AI-based Personalized Educational Content and Learning Environment Optimization System

Abstract

The present invention relates to an AI-based customized educational content and learning environment optimization system. It collects learning patterns, biosignals, user input, and feedback data in real time from a user terminal and analyzes them into quantitative indicators such as learning propensity, comprehension, and fatigue through an AI analysis module. Subsequently, the system maximizes the learner's learning effectiveness and provides a personalized learning experience through content recommendation and difficulty adjustment, multi-modal content conversion, real-time speech recognition-based subtitle generation and multi-language translation, learning environment optimization using IoT sensors, cloud-based integrated management, and gamification. The objective of the present invention is to maximize learning effectiveness by providing customized learning content through real-time analysis of the learner's status and feedback, and by optimizing the learning environment. Through this, optimal learning conditions are established by providing content tailored to the learner's learning tendencies and level of comprehension, while simultaneously adjusting the learning environment to account for fatigue and concentration. Furthermore, the invention aims to enhance learner motivation through gamification and social interaction. This invention can maximize learning effectiveness by analyzing learner data in real time and optimizing learning content and environments. Learners can manage their concentration and fatigue by receiving personalized content and environments, and can participate actively through various feedback and motivation systems. Furthermore, continuous learning improvement is possible by reflecting learner performance in real time through learning environment adjustments utilizing IoT sensors and cloud-based data management.

Description

Technology Field

[0001] This invention belongs to the field of combining educational technology, artificial intelligence (AI), and the Internet of Things (IoT). More specifically, it includes data collection and analysis used in existing online learning platforms or smart classroom systems, personalization utilizing AI, and indoor environment control technology through IoT sensors. Background Technology

[0002] With the advancement of digital technology, online and hybrid learning systems are rapidly spreading in the modern educational environment. However, most existing systems have limitations in that they provide the same content uniformly to all learners and fail to adequately reflect the characteristics and current state of individual learners. For instance, one student may require in-depth learning content due to high concentration, while another student may need simpler materials or summaries due to accumulated fatigue. Failure to effectively account for these differences poses a risk of significantly reduced learning efficiency and satisfaction.

[0003] Furthermore, as physical learning environments are often static or managed passively, various environmental factors within the classroom, such as lighting, temperature, and noise, frequently do not align with the optimal learning conditions for individual learners. For instance, if the classroom temperature is excessively high or low, learners' concentration may decrease, negatively impacting learning efficiency. Against this backdrop, there is a growing demand for systems capable of analyzing learners' biosignals and behavioral data in real time, and providing personalized content and environmental control tailored to those needs. The problem to be solved

[0004] The main challenges that this invention aims to solve can be broadly categorized into three areas. First, it seeks to effectively identify the real-time status of individual learners (e.g., concentration, comprehension, fatigue) and provide customized educational content tailored to them. Existing systems had limitations in providing the same content to all learners because they could not dynamically reflect the learners' states. Second, it aims to automatically optimize the physical learning environment so that environmental factors such as lighting, temperature, and noise within the classroom provide optimal learning conditions for the learners. This is intended to maximize learning efficiency and ensure a comfortable learning experience for the learners. Finally, to enhance learning motivation, the invention aims to form a learning community where learners can actively participate and collaborate by integrating gamification and social interaction elements.

[0005] For example, when a student consistently struggles with a specific topic and experiences increased fatigue, existing systems fail to reflect this immediately, which can negatively impact their learning. To address this issue, the present invention aims to significantly improve learning efficiency by analyzing the learner's state in real time, providing customized content and supplementary materials based on the results, and dynamically adjusting the learning environment. means of solving the problem

[0006] This invention solves the aforementioned challenges by integrating various sensors with AI and IoT technologies. First, by collecting learner behavioral data, biosignals, user input, and feedback in real time through a data collection module, learning patterns and current status can be closely identified. For example, a student's facial expressions and pulse can be measured via a webcam and a smartwatch, and learning concentration can be quantified by analyzing clickstream data.

[0007] Subsequently, the AI ​​analysis module processes the collected data in real time to calculate quantitative indicators such as the learner's preferred learning style, problem-solving performance, and fatigue levels. Based on this, the content recommendation and learning difficulty adjustment module automatically selects and provides more basic or advanced learning materials tailored to the student's understanding and learning tendencies. Additionally, the multimodal conversion module supports providing content in the optimal format among text, video, and audio, or converting it in real time, depending on network conditions or the learner's situation.

