A TRAINING SYSTEM AND METHOD WITH A PERSONALIZED SPACED REPETITION ALGORITHM AND DYNAMIC LEARNING ANALYTICS.

Abstract

The invention relates to a software-based training system and method with a personalized spaced repetition algorithm and dynamic learning analytics that analyzes users' learning performance in real-time, generates AI-assisted content, and dynamically optimizes the display times of flashcards using spaced repetition algorithms.

Description

PERSONALIZED SPACED REPETITION ALGORITHM AND AN EDUCATIONAL SYSTEM WITH DYNAMIC LEARNING ANALYTICS AND METHOD Technical Area: The invention uses artificial intelligence to analyze users' learning performance in real time. It produces supported content and uses spaced repetition algorithms to gather information. a software-based system that dynamically optimizes the display times of cards training personalized spaced repetition algorithm and dynamic learning analytics It is related to an education system and method. The system involves large-scale education. data processing, content standardization, personalized work plans data creation and continuous optimization of learning processes It utilizes analytics and machine learning techniques. The invention will benefit medical students, general practitioners, and residency trainees, particularly in the field of medicine. Preparation for knowledge-intensive exams in the medical field, including the Medical Specialization Exam (TUS). a computer-based training method developed to support processes It is a system. The system includes flashcards, question banks, digital educational content, and Personalized learning by managing user performance data in an integrated way. It offers an experience. As a result, the invention enables AI-powered content creation, Personalized spaced repetition algorithm, dynamic learning analytics, multi-layered data model and gamified learning infrastructure within a single system. It integrates the content. This integrated structure increases learning efficiency while enhancing the content. in terms of production, information retention and user-specific work planning, the previous It provides a significant technical advantage compared to other applications. 1 State of the Art: In the current state of the technology, various online education platforms exist. However, None of the platforms in question use active recall and spaced repetition methods for memory retention. It does not offer an educational model using flashcards. Applications used in the known state of the art include classic digital question banks, Online TUS training platforms, electronic grading systems, and spaced repetition. These are (spaced repetition) software programs. These systems facilitate information repetition and problem solving. along with providing support, an integrated and personalized approach to TUS preparation. It is unable to provide a learning infrastructure. The technical shortcomings identified in similar applications are listed below: 1. Individualized revision algorithm in classic question banks. It is not available. ○ Questions are usually either in a fixed subject order or by the user. They are presented in the form of specified lists. ○ The system monitors the user's forgetting curve or learning performance. by taking into account the repetition times, dynamically optimizing them. is unable to do so. 2. Spaced repetition exercises do not support the content structure specific to the TUS exam. ○ Although general-purpose systems can schedule card repetition, the TUS curriculum, many factors such as subject-subject relationships, exam year, question type, and clinical significance It is unable to evaluate layered training data together. 3. The content creation process is largely done manually. ○ Information cards for printed lecture notes or PDF resources The conversion is done individually by the user. ○ This situation leads to significant time loss, formatting discrepancies, and content issues. This leads to a breakdown in standardization. 4. There is no AI-powered content standardization. 2 ○ The cards created are in different formats by different users. data integrity and content quality as it can be prepared It cannot be protected. 5. Work priorities cannot be determined dynamically. ○ In current systems, the repeat order is usually only based on previous answers. It is created according to... ○ Weighting of past exam questions, importance of topics, clinical priority, and exam preparation. Parameters such as remaining time cannot be evaluated together. 6. Learning analytics has limitations. ○ User's long-term learning curve, forgetting rate, subject-based Its success and retention of information are not analyzed in detail. Therefore, the system is unable to optimize its own performance. 7. The content and the learning algorithm operate independently of each other. ○ Content management and repetition algorithms differ in most applications. They are designed as modules. ○ This situation involves bringing learning data back into the content production process. It prevents him from getting proper nutrition. 8. Gamification and the learning algorithm are not integrated. ○ Features include competition, scoring, or multiplayer gameplay. On these platforms, these mechanisms are usually only for motivation. It is used for this purpose. ○ Real-time data sharing with a learning algorithm Since it hasn't been done, it doesn't contribute to rescheduling. 9. The user's work schedule is not automatically optimized. ○ Daily work schedules are mostly manually created by the user. is being created. ○ The system includes target exam date, daily study time, and individual learning. It cannot automatically create a plan by considering the speed together. 