037] Embodiment 1 This embodiment exemplifies a semiconductor education enhancement system, which consists of the following modules: Data collection module: This module has two functions. One is to collect and organize teaching resources through Internet search or manual input to form a teaching resource database. The collected resources include, but are not limited to, the structures, principles, process designs, test methods of various semiconductor devices, as well as spin coating, lithography, development, etching, coating, oxidation, doping, and packaging processes involved in semiconductor micro-nano processing. The other is to be responsible for collecting students' learning data from multiple sources. Students' learning data includes, but is not limited to, students' classroom participation, assignment submission, exam scores, experimental operation records, and interactions on the online learning platform, etc. This module can be seamlessly integrated with existing systems such as the school's Student Information System (SIS), Learning Management System (LMS), and Laboratory Management System (LIMS). Data can be automatically collected through students' daily learning activities or manually input through teachers' observations and evaluations.

[0038] Artificial intelligence processing module: This module uses machine learning algorithms to analyze the collected data, generate teaching content, and evaluate students' learning progress and understanding. The algorithms used by the artificial intelligence processing module include, but are not limited to, decision tree, random forest, neural network, and convolutional neural network algorithms, etc. Input the process parameters and corresponding process effects of processes such as spin coating, lithography, development, dry etching, coating, oxidation, doping, wet etching, and cleaning in semiconductor processes in the data collection module into the artificial intelligence processing module. This module can perform deep learning on the input content through algorithms such as decision tree, random forest, neural network, and convolutional neural network, and can simulate process parameters and process results under different process requirements according to the learning results and store them in the database. The system arranges teaching content for students according to the content in the database and simultaneously collects students' learning achievements. Using a recurrent neural network, track students' learning content and learning achievements, and use the LSTM model to track the dynamic change process of students' knowledge proficiency over time. This module can also collect teaching resources and industry information in the Internet in real time, analyze and process the collected content in real time, update and improve the teaching content in real time, and save it in the data collection module. Set up an AI assistant in the system. Students can ask questions to the AI assistant by inputting text or files. The AI assistant analyzes the text and files, retrieves and processes data on the Internet, and obtains the best answer.

[0039] Personalized Teaching Recommendation Module: Based on the analysis results of the artificial intelligence processing module, personalized learning resources and experimental projects are recommended for each student. The recommendation system can generate recommendations based on the student's learning history, interests, and current learning status.

[0040] Virtual Laboratory Simulation Module: Using virtual reality (VR) technology, this module provides students with a simulated laboratory environment, enabling them to perform experimental operations without physical equipment. This module can simulate the key steps in the semiconductor manufacturing process, including lithography, etching, coating, oxidation, cleaning, and doping, which are micro-nano processing technologies used in the semiconductor industry.

[0041] Online Feedback and Evaluation Module: This module allows students to provide feedback after completing learning tasks, while teachers can monitor the students' learning progress and understanding. This module supports formative assessment, including self-assessment, peer assessment, and teacher assessment, to promote students' self-reflection and continuous improvement.

[0042] Knowledge Graph Construction Module: This module constructs a knowledge graph in the semiconductor field, connecting knowledge points, concepts, principles, and applications to form an interconnected knowledge network. The knowledge graph can help students better understand and master the overall framework of semiconductor technology.

[0043] Example 2 The process of semiconductor education using the semiconductor education enhancement system provided in Example 1 in this example is as follows: In the system background, the data collection module collects relevant books, materials, papers, and patents on semiconductor devices and manufacturing processes through web crawlers and manual input. The artificial intelligence processing module uses text processing technologies, such as regular expressions and natural semantic recognition technologies, to mine the key information in the collected books, materials, papers, and patents and perform format standardization processing, outputting it as teaching content. The output form of the teaching content can be text, pictures, AI pictures, audio, AI audio, video, or AI video, etc.

[0044] Such as the lithography process in semiconductor manufacturing. This process consists of three steps: spin coating, exposure, and development. These three process steps are interrelated and generally carried out in the order of spin coating, exposure, and development. The data collection module collects relevant process parameters through web crawlers and manual input, such as the spin coating, exposure, and development parameters corresponding to different photoresists; the results of normal and abnormal spin coating of photoresist; the process results of normal exposure, under-exposure, and over-exposure; and the process results of normal development, under-development, and over-development. Through deep learning of the above process content by the artificial intelligence processing module, for example, using a convolutional neural network to deeply learn the three different lithography states of normal exposure, under-exposure, and over-exposure, it can determine whether the process result corresponding to a certain photoresist under different exposure parameters is normal exposure, under-exposure, or over-exposure, and derive the process result diagram through AI drawing; it can also, conversely, infer from the actual exposure process result diagram whether the exposure result is normal exposure, under-exposure, or over-exposure, and then infer the range of exposure parameters used.