[0008] Furthermore, the learning environment optimization module utilizing IoT sensors continuously monitors various physical elements within the classroom, such as lighting, temperature, noise, and air quality, and automatically adjusts the environment through a central management server. In addition, the present invention enhances learning engagement by adding gamification and social interaction modules to provide motivational elements, such as reward points, badges, and leaderboards, based on learner performance, and by promoting feedback and collaboration among fellow learners. Effects of the invention

[0009] The effects of the present invention manifest in five main aspects. First, a personalized learning experience is provided, offering content tailored to each learner's current state and preferences, thereby significantly improving learning efficiency and achievement. For example, a student with high levels of fatigue can learn more effectively by receiving concise summary content and supplementary materials.

[0010] Second, the learning environment optimization feature automatically adjusts physical conditions within the classroom, such as lighting, temperature, and noise, maintaining a comfortable and focused learning environment. This significantly contributes to enhancing students' concentration and efficiency. Third, the integration of gamification and social interaction features motivates learners through a reward system and enhances their engagement through competition and collaboration with peers.

[0011] Furthermore, real-time subtitle generation and multilingual translation functions enable learners from various language backgrounds to easily understand lectures, allowing for effective adaptation to international educational environments. Finally, the real-time monitoring and feedback system via the integrated management server module facilitates immediate responses to issues, contributing to the stability and reliability of the overall learning environment. Specific details for implementing the invention

[0012] Hereinafter, preferred embodiments will be described with reference to the attached drawings to enable a more specific understanding of the present invention.

[0013] The present invention is capable of various modifications and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the present invention to specific embodiments, and it should be understood that it includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention.

[0014] Terms such as "first," "second," etc., may be used to describe various components, but said components should not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component. The term "and / or" includes a combination of a plurality of related described items or any of a plurality of related described items.

[0015] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

[0016] The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to specify the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0017] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.

[0018] Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the attached drawings. In order to facilitate an overall understanding of the present invention, the same reference numerals are used for identical components in the drawings, and redundant descriptions of identical components are omitted.

[0020] As an example according to the present invention, an AI-based customized educational content and learning environment optimization system comprises: (a) a data collection module that collects learning patterns, biosignals, user input, and feedback data in real time from a user terminal; (b) an AI analysis module based on machine learning or deep learning that processes the collected data in a streaming manner and calculates quantitative indicators including learning propensity, comprehension, and fatigue; (c) a content recommendation and difficulty adjustment module that recommends content and automatically adjusts the learning difficulty based on the quantitative indicators; (d) a multimodal conversion module that provides pre-produced multimodal content in an optimal form through real-time AI conversion according to the learner's current state; (e) an AI-based language processing module that automatically inserts multilingual subtitles and term explanations according to the learner's level through real-time speech recognition-based subtitle generation and neural network machine translation for the multimodal content provided by the multimodal conversion module; and (f) a learning environment optimization module that collects learning environment data using IoT sensors and controls the learning environment according to AI analysis results and predefined criteria. (g) a cloud-based integrated management server module that manages the data of the modules (a) to (f) through an encrypted API and processes control commands and system logs; (h) a system that additionally includes a gamification and social interaction module in addition to the integrated management server module, provides a reward system including points, badges, or leaderboards based on learner performance and participation data, and strengthens learning motivation through feedback and collaboration functions among peer learners.

[0022] The above data collection module collects learning-related data in real time from a user terminal, and the learning-related data includes various types of information such as learning patterns, biosignals, user input, and feedback.

[0023] The above data collection module periodically (e.g., every 5 minutes) calls the webcam, smartwatch, built-in sensor, and software logs to collect the user's learning patterns, biosignals, user input, and feedback data in real time, continuously monitors changes in the learner's fatigue and concentration based on this, and automatically provides supplementary materials or feedback when necessary.

[0024] Furthermore, the data collection module fuses data from various sources to enhance analysis accuracy, updates the status in real-time to prepare for unexpected situations, and provides real-time feedback to users. For example, it offers features such as immediately recommending relevant materials to users who provide feedback or resolving errors that occur during learning in real-time.

[0025] In addition, this module collects from real-time chat, comments, survey responses, etc. Text-based feedback It analyzes the learner's subjective learning experience and emotional state to evaluate them. For example, if a student provides feedback such as "This part is difficult," it can support learning by automatically providing supplementary explanatory materials or quizzes for that section. Real-time collected feedback is used to provide customized content that reflects the learner's preferences.