10. Integrated management of large-scale educational data is limited. ○ Numerous flashcards, topic relationships, labels, and user performance features. data that can be effectively managed within the same data model These systems are not widespread. 3 Consequently, the current solutions used in the known state of the technique are: question banks, Specifically, as a digital lesson platform or a general-purpose spaced repetition application. It meets the needs, but AI-powered content creation is personalized. FSRS-based rescheduling, TUS-specific multi-layered data model, learning analytics and gamified learning infrastructure within a single integrated system. He is unable to bring them together. Purpose of the Invention: The invention involves AI-powered content generation and a personalized spaced repetition algorithm. Dynamic learning analytics, multi-layered data modeling, and gamified learning. by integrating its infrastructure within a single system, it increases learning efficiency. to increase content creation, information retention, and user-specific work planning. It aims to... In addition, the invention includes flashcards, question banks, digital educational content, and Personalized learning by managing user performance data in an integrated way. It offers an experience. In conclusion, the invention enables AI-powered content generation and personalized, intermittent streaming. repetition algorithm, dynamic learning analytics, multilayer data model and by integrating gamified educational infrastructure into a single system It differs technically from digital TUS preparation platforms. This integrated The structure enhances learning efficiency while improving content production, information retention, and user experience. Significant technical superiority over previous applications in terms of customized work planning. It provides. Description of the Invention: The invention is a dynamic system developed for medical education and preparation for specialty exams. It is a learning management system that includes existing digital question banks and traditional spaced repetition. 4 It solves various technical problems that existing applications cannot solve. The invention... The technical advantages it provides are listed below: 1. It uses a dynamic learning model. ○ The system takes into account the feedback the user gives to each card (correct-incorrect, By analyzing each information card (response time, confidence level, etc.). It calculates the repetition time individually. ○ Unlike fixed-period repetition systems, for each card It creates an independent learning model. 2. FSRS (Free Spaced Repetition Scheduler) The training data of the advanced spaced repetition algorithm based on (timer) adaptation ○ Card display intervals are determined based on the user's past performance. It is optimized using the obtained parameters. ○ This results in higher information retention within the same study period. ○ FSRS is the process of cognitive learning and the retrieval of information retained in memory. optimization is the most efficient result of scientific research. It is an algorithm that enables the generation of a series of randomized double-blind results. Compilation of the outcomes of placebo-controlled scientific studies The effectiveness of the algorithm that created it has been scientifically proven. 3. Content creation and standardization with artificial intelligence. ○ Printed lecture notes, PDF documents, or training materials automatically analyzed and standardized into information cards It is transformed. ○ The generated cards follow a defined template, label structure, and content. It is prepared in accordance with the rules. ○ This process significantly reduces the time required for manual card preparation. 4. Breaking down information into atomic learning units. ○ Artificial intelligence can break down long texts into smaller pieces to reduce cognitive load. It divides learning into units. ○ Learning efficiency is increased because each card contains a single learning objective. 5. Exam-focused content weighting ○ The cards are not only based on learning success; ■ topic, ■ subtopic, ■ Frequency of appearance in past exams, ■ question type, ■ Clinical significance It is evaluated along with parameters such as these. ○ This allows work priorities to be dynamically rearranged. 6. Multi-layered labeling architecture ○ Each flashcard includes information on the topic, subtopic, course, exam year, exam period, and question type. It can be indexed by learning level and user-defined tags. ○ This structure enables fast filtering and personalized processing in large datasets. It enables the creation of a work plan. 7. Real-time learning analytics ○ The system continuously analyzes user performance to prevent forgetting. It calculates the curves. ○ Learning progress can be monitored graphically and statistically. 8. Automatic generation of a personalized study plan. ○ The system; ■ remaining time, ■ Target exam date, ■ daily work capacity, ■ learning speed by taking parameters such as these into account, it automatically performs the daily repetition schedule. It creates. 9. AI-powered quality control mechanism ○ The generated flashcards lack content consistency, repetitive phrases, and missing information. It is automatically analyzed in terms of information and format standards. ○ This will reduce the need for manual content quality review. is increased. 10. Integration of gamified learning infrastructure with spaced repetition system. ○ Competitive work modes, user rankings, and multiplayer Learning scenarios are directly integrated with the spaced repetition algorithm. It works. 