[0045] On this basis, the artificial intelligence processing module forms systematic lithography-related teaching content according to the materials and information collected by the data collection module as follows: The definition and principle of the lithography process, generating text, schematic diagrams, or animations of the lithography process principle based on the text, pictures, and video materials in the data collection module; The process flow of the lithography process, generating text, schematic diagrams, or animations of the lithography process flow based on the text, pictures, and video materials in the data collection module and introducing the three process steps of spin coating, exposure, and development in stages; The spin coating process step, summarizing different spin coating methods and their corresponding equipment based on the text, pictures, and video materials in the data collection module and displaying relevant pictures or preparation operation videos. Summarize the impacts of different factors on spin coating, and generate effect diagrams of the impacts of different factors on spin coating through algorithms or models such as the Diffusion model, DALL-E network, and image stitching technology; The exposure process step, summarizing different exposure methods and their corresponding equipment based on the text, pictures, and video materials in the data collection module and displaying relevant pictures or preparation operation videos. Summarize the impacts of different factors on exposure, analyze how under-exposure and over-exposure occur and how to optimize them, and generate effect diagrams of the impacts of different factors on exposure as well as effect diagrams of normal exposure, under-exposure, and over-exposure through algorithms or models such as the Diffusion model, DALL-E network, and image stitching technology; Developing process steps: Summarize different developing methods and their corresponding equipment based on the text, pictures, and video materials in the data collection module, and display relevant pictures or preparation operation videos. Summarize the impacts of different factors on development, analyze how underdevelopment and overdevelopment occur and how to optimize them, and generate effect diagrams of the impacts of different factors on development, as well as effect diagrams of normal development, underdevelopment, and overdevelopment through algorithms or models such as the Diffusion model, DALL-E network, and image stitching technology. Analyze the mutual influences among the three related processes of spin coating, exposure, and development, and summarize and list the final process results caused by the mutual influences of the three process steps through permutation and combination, and give optimization and solution methods when there are poor process results.

[0046] In addition to the above teaching content, it is also possible to generate content such as the latest progress in lithography technology, the development of lithography equipment, and the latest information in the lithography industry.

[0047] The teaching content is updated in real time with the continuous collection of data and the continuous operation of the artificial intelligence processing module.

[0048] At the beginning of the semester, the teacher starts the system of the present invention through the school's teaching management platform and sets parameters and goals related to the course.

[0049] The data collection module automatically collects students' background information, previous academic achievements, course registration information, etc. from the Student Information System (SIS), Learning Management System (LMS), and Laboratory Management System (LIMS). During the course, this module continuously collects students' real-time data, including classroom attendance records, assignment submission situations, online discussion participation degrees, experimental operation records, etc. The collected data is input into the artificial intelligence processing module, which analyzes the students' data using preset machine learning algorithms. The algorithms analyze students' participation degrees, learning habits, assignment correct rates, knowledge point mastery situations, etc. to evaluate students' learning progress and understanding levels. The artificial intelligence processing module can also identify differences among students and create personalized learning profiles for each student.

[0050] Based on the analysis results of the artificial intelligence processing module, the personalized teaching recommendation module recommends resources suitable for each student's learning needs and styles. The recommended resources may include video lectures, reading materials, online exercises, virtual laboratory experiments, etc., aiming to supplement and strengthen students' learning in class. The teacher can, with the help of the system, adjust the teaching plan according to the recommendation results to better meet the personalized needs of students.

[0051] Students access a simulated semiconductor manufacturing environment through the virtual laboratory simulation module to perform various experimental operations. The module provides an operation interface and feedback mechanism similar to a physical laboratory, enabling students to practice and master key skills without physical equipment. The operation data of students in the virtual laboratory is collected and sent back to the artificial intelligence processing module for further learning analysis and feedback. After completing the learning tasks, students provide feedback through the online feedback and evaluation module, including self-assessment, peer evaluation, and teacher evaluation. Teachers can monitor the learning progress and understanding of students in real time and adjust teaching strategies and content in a timely manner. The system supports formative assessment, encouraging students to engage in self-reflection and peer learning.

[0052] The knowledge graph construction module constructs a knowledge graph in the semiconductor field based on the course content and the learning progress of students. The knowledge graph shows the connections between knowledge points in a graphical way, helping students build a knowledge system and understand complex concepts. Teachers can use the knowledge graph for teaching design, enabling students to better understand the course content.

[0053] The system self-optimizes according to the feedback and learning outcomes of students, adjusting algorithms and recommendation strategies to improve teaching effectiveness. Teachers and educational researchers can use the large amount of data collected by the system for educational research, continuously improving teaching methods and curriculum design.