[0026] For example, the LMS REST API is called every 5 minutes to receive new lecture information and test results in JSON format, and duplicate data is removed using a data filtering algorithm. For instance, learning patterns can measure which content a user selects while watching a lecture, where they stop, and the playback time of the content.

[0027] Biosignals can collect heart rate, skin conductivity, body temperature, etc., measured by smartwatches, wearable devices, etc., in real time. For example, it can track the pattern of heart rate increase when a user feels fatigued.

[0028] User input tracks keyboard or mouse clicks, touchscreen events, or other user actions in real time. For example, frequently used keys can be recorded and utilized for preference analysis.

[0029] Feedback is collected and analyzed through surveys or rating systems that provide feedback to learners regarding the lecture.

[0031] The above AI analysis module processes collected data in a streaming manner and plays the role of calculating quantitative indicators such as learning propensity, comprehension, and fatigue.

[0032] The AI ​​analysis module processes real-time learner data to predict learning tendencies, comprehension, and fatigue, as well as behavior based on learning patterns, thereby providing a predictive model that optimizes future learning plans.

[0033] In addition, the above AI analysis module is Past training data By analyzing, learners will future Topics or areas likely to cause difficulties Predicts.

[0034] For example, if a student's error rate in a specific subject exceeds 40%, the system provides supplementary materials or additional questions on the topic in advance and adjusts the study plan to improve the learner's understanding. By predicting expected learning outcomes or progress based on the learner's performance, Automatically generate customized learning strategies does.

[0035] In addition, the AI ​​analysis module analyzes the learner's emotional state in real time based on facial images, heart rate, and blink frequency data. For example, if the heart rate increases by more than 20% or the blink frequency changes by more than 30% compared to normal, it determines that the learner is feeling anxious or stressed and provides a function to reduce the learning speed or recommend a break time.

[0036] The aforementioned learning tendency involves analyzing the user's past learning history and preferred content types to determine which method (video, text, quizzes, etc.) they prefer to learn.

[0037] The aforementioned comprehension is evaluated through the accuracy of problem-solving, reaction time, and error rate; for example, it tracks how quickly and accurately a student answers each time they solve a problem and calculates their comprehension based on this data.

[0038] The aforementioned fatigue is calculated based on biosignals (e.g., heart rate, blinking frequency) and user input patterns (e.g., slowdown, reduced click frequency, etc.), and, for example, the point at which the user feels fatigue can be identified through changes in heart rate or a decrease in click speed.

[0040] The above content recommendation and difficulty adjustment module recommends learning content and automatically adjusts the learning difficulty based on analyzed quantitative indicators.

[0041] To elaborate, the aforementioned content recommendation and difficulty adjustment module comprehensively analyzes the learner's past lecture history, quiz scores, assignment evaluation data, and real-time comprehension and fatigue indicators to classify and recommend learning content through recommendation algorithms (e.g., collaborative filtering and weighted ranking). Furthermore, if a user repeatedly records incorrect answers, the module automatically lowers the difficulty level to prioritize the provision of supplementary learning materials. For instance, if the accuracy rate is 70% or lower, or the average response time is 10 seconds or longer, content with reduced difficulty is provided.

[0042] For example, content recommendation involves suggesting suitable content based on a learner's learning style and comprehension. This could involve, for instance, analyzing subjects or concepts that a student finds difficult and automatically recommending supplementary learning materials.

[0043] For example, difficulty adjustment automatically adjusts the difficulty of the content after analyzing the learner's fatigue and comprehension. For instance, if a user struggles to keep up, materials with a lower difficulty level are provided; conversely, if learning progresses well, the difficulty can be increased to offer challenging content.

[0045] The aforementioned multimodal conversion module is a module that provides pre-produced multimodal content in an optimal form through real-time AI conversion tailored to the learner's current state. In other words, the multimodal conversion module provides pre-produced text, image, and video content in an optimal form (e.g., text summary from video or interactive quiz) through real-time AI conversion based on the learner's current concentration, fatigue level, and real-time interactive feedback. For example, if the learner's concentration drops below 30%, the video lecture is converted into text summary content and provided.

[0046] The above multimodal conversion module includes a function that automatically switches to the most effective learning method by analyzing the learner's current concentration and fatigue levels, as well as the distribution of attention during learning.

[0047] The above AI transformation modifies or adds learning materials in real time based on the learner's interactive feedback, providing a more personalized learning experience.

[0048] Here, pre-produced content refers to various forms of content, such as text, images, and videos, that are prepared in advance.