6 ○ This way, user motivation is increased while preserving educational effectiveness. 11. Continuously self-optimizing learning system ○ Algorithm based on long-term user performance analysis The parameters are updated. ○ The system adapts to the user's learning behavior over time. This produces more accurate repetition times. 12. Scalable digital education infrastructure ○ The same system handles millions of information cards and numerous users. with data structures and indexing mechanisms that can manage it It has been designed. ○ This allows for low-volume processing of high-volume educational content. Work can be carried out with a delay. In conclusion, the invention enables AI-powered content generation and personalized, intermittent streaming. repetition algorithm, dynamic learning analytics, multilayer data model and by integrating gamified educational infrastructure into a single system It differs technically from digital TUS preparation platforms. This integrated The structure enhances learning efficiency while improving content production, information retention, and user experience. Significant technical superiority over previous applications in terms of customized work planning. It provides. How the Learning Algorithm Works: In the system, each flashcard is kept as an independent learning object. For each card; ● learning situation, ● number of repetitions, ● the possibility of forgetting, ● response time, ● User rating, ● last repeat date, ● next repeat date They are stored separately. 7 After the card is answered, the system; 1. User's answer, The second card shows the previous learning history. The difficulty level of the 3rd card, 4. user performance By evaluating them together, they calculate the new repetition time. This calculation creates a different repetition interval for each card. Thus... Information that is easy to learn occurs less frequently, while information that is difficult to learn occurs more frequently. It is shown. Technical description of the algorithm: FSRS (Free Spaced Repetition Scheduler) is based on the traditional Anki algorithm SM- Replacing model 2, the cognitive models of human memory (Sudden Forgetting and Memory) A modern and open-source interval recurrence based on (scatter curves) It is an algorithm. FSRS is a memory model that mathematically processes memory using three fundamental components: DSR. It is framed within the (Difficulty, Stability, Retrievability) framework: 1. D (Difficulty): How difficult it is to understand and remember the content of the card. This shows how difficult it is (ranging from 1 to 10). 2. S (Stability): How long information can remain in memory, i.e. It expresses its resistance to deterioration in days. As stability increases, it repeats. The intervals lengthen. 3. R ​​(Retrievability - Retention Rate): The user's immediate interaction with the card. It expresses the probability (between 0 and 1) of correctly remembering it when encountered. 1. Basic Formulas and Mathematical Background A. Retrievability Rate (R) 8 In the FSRS model, R is the relationship between elapsed time (t, in days) and stability (S). The calculation is done using the Power Law of Forgetting formula: R(t) = (1 + (19 / 81) * (t / S)) ^ (-0.5) Here, t is the time elapsed since the last day. By definition, R is the recall rate when t = S (elapsed time equals stability). R(S) is calibrated to be 0.90 (90%). Therefore, FSRS is the default. the user's target recall rate will be 90% (Desired Retention = 0.90) It optimizes. B. New Interval Calculation The next recurrence date of a card, target retrievability (R_target), and current Based on the determination (S), it is derived as follows: I(R_target) = (S / (19 / 81)) * ((R_target ^ -2) - 1) The user can set the target recall rate in the settings (e.g., 92% instead of 90%). When you increase it, the intervals shorten according to the formula; when you decrease it, the intervals lengthen. 2. User Feedback and Weights FSRS offers 17 personalized options, optimized in the background (machine learning). It has weight parameters (w0 to w16). The user provided 4 different responses. (Again, Hard, Good, Easy) These parameters filter the card to select D and S. It updates its values. A. Assigning Initial Values ​​(New Cards) When a card is worked on for the first time (initial rating), the starting difficulty (D0) and The initial stability (S0) is assigned using the following formulas: Initial Difficulty (D0): Based on the first key press by the user (Again=1, Hard=2, (Good=3, Easy=4) is calculated using parameters w0 and w3: D0(G) = w0 - (G - 3) * w1 9 (The resulting value is limited to between 1 and 10). Initial Stability (S0): Directly from the first 4 weights (w0, depending on the initial response (G), (w1, w2, w3) one is assigned: S0(G) = w_(G-1) So if a card seen for the first time is labeled "Easy," then its stability is similar to a card labeled "Again." It is initially set much higher than expected. B. Updating Existing Cards (Review Cards) If the card has been worked on before, the R-value at the time of review will be adjusted to the current difficulty. (D) and updates are made according to the given note (G). 1. Difficulty Update (D_new) The difficulty level is slightly higher in each review compared to the previous difficulty (D) and the given answer. shifted (via mean reversion mechanism): Intermediate change of D: ΔD = -w5 * (G - 3) New challenge: D_new = D - w5 * (G - 3) However, this value is not taken directly; it is used with the initial value (D0) to maintain system equilibrium. It is weighted and averaged (along with the w4 parameter). Outside the limits of 1 and 10. Its exit is prevented. 