An intelligent planning and evaluation system for personalized learning paths

Abstract

The application discloses an intelligent planning and evaluation system for personalized learning paths, and relates to the technical field of digital data processing.The system comprises the following modules: a user portrait module that constructs and dynamically optimizes a user model through multidimensional data collection and algorithm analysis; a resource library module that constructs a structured system according to disciplines, levels and types and correlates indexes; a path planning engine module that generates personalized learning paths by fusing algorithms; a learning process tracking module that collects and analyzes learning behavior data in real time; a dynamic evaluation module that establishes a multi-dimensional index system to quantify learning effects; a feedback adjustment module that generates targeted optimization suggestions; and a visual interactive module that graphically displays information and supports multi-terminal adaptation and path customization.The application realizes closed-loop optimization of the learning process by combining precise user portraits and personalized path planning with dynamic evaluation and feedback adjustment, and the resource adaptation is accurate, the learning efficiency and effects are significantly improved, and the personalized learning needs of different learners are effectively met.

Description

Technical Field

[0001] This invention relates to the field of digital data processing technology, and in particular to a personalized learning path intelligent planning and evaluation system. Background Technology

[0002] With the popularization of online education and the concept of lifelong learning, learners' demand for personalized learning services is becoming increasingly urgent. Traditional learning systems often adopt a uniform learning path design, applying the same learning content, schedule, and assessment standards to all users, completely ignoring individual differences among learners in terms of learning foundation, knowledge mastery, learning ability, and learning preferences. This model leads to learners with weak foundations struggling to keep up, while learners with strong abilities cannot obtain sufficient expansion and improvement, greatly affecting learning outcomes and user motivation. At the same time, existing learning systems' resource recommendations are mostly based on simple keyword matching, lacking in-depth analysis of resource and user suitability, making it difficult to accurately meet the personalized needs of different learners.

[0003] Existing learning systems that claim to offer personalized features still suffer from numerous technical shortcomings. Regarding user profiling, most only collect limited basic data, lacking continuous tracking and in-depth analysis of multi-dimensional data such as learning behavior, ability changes, and goal dynamics. This results in significant discrepancies between user profiles and actual situations, failing to provide reliable support for personalized services. In the path planning stage, existing systems employ rigid planning logic, often generating paths based on fixed knowledge hierarchy sequences. They lack dynamic responses to users' real-time learning status, progress deviations, and changes in interest, making flexible adjustments difficult based on actual learning progress. In terms of assessment and feedback, evaluation indicators are singular, focusing primarily on summative assessments of knowledge mastery, lacking comprehensive consideration of dimensions such as learning ability and efficiency. Furthermore, feedback adjustments are often limited to simple resource recommendations, lacking targeted path optimization, methodological guidance, and other in-depth support, failing to fundamentally address learners' problems.

[0004] Furthermore, existing systems suffer from insufficient collaboration and limited adaptability. Some systems lack effective learning interaction mechanisms, making it difficult for learners to obtain peer support and collaboration opportunities, resulting in a relatively isolated learning process. Most systems are only compatible with a single subject or a specific learning stage, failing to meet the diverse and interdisciplinary learning needs of learners. Simultaneously, the systems' data security and privacy protection mechanisms are inadequate, posing a risk of leakage of sensitive user information. Some systems lack self-learning and optimization capabilities, leading to performance degradation after long-term use and difficulty in adapting to constantly changing learning needs and scenarios. These technological limitations result in low levels of personalized service in existing systems, failing to fully leverage the advantages of digital learning and hindering the high-quality development of online education. Summary of the Invention