[0049] Furthermore, AI transformation refers to changing the form of content based on the learner's state (e.g., concentration, fatigue, learning style). For instance, this involves converting video lectures into text summaries or concept explanations into interactive quizzes, thereby providing optimal learning methods to enhance the learner's comprehension.

[0051] The above AI-based language processing module provides multilingual subtitles through real-time speech recognition and machine translation, and automatically inserts term explanations tailored to the learner's level.

[0052] In other words, the AI-based language processing module converts voice data within the learning material into text subtitles using a real-time speech recognition engine, translates them into multiple languages ​​using a neural network machine translation algorithm, and automatically inserts simple explanations or examples for difficult terms according to the learner's level and preferred language. For example, if the subtitles for an English lecture have a BLEU score of 0.8 or higher, they are provided as is, and if an error rate of 50% or higher is confirmed for difficult technical terms, a term explanation is automatically added.

[0053] In addition, by providing customized translations and interpretations tailored to the learner's preferred language and cultural background, it enhances user comprehension and optimizes learning efficiency in a global learning environment.

[0054] Here, real-time speech recognition-based subtitle generation refers to converting audio or video included in learning materials into text subtitles in real time. For example, subtitles matching a student's pronunciation can be generated in real time during a lecture.

[0055] In addition, multilingual subtitles through machine translation refer to translating speech-recognized subtitles into various languages ​​to provide content suitable for learners who use various languages.

[0056] In addition, glossaries automatically add explanations for terms that may be difficult to understand, tailored to the learner's level. For example, simple explanations or examples can be automatically inserted for difficult scientific terms.

[0058] The above-mentioned learning environment optimization module collects learning environment data using IoT sensors and controls the learning environment based on AI analysis results and predefined criteria.

[0059] In other words, the aforementioned learning environment optimization module utilizes IoT sensors that collect data such as illuminance, temperature, noise, humidity, air quality, CO2 levels, and indoor occupancy in real time, and controls heating and cooling, lighting, humidifiers, air purifiers, noise blockers, and security devices in multi-stage (e.g., 5-level lighting control, 2) according to AI analysis results and predefined criteria (e.g., temperature 22–24°C, CO2 1,000 ppm or less). Controlled by unit temperature control.

[0060] In particular, the learner's Concentration and Fatigue According to, Personalized learning environment It adjusts in real time. For example, the learner's If fatigue increases Lower the lighting brightness, or Automatically adjusts the temperature Therefore, it provides a comfortable learning environment. In addition, if the learner's concentration drops Adjust the noise level Optimize learning efficiency by doing so or providing break time notifications.

[0061] In addition, it provides a personalized learning space by continuously improving the environment control algorithm through real-time reflection of user feedback. For example, if the light sensor detects that the brightness in the classroom is less than 20%, it automatically adjusts the lighting to be brighter by one step out of five.

[0062] The above learning environment optimization module also collects user feedback on changes in the learning environment in real time and continuously improves environment optimization through this to provide a personalized learning space.

[0063] Here, IoT sensors collect environmental data in real time, such as illuminance, temperature, noise, humidity, CO2 levels, and indoor occupancy. For example, if the illuminance in a classroom is too low, the lighting can be automatically adjusted, or if the CO2 concentration is high, ventilation can be controlled.

[0064] Learning environment control refers to optimizing the environment based on the learner's fatigue or concentration level. For example, users with high fatigue levels may be automatically recommended a rest area or have their lighting adjusted.

[0066] The above-mentioned cloud-based integrated management server module is responsible for managing data collected from modules (a) through (f) in an integrated manner via an encrypted API, and for processing control commands and system logs.

[0067] additionally, Data Visualization and Dashboards By providing a module, administrators and instructors can access data such as the learning environment, user behavior, and system logs in real time. Visually check in graph and chart formats It is possible. For example, each learner's Attendance rate, learning progress, content error detection rate It allows for real-time monitoring of the back on the dashboard, enabling a quick response in the event of a problem.

[0068] In other words, the aforementioned cloud-based integrated management server module manages data collected from user terminals and each module through TLS / SSL-encrypted APIs, and adopts a distributed architecture to ensure data flow optimization and system scalability. Furthermore, all control commands and system logs are recorded in a blockchain-based logging system when necessary, enabling future auditing and verification of data integrity. For example, lighting and temperature control commands are automatically recorded along with student F's attendance information, facilitating data recovery and tracing even in the event of a system failure.

[0069] Here, the cloud-based integrated management server module adopts a distributed architecture to optimize data flow, enhance system scalability and resilience, and automatically provide countermeasures against server failure or data loss.