2. Updating Stability (S_new) FSRS's strongest feature is its increase in stability (S_Inc - Stability Increase) compared to the card's current state. It is dynamic scaling depending on the situation: Successful Recall (Hard, Good, Easy): If the user remembers the card, the stability never drops (S_Inc >= 1). Rate of increase. (S_Inc), current stability (S), difficulty (D), recall rate at pitch (R), and It depends on the rating: S_Inc(D, S, R, G) = 1 + Factors * (w parameters) Important Logical Rules: The stability increase of difficult cards is slow: the higher the D, the smaller the S_Inc. Stability saturation: The higher the S value, the more difficult it is to increase new stability. (S reaches saturation). The reward on the verge of being forgotten: The lower the R (i.e., the closer you are to forgetting the card). (if you remember), it's such a huge reinforcer for the brain; hence S_Inc is so important. It will be big. Parameters w15 and w16 represent the coefficients of "Easy" and "Hard" responses at this stage. It determines the factors. Failure / Forgetting (Again): If the user presses the "Again" button, the memory chain is broken (lapse). In this case Stability decreases (S_fail). S_fail = w11 * (D ^ -w12) * ((S + 1) ^ w13) * ((1 - R) ^ w14) Critical Security Rule: The newly calculated failure stability is higher than the card's previous performance. It cannot be greater than the determination of a forgotten card (min(S_fail, S)). That is, the determination of a forgotten card. It will never be written higher than before. 3. FSRS Optimizer (Weight Optimization) FSRS's 17 parameters (w0 to w16) come with default universal values. However, each user's memory structure and the type of material they memorize (language, medicine, coding) etc.) are different. 11 How it works: The user has a sufficient number of Anki members (usually a minimum of 1000+) When you access the review history, press the "Optimize" button to optimize the FSRS archive. It scans the data. Loss Function: The algorithm calculates the loss function for each card from the past. Treating "successful / unsuccessful" results as a binary classification. It takes this and tries to minimize the Log-Loss (Binary Cross-Entropy) metric. Gradient Descent / Optimization: Minimizing errors in historical data. A suitable set of 17 coefficients can be optimized using mathematical optimization methods, according to the user's own... It is recalculated and processed into the system according to brain performance. AI-Powered Content Creation: The artificial intelligence module in the system; ● PDF lecture notes, ● written educational materials, ● question banks, ● explanatory texts It analyzes and converts data into standard information cards. The cards created; ● single learning objective, ● standard format, ● topic tag, ● subtopic tag, ● course tag, ● description area It is automatically configured. This process significantly reduces the time required for manual card preparation. Gamified Learning Mechanism: 12 The Arena module differs from classic question-solving systems in that it focuses on learning. It works directly in conjunction with the algorithm. The user's; ● Flashcard success rate, ● response speed, ● long-term performance, ● ELO rating They are stored in the same database. This data is used to perform competitor matching and the user A competitive environment appropriate to the skill level is created. Elo rating is determined by the scores of the two players. a calculator that takes into account the relative skill levels of the two teams and the expected match outcome. It is a rating system. Technical Contribution of the Invention: As a result of the modules described above working together, the system; ● creating a completely personalized recurring plan for the user, ● dynamically sorting the flashcards, ● Optimizing repetition times according to the forgetting curve, ● Automating content production with the help of artificial intelligence, ● It increases user motivation through a gamified competition system, ● Improving system performance by continuously updating learning analytics It improves. Thanks to this integrated structure, the invention is not just a question bank or spaced repetition. Not an application; AI-powered adaptive learning, personalized repetition. combining planning and online competition mechanisms on a single technical platform. It operates as an integrated digital education system. The invention facilitates the learning process by continuously analyzing the user's level of knowledge. as a personalized, AI-powered adaptive digital education system The system is working on content creation, flashcard management, and personalized revisions. planning, performance analysis and competition modules working together It is based on. 13 The process consists of the following steps: 1. Creating a user account: After the user registers with the system target exam, study time, and the subject or topic they want to learn. The system selects the topics. It creates a personal learning profile for the user. It starts storing all performance data under this profile. 2. Preparation of training content: Flashcards within the system; content prepared by experts or AI-powered content It consists of standard data cards generated by the production module. income. Each information card; ● topic, ● subtopic, ● lesson, ● label, ● explanation, ● learning parameters It is recorded in the database along with it. 3. Creating the daily work list: At the start of each session, the system; ● cards the user has worked with previously, ● the possibility of being forgotten, ● cards that are due again, ● flashcards for learning new things It automatically generates a daily work schedule by analyzing the data. Thus... The user only sees the information cards they need to work with that day. 4. Displaying information cards: The system displays information cards to the user one by one. It displays. Initially, only the question or missing information is displayed. User After forming the answer in your mind, use the "Show Answer" command. It displays the correct answer and explanation. 