[0054] Based on the above embodiments, it is clear that the semiconductor education enhancement system provided by the embodiments of the present invention sets up multiple modules, combines cyclic feedback learning and artificial intelligence models, and through automated data collection and analysis, teachers can more effectively understand the learning situation of students, thus saving time and improving teaching quality; it can provide customized learning resources and activities according to the characteristics and needs of each student, thereby improving learning effects; through virtual laboratory simulation, it reduces the demand for physical laboratory resources and lowers educational costs; through interactive and immersive learning experiences, it improves students' participation and learning motivation; in addition, the collected data can be used for educational research to improve teaching methods and curriculum design.

[0055] It should be understood that the above embodiments are only used to illustrate the technical concept and characteristics of the present invention, and their purpose is to enable those familiar with this technology to understand the content of the present invention and implement it accordingly, and it should not be used to limit the protection scope of the present invention. Any equivalent changes or modifications made according to the spirit and essence of the present invention should be covered within the protection scope of the present invention.

Claims

1. An artificial intelligence-based semiconductor education enhancement system, characterized in that, Including: A data collection module for collecting learning materials and students' learning data, where the learning data is related to the students' learning process of semiconductor process knowledge; An artificial intelligence processing module for analyzing the students' learning progress and understanding degree of the semiconductor process knowledge based on the learning data; A personalized teaching recommendation module for generating personalized recommendations for the students based on the analysis results of the artificial intelligence processing module, where the personalized recommendations include selected learning materials and corresponding experimental projects; A virtual laboratory module for providing a virtual laboratory simulating the semiconductor manufacturing process for the students by using virtual reality technology, and conducting simulation experiments based on the experimental projects and outputting experimental results to the students; An online feedback and evaluation module for collecting the students' learning feedback on the learning materials and experimental projects and updating the students' learning data; A knowledge graph construction module for constructing a knowledge graph in the semiconductor field, where the knowledge graph is used to assist the students in understanding the learning data and experimental projects.

2. The semiconductor education enhancement system according to claim 1, wherein The artificial intelligence processing module includes a recurrent neural network and an LSTM model. The recurrent neural network is used to track the students' learning progress, and the LSTM model is used to track the students' understanding degree.

3. The semiconductor education enhancement system according to claim 2, characterized in that, The algorithms adopted by the artificial intelligence processing module include any one or a combination of two or more of decision tree, random forest, neural network, and convolutional neural network.

4. The semiconductor education enhancement system according to claim 1 or 2, characterized in that, The analysis content of the artificial intelligence processing module includes the students' classroom performance, homework scores, and experimental operation data.

5. The semiconductor education enhancement system according to claim 4, wherein The semiconductor process includes spin coating, lithography, development, dry etching, coating, oxidation, doping, heat treatment, wet etching, cleaning, and corresponding process parameters and process effects; The artificial intelligence processing module also analyzes the connections between different processes based on the semiconductor process to evaluate the students' understanding of the connections between processes.

6. The semiconductor education enhancement system according to claim 1, wherein The algorithms adopted by the personalized teaching recommendation module include any one or a combination of two or more of clustering algorithm, classification algorithm, regression algorithm, and ensemble learning.

7. The semiconductor education enhancement system according to claim 4, wherein The algorithms adopted by the personalized teaching recommendation module also include a feedback loop algorithm, which is used to analyze the error-prone knowledge points in the students' previous learning and recommend the relevant content of the error-prone knowledge points in the subsequent learning.

8. The semiconductor education enhancement system according to claim 1, wherein The scoring system of the personalized teaching recommendation module includes teacher scoring, peer scoring, and self-scoring by the students; And / or, the knowledge graph constructed by the knowledge graph construction module includes the connections between different processes and the students' understanding degree of the connections between different processes.

9. A semiconductor education method based on the semiconductor education system according to any one of claims 1-8, characterized in that, Including the following steps: Step 1: Collect learning materials and students' learning data through the data collection module; Step 2: Analyze the students' learning progress and understanding degree of the semiconductor process knowledge based on the learning data through the artificial intelligence processing module; Step 3: Generate personalized recommendations for the students based on the analysis results of the artificial intelligence processing module, where the personalized recommendations include selected learning materials and corresponding experimental projects; Step 4: Use virtual reality technology to provide a virtual laboratory for the student to simulate the semiconductor manufacturing process, conduct simulation experiments based on the experimental project, and output the experimental results to the student; Step 5: Collect the learning feedback of the student on the learning materials and experimental project, and update the learning data of the student; Step 6: Construct a knowledge graph in the semiconductor field to assist the student in understanding the learning data and experimental project.

10. A readable storage medium, characterized in that, The computer program is stored in the readable storage medium, and when the computer program is run, it executes the steps of the semiconductor education method described in claim 9.

Images