[0070] In addition, the cloud-based integrated server module also provides each user with the latest content and settings through real-time data synchronization, and offers automated monitoring and alert functions that enhance system stability.

[0071] For example, this involves managing all data in the cloud and efficiently integrating the data flow between modules; data collected from user terminals can be aggregated and analyzed on the cloud server to provide content necessary for learning.

[0072] In addition, cloud servers can ensure data security and traceability by processing all data through encrypted channels and managing system logs.

[0074] The aforementioned gamification and social interaction module provides a reward system including points, badges, and leaderboards based on learner performance and participation data, and plays a role in reinforcing learning motivation through feedback and collaboration functions among peer learners.

[0075] In other words, the aforementioned gamification and social interaction module analyzes learner performance and participation data in real time to award points, badges, and leaderboards upon achieving certain performance goals. It also strengthens learning motivation through collaborative functions such as Q&A, mentoring, and joint task execution among peer learners, and provides a reward system that simultaneously promotes competition and cooperation.

[0076] In addition, whenever a learner achieves a goal set for each stage, their performance can be measured through real-time diagnosis, and timely learning rewards can be provided along with immediate feedback.

[0077] For example, if a learner solves a specific quiz with an accuracy of 80% or higher, their performance is immediately recorded, points and badges are awarded through a learning reward system, and the difficulty of subsequent content is personalized to encourage more challenging learning. In this process, the learner's sense of accomplishment is enhanced while providing continuous motivation for learning.

[0078] For example, students who achieve an accuracy rate of 80% or higher on a specific topic earn additional points and badges, and their rankings on the leaderboard rise, encouraging competition among learners.

[0079] The above gamification and social interaction modules can also encourage learners to participate voluntarily through a system that induces competition and collaboration among peer learners based on learner performance.

[0080] In other words, it provides a reward system that induces competition and motivation by awarding points or badges when learners achieve certain results and displaying their rankings on a leaderboard, and facilitates social interaction by enabling users to exchange information and cooperate with other learners through feedback and collaboration among peers. For example, real-time feedback features can promote Q&A among students or encourage collaborative task completion.

[0081] As an example according to the present invention, a system may be disclosed in which the data collection module collects data using a webcam, a smartwatch, an embedded sensor, or a software log, wherein the learning pattern is collected as a clickstream, access time, or problem-solving behavior, the biosignal is collected as emotion recognition by facial image analysis, PPG-based pulse, or an eye-tracking camera, the user input is collected as the frequency and interval of keyboard, mouse, or touch events, and the feedback is collected through survey and rating functions, respectively.

[0082] Regarding the collection of data using webcams, smartwatches, built-in sensors, and software logs, webcams can capture the user's face or gestures in real time and analyze them to track the learner's emotions, fatigue, and concentration. For example, facial recognition can be used to analyze whether the user is tired or focused.

[0083] In addition, smartwatches or wearable devices monitor the user's biosignals (e.g., heart rate, body temperature, activity level, etc.) in real time. This allows for the identification of the learner's physical condition and can be utilized to manage fatigue or stress during learning.

[0084] Furthermore, user behavior patterns can be tracked through sensors embedded in smartphones or other devices. For example, accelerometers or gyroscopes can be used to detect and analyze a user's posture or movements during learning. In addition, users' digital behavior is recorded. For instance, various actions such as clicks, page transitions, and button clicks that occur when a user interacts with a website or application are logged, and this log data allows for the analysis of user behavior on the interface.

[0085] To collect learning patterns, clickstreams, access times, problem-solving behaviors, etc., can be identified and collected.

[0086] For example, a clickstream tracks which pages a user visits and which content they click when using a site or application, allowing for the identification of their interests or preferences.

[0087] In addition, it tracks how long users spend on each piece of content based on their access time. For instance, if a user spends a significant amount of time on a specific lecture or material, it reveals the learner's interest in or need for that content. Furthermore, it tracks user behavior when solving problems. For instance, by analyzing reaction times, error rates, and accuracy rates while solving problems, the user's level of comprehension can be evaluated.

[0088] Next, for the collection of biosignals, emotion recognition based on facial image analysis, PPG-based pulse, and eye-tracking cameras can be utilized.

[0089] First, as emotion recognition based on facial image analysis, the user's facial expressions are analyzed to track their emotional state (e.g., happiness, stress, boredom, fatigue, etc.). For instance, if a user feels bored or stressed during learning, this can be detected from their facial expressions.