5. Obtaining user feedback: After the answer is displayed, the user card; ● Again, 14 ● Difficult, ● Good, ● Easy It scores using one of the evaluation options as follows. This evaluation, feedback indicating the extent to which the user has learned the card It is recorded in the system. 6. Calculating the repetition time: The system uses the feedback received from the user, along with the card's previous learning history and performance data By evaluating the relevant information card, it automatically determines the next display time. They calculate it as follows. Flashcards that are easy to learn are not repeated even after a longer period of time. When the cards are shown, the ones that are harder to learn are reviewed at shorter intervals. It is added to the list. This process is performed independently for each information card. 7. Updating the performance analysis: After each work session, the system; ● the correct learning rate, ● repeat success rate, ● subject-based performance, ● working time, ● daily progress It updates the user profile by calculating the data. This data is then used for further work. It is used in the creation of plans. 8. How Arena (competitive) mode works: The user can switch to Arena mode if they wish. The user participates in the online competition session. The system calculates the user's ELO rating. Taking this into account, it automatically matches opponents of similar skill levels. Each The same information cards are presented to the player within certain rules. The match finally; ● number of correct answers, ● response time, ● success rate ELO ratings are updated based on this evaluation, and the user's rank is readjusted. It is calculated. In this way, the competition system provides an additional support for the learning process. It functions as a motivational mechanism. Example use case: For example, the user starts their daily study session by selecting the "Biochemistry" course. The system analyzes the user's past work data to identify when it's time to repeat the process. It identifies the new flashcards to be learned. After the user answers the first card... The system displays the response and rates the card as "Good". The system then moves on to the next card associated with this card. When automatically scheduling the repeat date to a later date, select "Repeat". The marked cards are added back to the worklist later that same day. At the end of the day... The user's performance statistics are updated, along with the next day's work schedule. It is automatically regenerated. The difficulty the user assigns to the card in remembering. If the feedback is about how difficult the card is to remember, then the card will be used by the user in the future. It is presented to him so often. The card is judged on how easy it is to remember. It is presented so rarely in the following days. Thus, the user is put at a disadvantage. To help them memorize the cards better, they should be shown more frequently, using cards they can easily remember or already have. by making sure that the flashcards he memorized are seen less frequently in order to reduce his workload Optimal repetition is ensured for each piece of information. (e.g., a user who gives difficult feedback) Use the card every two days until it gives good feedback, and don't forget the card that gives easy feedback. (Not frequent enough, for example, once a week.) This programming applies to thousands of cards. When implemented, it optimizes the user's access to a vast pool of information to meet their needs. It enables them to work and manage. If the user wishes, they can join Arena mode on the same day to compete against others with a similar level of knowledge. Users can compete with each other. ELO is determined based on the success achieved in the match. Their score and rank are updated. Thus, the system integrates individual learning processes with competition. integrated learning-based approach on a single platform is carrying out. The invention relates to a personalized spaced repetition algorithm and dynamic learning analytics. It is an education system characterized by the fact that easily learned information is less common, and difficult information is presented less frequently. to ensure that the learned information is shown more frequently. that learning situation, number of repetitions, probability of forgetting, response time, User rating, last revision date, next revision date Each data card, where data is stored separately, undergoes independent learning. kept in the system as an object, 16 After that information card is completed, the user's answer is reflected in the card's previous version. learning history, the difficulty level of the card, the user their performance will be evaluated together and the time for a new repeat will be determined. calculated and as a result of this calculation, a different repetition for each card. the range was created - To the learning algorithm module; - To significantly reduce the time spent on manual card creation, PDF tutorials are available. notes, written educational materials, question banks, explanatory texts by analyzing and converting into standard information cards the aforementioned Flashcards with a single learning objective, standard format, topic label, and subtopic. artificial