[0005] This invention proposes a personalized learning path intelligent planning and evaluation system to solve the problems mentioned in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a personalized learning path intelligent planning and evaluation system, comprising the following modules: User profile building module: Collects data covering learning foundation, knowledge mastery, learning ability, learning preferences, time arrangement and goal requirements, uses K-means clustering algorithm to classify users, and combines decision tree algorithm to mine the association rules between user learning behavior and goal achievement, and continuously updates data to dynamically optimize profiles; Learning Resource Library Module: Constructs a structured resource system, labels each resource with knowledge point tags, difficulty level, learning time and suitability level, and establishes a resource association graph to realize the association index between knowledge points; Path planning engine module: Based on user profiles and learning objectives, it integrates reinforcement learning and collaborative filtering algorithms to generate personalized learning paths, sets differentiated path generation strategies for different user types, and reserves a path adjustment interface to support dynamic optimization; Learning process tracking module: Collects user learning behavior data in real time, uncovers user learning patterns and potential problems, monitors deviations between learning progress and plan through a sliding window algorithm, marks abnormal learning behaviors and triggers alerts; Dynamic evaluation module: Establish a multi-dimensional evaluation index system, use fuzzy comprehensive evaluation method to quantitatively evaluate the user's learning effect, generate evaluation reports in real time, and analyze the learning progress trend and existing shortcomings by comparing the evaluation results of users at different stages. Feedback and Adjustment Module: This module can intelligently generate adjustment suggestions. When the evaluation results show that the knowledge points are not up to standard, it automatically recommends targeted reinforcement resources; when the learning progress is lagging behind, it optimizes the subsequent learning plan and adjusts the task intensity; when it detects changes in the user's learning preferences, it updates the recommended resource types and path arrangements in a timely manner. Visualized Interactive Module: Displays personalized learning paths, learning progress, evaluation results, and adjustment suggestions through a graphical interface. It supports users in viewing learning data statistics and analysis reports, provides a path customization and modification function, and supports multi-terminal adaptation for mobile and PC.

[0007] Furthermore, it also includes a comprehensive user capability assessment unit. This unit constructs a capability assessment model based on user learning data and calculates the user's comprehensive capability value through multi-dimensional weighted indicators. The assessment model expression is as follows: ,in This represents the user's overall ability score. To assess the number of indicators, For the first The weighting coefficients of each indicator For the first The raw scores of each indicator For the first The normalization coefficient of each indicator, This is the indicator number, and the model enables accurate quantitative assessment of user capabilities.

[0008] Furthermore, it also includes a learning resource intelligent adaptation module. This module builds an adaptation calculation model based on user profiles and resource attributes, and evaluates the degree of adaptation between resources and users from three dimensions: knowledge point matching, difficulty adaptation, and preference matching. It also establishes a dynamic resource update mechanism, regularly collects high-quality learning resources, and performs tagging and quality assessment, eliminates outdated and inefficient resources, and optimizes resource recommendation strategies based on user feedback on resources.

[0009] Furthermore, it also includes a learning objective decomposition unit, which breaks down the user-defined long-term learning objectives into phased sub-objectives according to the time dimension and knowledge level. The decomposition model expression is as follows: ,in For the first Each stage of sub-goals, For users' long-term learning goals, For the first The time percentage coefficient for each stage For the first Knowledge weight coefficients for each stage Each stage is numbered, and each stage's sub-goal clearly defines the corresponding knowledge points, learning resources, completion deadlines, and assessment standards. The gradual achievement of sub-goals promotes the realization of long-term goals, while also supporting the dynamic adjustment of sub-goals.

[0010] Furthermore, it includes a learning behavior anomaly diagnosis module. This module constructs a normal learning behavior model, compares and analyzes the deviation between the user's real-time learning behavior data and the model, and sets differentiated deviation thresholds to adapt to different learning stages and user types. When abnormal behavior is detected, it analyzes the cause of the anomaly and generates targeted solutions. If the anomaly is caused by excessive learning difficulty, it recommends basic reinforcement resources; if the anomaly is caused by insufficient learning interest, it recommends related resources with strong interest; if the anomaly is caused by unreasonable time arrangement, it optimizes the learning plan and time allocation suggestions.

[0011] Furthermore, it also includes a collaborative learning interaction module, which matches like-minded learning partners based on user profiles and learning goals to build small learning groups; at the same time, it sets group tasks and collective goals, establishes an interactive behavior evaluation mechanism, and incorporates the evaluation results into the user's comprehensive evaluation system as a reference for path adjustment and resource recommendation.

[0012] Furthermore, it also includes a learning method recommendation module. This module analyzes the strengths and weaknesses of users' existing learning methods based on user learning behavior data and evaluation results, and recommends suitable learning methods in combination with the characteristics of different subjects and knowledge types. At the same time, it provides guidance and practical cases on learning methods, tracks the changes in the effect of users after using new learning methods, and dynamically adjusts the recommendation strategy according to the evaluation results.

[0013] Furthermore, it also includes a data security and privacy protection module. This module uses encryption technology to protect the storage and transmission of sensitive information, sets up an access control mechanism to authorize only users and system administrators to access relevant data, establishes a data backup and recovery mechanism to regularly back up user data and system data, and follows the principle of data minimization to collect only the user data necessary to achieve system functions.