[0090] In addition, PPG-based pulse: The user's heart rate is measured using Photoplethysmography (PPG) technology. Heart rate is an important indicator of a user's fatigue or stress levels; for example, a rapid increase in heart rate can suggest tension or stress related to learning.

[0091] Eye-tracking cameras analyze which parts a user is paying attention to through eye tracking. This allows for determining the degree to which a user is focusing on lectures or materials during learning.

[0092] Next, user input collection involves gathering the frequency and intervals of keyboard, mouse, or touch events. In other words, it tracks the speed and frequency of user input on the keyboard and other devices. For instance, frequent mouse clicks during learning can indicate potential issues with comprehension or concentration. Additionally, touch events are primarily used in mobile environments and are useful for evaluating a learner's reaction speed or skill level.

[0093] Next, to collect feedback, subjective evaluations of the learning experience can be gathered through surveys or rating functions provided to learners. For example, students can be provided with a survey after a lecture to evaluate their understanding, usefulness, and the instructor's delivery.

[0095] As an example according to the present invention, a system is disclosed in which the AI ​​analysis module processes real-time streaming and batch method data, classifies learning tendencies based on past learning records and content preferences, quantifies comprehension as problem-solving accuracy, reaction time, or error rate, and calculates fatigue by combining heart rate variability, eye blinking frequency, and user input pattern degradation.

[0096] Regarding the processing of real-time streaming and batch data, real-time streaming data involves processing data that occurs immediately; by analyzing data generated during the learner's content consumption process in real time, it is possible to provide immediate feedback. For example, click streams, response times, or error rates generated while a student solves problems can be analyzed in real time. In the case of batch data, the batch method involves collecting data at regular intervals or specific times for collective processing. For instance, a method can be used to aggregate learning records or content preferences collected over a day and analyze them after a set period.

[0097] Next, regarding the analysis of learning tendencies, these tendencies are classified based on users' past learning records and content preferences. This is utilized to identify what content users focused on in the past and what types of learning they prefer, thereby providing customized content tailored to those needs. For example, if Student A prefers video lectures and achieved high scores on quizzes, the AI ​​can analyze that this student has a strong inclination toward theory-based content. The AI ​​analysis module estimates the student's preferences based on data obtained from past learning records and incorporates this into the recommendation system to provide personalized content.

[0098] Next, regarding the quantification of comprehension, it is measured through problem-solving accuracy, reaction time, or the error rate. This method evaluates how well a learner understands the material based on their successful outcomes and reaction speed when solving problems. Problem-solving accuracy assesses comprehension through the correct answer rate of the questions solved; for instance, if a student has a high correct answer rate, it can be judged that they have demonstrated a high level of understanding of the content. Reaction time—the time taken to solve a problem—is also an important indicator; taking too long may suggest low comprehension, while solving it quickly may indicate high comprehension. Furthermore, the error rate, which represents the percentage of incorrect answers, allows for the identification of areas where the learner struggles. For instance, repeatedly getting questions wrong can be interpreted as a signal that the student lacks understanding in that specific area.

[0099] Next, regarding the calculation of fatigue, it is determined by combining various physiological and behavioral data, including heart rate variability, blink frequency, and a decline in user input patterns. Here, heart rate can be utilized as an indicator of a learner's stress level or fatigue; if sudden changes in heart rate are detected, the learner may be experiencing fatigue or stress. Additionally, blink frequency can increase or decrease when a person is fatigued, and tracking this allows for the identification of whether the learner is feeling tired. Furthermore, changes in input patterns when using a keyboard or mouse can be a sign of fatigue; for example, if input speed slows down or repetitive errors increase, it is highly likely that the learner is experiencing a decline in concentration and fatigue.

[0101] As an example according to the present invention, a system is disclosed in which the content recommendation / difficulty adjustment module and the multimodal conversion module analyze the user's past lecture history, quiz scores, and assignment evaluation data to classify content into problem-solving videos, concept explanation texts, interactive quizzes, etc., readjust recommendation rankings in real time through an AI algorithm, and provide supplementary learning materials first when repeated incorrect answers occur.

[0102] Regarding the analysis of past lecture history, quiz scores, and assignment evaluation data, user lecture history refers to a record of the content and format of previously taken courses; this information serves as important data for understanding the user's preferred learning style, teaching methods, and topics. Additionally, quiz scores provide information regarding the accuracy of the questions answered, which is used to assess the learner's comprehension and learning progress. Furthermore, assignment evaluation data allows for the assessment of a learner's learning outcomes and problem-solving abilities. If an assignment is performed well, it can be determined that the learner possesses a high level of understanding of the relevant topic, which can then be utilized for content recommendations and difficulty level adjustments.