intelligence that automatically configures the tag, course tag, and description field. to the intelligence-assisted content creation module; - works directly in conjunction with the learning algorithm module and allows the user flashcard success rate, response speed, long-term performance, ELO rating companies that store their data in the same database and use that data to compete a competitive environment that facilitates matching and is appropriate to the user's level. gamified learning module that creates It is having. The invention relates to a personalized spaced repetition algorithm and dynamic learning analytics. It is an educational method that has the following characteristics: - Creating a user account, After the user registers with the system, they can choose their target exam and study duration. and selects the course or topic they want to learn, and the system By creating a personal learning profile for the user, all performance... It starts storing its data under this profile, - Preparation of training content, The information cards within that system are analyzed by experts. from prepared content or AI-powered content creation consists of standard information cards generated by the module. Each information card is brought in and includes: topic, subtopic, lesson, label, description, The learning parameters are recorded in the database. - Creating a daily work schedule, 17 Only the information cards that the user needs to work with on that day. To enable them to see it, the system starts each session; the cards the user has worked with before, the possibility of forgetting them, again By analyzing the cards that are due, and the new cards to be learned, daily... It automatically generates the work list. - Showing information cards, The system displays the information cards to the user one by one, and in the first stage... Only the question or missing information is displayed; the user has the answer in mind. After creating it, use the "Show Answer" command to select the correct answer. It displays the answer and explanation. - Gathering user feedback, After that answer is displayed, the user card will show: again, hard, good, easy scores are given using one of the evaluation options in this format; The evaluation indicates the extent to which the user has learned the card. It is recorded in the system as feedback. - Calculating the recurrence time, The system uses the feedback received from the user to analyze the card's previous learning. By evaluating the history and performance data, relevant information It automatically calculates the next showing time of your card, easy. The cards that have been learned are shown again after a longer period of time, while the difficult ones... The learned flashcards are added back to the study list at shorter intervals. This process is performed independently for each information card. - Updating the performance analysis, After each study session, the system calculates the correct learning rate, repeating the process. success rate, subject-based performance, study time, daily It updates the user profile by calculating progress, this data It is used in creating subsequent work plans. - Working with the gamified learning module, The system takes the user's ELO rating into account and compares them to similar levels. It automatically matches opponents, giving each player the same information cards. It is presented within a specific set of rules; at the end of the competition, the correct answer is given. ELO scores are calculated by evaluating the number of responses, response time, and success rate. The system is updated and the user's rank is recalculated. 18 It has procedural steps. Industrial Application of the Invention: This invention can be used in various educational systems, primarily the TUS (Medical Specialization Examination). 19

Claims

1. The invention relates to a personalized spaced repetition algorithm and dynamic learning analytics. It is an education system characterized by the fact that easily learned information is less common, and difficult information is presented less frequently. to ensure that the learned information is shown more frequently. that learning situation, number of repetitions, probability of forgetting, response time, User rating, last revision date, next revision date Each data card, where data is stored separately, undergoes independent learning. kept in the system as an object, After that information card is completed, the user's answer is reflected in the card's previous version. learning history, the difficulty level of the card, the user their performance will be evaluated together and the time for a new repeat will be determined. calculated and as a result of this calculation, a different repetition for each card. the range was created - To the learning algorithm module; - To significantly reduce the time spent on manual card creation, PDF tutorials are available. notes, written educational materials, question banks, explanatory texts by analyzing and converting into standard information cards the aforementioned Flashcards with a single learning objective, standard format, topic label, and subtopic. artificial intelligence that automatically configures the tag, course tag, and description field. to the intelligence-assisted content creation module; - works directly in conjunction with the learning algorithm module and allows the user flashcard success rate, response speed, long-term performance, ELO rating companies that store their data in the same database and use that data to compete a competitive environment that facilitates matching and is appropriate to the user's level. gamified learning module that creates It is having.