[0103] Next, regarding content classification, problem-solving videos provide visual aids for users to solve problems and are primarily used to enhance learning effectiveness through the problem-solving process or visual explanations. Conversely, concept explanation text content focuses on theoretical background or conceptual explanations and is useful for re-explaining concepts that users have not fully grasped. Additionally, interactive quizzes are quiz-style content designed for active student participation, serving as a method to immediately assess how well learners understand given concepts. This content is automatically classified based on the user's learning level or needs. For example, problem-solving videos may be recommended when a student is struggling to solve problems independently, while concept explanation texts are provided when there is a need to review theory.

[0104] Next, regarding the real-time readjustment of recommendation rankings through AI algorithms, the AI ​​algorithm analyzes the user's learning patterns, successful learning records, and incorrect answer patterns to predict which content is most suitable for the student. The ranking of recommended content is adjusted in real-time based on the learner's response; for example, if a student struggles with a specific type of problem, the AI ​​can prioritize presenting that type of problem or provide supplementary materials.

[0105] Furthermore, regarding the provision of supplementary learning materials in the event of repeated errors, if a learner repeatedly makes mistakes on a problem or records continuous errors, the AI ​​can recognize this and prioritize the provision of supplementary learning materials. These supplementary materials focus on reinforcing concepts or problem types that the user has not understood, and automatically provide appropriate materials to enhance learning effectiveness.

[0107] As an example according to the present invention, the integrated management server module collects learning environment data including illuminance, temperature, noise, humidity, air quality, CO2, or the number of indoor personnel from IoT sensors, and controls a learning environment control device including a heater, lighting, a humidifier, an air purifier, a noise blocker, and a security device in multiple stages based on the collected data; the integrated management server module processes data through distributed computing and records all control commands and system logs in an encrypted channel and, if necessary, in a blockchain-based log system; furthermore, the system is characterized by including a function to manage a real-time feedback and reward system among learners in conjunction with a gamification and social interaction module. A system is disclosed.

[0108] In this context, IoT sensors control the learning environment by detecting various physical elements in real time. The collected data includes, but is not limited to, environmental data such as illuminance, temperature, noise, humidity, air quality, CO2, and the number of people inside. For example, illuminance sensors measure the brightness of the lighting in the learning space to maintain an environment where students can study most efficiently; temperature sensors help maintain an appropriate temperature; noise sensors detect noise levels to provide an environment conducive to concentration; and humidity and air quality sensors manage the learning environment to ensure it is comfortable and suitable for students' health. Additionally, carbon dioxide sensors measure indoor air quality to ensure students receive an optimal oxygen supply, and the number of people inside can detect the density of the learning space to adjust the space appropriately.

[0109] Next, as a learning environment control device, it controls various control devices in a multi-stage manner based on the data collected above. Here, heating and cooling units automatically adjust the indoor temperature according to the ambient temperature, and lighting units automatically adjust brightness according to illuminance to ensure students can study in a comfortable lighting environment. Humidifiers and air purifiers detect and regulate air quality and humidity to provide a pleasant environment, while noise blockers activate noise reduction functions based on noise levels to enhance study concentration. Additionally, a security system is included that can automatically adjust based on the number of detected personnel for indoor security.

[0110] Next, distributed computing refers to a method of processing data by combining cloud and edge computing. This approach enables the efficient and rapid processing of data collected in real time. Edge computing provides fast responses by immediately processing data at locations close to the learning environment, while cloud computing can analyze and store large-scale data and perform more complex analysis and data processing tasks.

[0111] Next, for data security and record-keeping, all control commands and system logs are transmitted via encrypted channels. This measure is intended to guarantee the safety and integrity of the data. Additionally, all control commands (e.g., temperature control, lighting control, etc.) and system logs are recorded in a blockchain-based logging system. This prevents data alteration or forgery and ensures that subsequent auditing and tracing are possible.

[0112] Next, the gamification and social interaction modules provide functions that increase learner engagement and strengthen learning motivation.

[0113] The aforementioned gamification and social interaction module promotes learning motivation through a mentoring system and group activities, enabling learners to support and cooperate with one another even in a competitive environment.