2. The personalized spaced repetition algorithm and dynamics mentioned in Claim 1. It is an educational system that incorporates learning analytics; its characteristic feature is: - User's information card can be categorized as "again", "difficult", "good" or "easy". As a result of the evaluation, the system learned about the relevant information card. it updated its parameters and used the updated parameters to create a new one. It has a learning algorithm module that automatically calculates the repetition time. It is the fact that.

3. The personalized spaced repetition algorithm mentioned in claim 1 or claim 2 and It is an educational system with dynamic learning analytics, the characteristic of which is; - Cards that are due again, new cards to be taught, user's daily study capacity, target exam date, previous performance data Learning that automatically generates a daily work list by evaluating them together. It has an algorithm module.

4. Personalized intervals mentioned in any of the above requests It is a training system with a repetition algorithm and dynamic learning analytics. feature; - After each user interaction; success rate, topic-based performance, by calculating response time, repetition success, and learning tendency data User profiles are automatically updated and in subsequent learning plans. It has a learning algorithm module that it uses.

5. Personalized intervals mentioned in any of the above requests It is a training system with a repetition algorithm and dynamic learning analytics. feature; - By analyzing PDFs, digital notes, lecture documents, or similar educational content. converting them into standard flashcards, separating them into learning objectives, topics and subtopics. Creates subject tags, prepares description fields, standardizes data. It has an AI-powered content creation module that records data into its structure. It is the fact that.

6. The invention relates to a personalized spaced repetition algorithm and dynamic learning analytics. It is an educational method that has the following characteristics: - Creating a user account, 21 After the user registers with the system, they can choose their target exam and study duration. and selects the course or topic they want to learn, and the system By creating a personal learning profile for the user, all performance... It starts storing its data under this profile, - Preparation of training content, The information cards within that system are analyzed by experts. from prepared content or AI-powered content creation consists of standard information cards generated by the module. Each information card is brought in and includes: topic, subtopic, lesson, label, description, The learning parameters are recorded in the database. - Creating a daily work schedule, Only the information cards that the user needs to work with on that day. To enable them to see it, the system starts each session; the cards the user has worked with before, the possibility of forgetting them, again By analyzing the cards that are due, and the new cards to be learned, daily... It automatically generates the work list. - Showing information cards, The system displays the information cards to the user one by one, and in the first stage... Only the question or missing information is displayed; the user has the answer in mind. After creating it, use the "Show Answer" command to select the correct answer. It displays the answer and explanation. - Gathering user feedback, After that answer is displayed, the user card will show: again, hard, good, easy scores are given using one of the evaluation options in this format; The evaluation indicates the extent to which the user has learned the card. It is recorded in the system as feedback. - Calculating the recurrence time, The system uses the feedback received from the user to analyze the card's previous learning. By evaluating the history and performance data, relevant information It automatically calculates the next showing time of your card, easy. The cards that have been learned are shown again after a longer period of time, while the difficult ones... The learned flashcards are added back to the study list at shorter intervals. This process is performed independently for each information card. 22 - Updating the performance analysis, After each study session, the system calculates the correct learning rate, repeating the process. success rate, subject-based performance, study time, daily It updates the user profile by calculating progress, this data It is used in creating subsequent work plans. It has procedural steps.

7. The personalized spaced repetition algorithm and dynamics mentioned in Claim 6. It is an educational method that incorporates learning analytics; its characteristic feature is: - How the gamified learning module works, The system takes the user's ELO rating into account and compares them to similar levels. It automatically matches opponents, giving each player the same information cards. It is presented within a specific set of rules; at the end of the competition, the correct answer is given. ELO scores are calculated by evaluating the number of responses, response time, and success rate. The system is updated and the user's rank is recalculated. It has a processing step. 23