[0114] Furthermore, it provides learners with the opportunity to receive real-time feedback on their performance and improve accordingly. In addition, it strengthens motivation by offering points, badges, and leaderboards based on learners' performance, and promotes learning motivation through interaction by encouraging mutual learning via collaboration and feedback. This enables the creation of a collaborative learning environment.

[0115] The device described above may be implemented as a hardware component, a software component, and / or a combination of a hardware component and a software component. For example, the device and components described in the embodiments may be implemented using one or more general-purpose or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing unit may execute an operating system (OS) and one or more software applications executed on said operating system. Additionally, the processing unit may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing unit may be described as being used as a single unit, but those skilled in the art will understand that the processing unit may include a plurality of processing elements and / or a plurality of types of processing elements. For example, the processing unit may include multiple processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are also possible.

[0116] Software may include computer programs, code, instructions, or combinations of one or more of these, and may configure a processing unit to operate as desired or instruct the processing unit independently or collectively. Software and / or data may be embodied in any type of machine, component, physical device, virtual equipment, computer storage medium, or device so as to be interpreted by the processing unit or to provide instructions or data to the processing unit. Software may be distributed over networked computer systems and may be stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

[0117] The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and may be recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., either alone or in combination. The program instructions recorded on the medium may be those specifically designed and configured for the embodiment, or they may be those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

[0118] Although the embodiments have been described above with reference to limited examples and drawings, those skilled in the art can make various modifications and variations from the description above. For example, suitable results may be achieved even if the described techniques are performed in a different order than described, and / or if the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from described, or replaced or substituted by other components or equivalents.

[0119] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims set forth below.

Claims

Claim 1 As an AI-based customized educational content and learning environment optimization system, comprising: (a) a data collection module that collects learning patterns, biosignals, user input, and feedback data in real time from a user terminal; (b) an analysis module that processes the collected data in a streaming manner and calculates quantitative indicators including learning propensity, comprehension, and fatigue using a machine learning or deep learning-based AI analysis module; (c) a content recommendation and difficulty adjustment module that recommends content and automatically adjusts learning difficulty based on the quantitative indicators; (d) a multimodal conversion module that provides pre-produced multimodal content in an optimal form through real-time AI conversion according to the learner's current state; (e) an AI-based language processing module that automatically inserts multilingual subtitles and term explanations according to the learner's level through real-time speech recognition-based subtitle generation and neural network machine translation for the multimodal content provided by the multimodal conversion module; (f) a learning environment optimization module that collects learning environment data using IoT sensors and controls the learning environment according to AI analysis results and predefined criteria; (g) an integrated management of the data of modules (a) through (f) via an encrypted API, and control commands and a system A system characterized by a cloud-based integrated management server module for processing logs; (h) additionally including a gamification and social interaction module in addition to the integrated management server module, providing a reward system including points, badges, or leaderboards based on learner performance and participation data, and reinforcing learning motivation through feedback and collaboration functions among fellow learners. Claim 2 The system according to claim 1, wherein the data collection module collects data using a webcam, smartwatch, built-in sensor, or software log, wherein the learning pattern is collected as a clickstream, connection time, or problem-solving behavior; the biosignal is collected as emotion recognition by facial image analysis, PPG-based pulse, or eye-tracking camera; the user input is collected as the frequency and interval of keyboard, mouse, or touch events; and the feedback is collected through survey and rating functions, respectively. Claim 3 The system according to claim 1, wherein the AI ​​analysis module processes real-time streaming and batch method data, classifies learning tendencies based on past learning records and content preferences, quantifies comprehension as problem-solving accuracy, reaction time, or error rate, and calculates fatigue by combining heart rate variability, blink frequency, and degradation of user input patterns. Claim 4 In claim 1, the content recommendation / difficulty adjustment module and the multimodal conversion module analyze the user's past lecture history, quiz scores, and assignment evaluation data to classify the content into problem-solving videos, concept explanation texts, interactive quizzes, etc., readjust the recommendation ranking in real time through an AI algorithm, and provide supplementary learning materials first when repeated incorrect answers occur. Claim 5 The system according to claim 1, wherein the integrated management server module collects learning environment data including illuminance, temperature, noise, humidity, air quality, CO2, or the number of indoor people from IoT sensors, and controls a learning environment control device including a heater, lighting, humidifier, air purifier, noise blocker, and security device in multiple stages based on the collected data, wherein the integrated management server module processes data through distributed computing and records all control commands and system logs in an encrypted channel and, if necessary, in a blockchain-based log system, and further, includes a function to manage a real-time feedback and reward system among learners in conjunction with a gamification and social interaction module.