[0014] Furthermore, it also includes a system self-learning optimization module. This module uses deep learning algorithms to continuously optimize the core algorithm, analyzes the learning effect data of different user types to discover the optimal path planning pattern and resource recommendation combination, regularly updates the user profile construction algorithm and learning behavior analysis model, and supports online upgrades and iterations of the algorithm model, which can be optimized without interrupting system operation.

[0015] Furthermore, it also includes a multilingual and multidisciplinary adaptation module, which supports multiple language switching, expands the learning resource library and evaluation index system by subject category, covering basic and professional subjects; and optimizes path generation strategies and evaluation methods based on the knowledge characteristics and learning patterns of each subject.

[0016] Compared with existing technologies, the beneficial effects of this invention are: The user profile building module uses multi-dimensional data collection and algorithm analysis to accurately build and dynamically optimize user models, ensuring that profiles can match users' actual situations in real time, laying a solid foundation for subsequent personalized services. The learning resource library module builds a structured resource system and association graph, supporting accurate retrieval and matching based on user profiles, ensuring the relevance and suitability of recommended resources. The path planning engine module integrates advanced algorithms and differentiated strategies to generate personalized learning paths that meet user needs, satisfying both the consolidation needs of users with weak foundations and the expansion needs of users with strong abilities.

[0017] The learning process tracking module collects and analyzes user learning behavior data in real time, accurately uncovering learning patterns and potential problems to provide data support for dynamic adjustments. The dynamic evaluation module establishes a multi-dimensional evaluation indicator system, combining formative and summative assessments to comprehensively quantify learning outcomes and clearly present progress trends and shortcomings. The feedback and adjustment module generates targeted suggestions such as resource replacement, path optimization, and methodological guidance based on process data and evaluation results, achieving closed-loop optimization of the learning process and effectively solving various problems in learning.

[0018] The visual interaction module displays learning-related information through an intuitive graphical interface, supports customized path modification and multi-terminal adaptation, enhancing user experience and learning convenience; the comprehensive user ability assessment unit enables precise quantification of user capabilities, providing a scientific basis for path planning and adjustment; the intelligent learning resource matching module ensures the timeliness, quality, and suitability of recommended resources, further improving learning effectiveness; and the learning goal decomposition unit breaks down long-term goals into actionable, phased sub-goals, reducing learning difficulty and increasing goal achievement rate.

[0019] The learning behavior anomaly diagnosis module promptly identifies and resolves abnormal learning states, ensuring the smooth progress of the learning process; the collaborative learning interaction module enriches learning formats, enhancing the fun and initiative of learning; the learning method recommendation module helps users optimize their learning methods and improve learning efficiency; the data security and privacy protection module comprehensively protects user data security and rights; the system self-learning optimization module ensures that system performance continuously improves over time, adapting to ever-changing needs; and the multi-language and multi-disciplinary adaptation module expands the system's applicability, meeting diverse learning needs.

[0020] Overall, this invention, through the collaborative efforts of multiple modules, constructs a comprehensive personalized learning service system that encompasses user profiling, resource matching, path planning, process tracking, evaluation and feedback, and interactive optimization. It effectively addresses the shortcomings of existing systems, such as insufficient personalization, poor adaptability, and untimely feedback, significantly improving the accuracy, efficiency, and engagement of learning. It fully meets the diverse needs of different learners, provides high-quality technical support for online education and lifelong learning, and possesses broad application value. Attached Figure Description

[0021] Figure 1 This is a schematic block diagram of the personalized learning path intelligent planning and evaluation system proposed in this invention. Figure 2 A comparison chart of learning outcomes across different subjects; Figure 3 The effect diagram shows the dynamic adjustment of the learning path; Figure 4 Use frequency analysis charts for learning resource types; Figure 5A radar chart for multi-dimensional assessment of user learning ability. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0024] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly; for example, they may refer to a fixed connection, a detachable connection, or an integral connection; they may refer to a mechanical connection or an electrical connection; they may refer to a direct connection or an indirect connection through an intermediate medium; and they may refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The invention will now be described in further detail with reference to the accompanying drawings.

[0025] Reference Figures 1 to 5 A personalized learning path intelligent planning and evaluation system, comprising the following modules: User profile building module: Constructs accurate user models through multi-dimensional data collection and analysis. The collected data covers learning foundation, knowledge mastery, learning ability, learning preferences, time arrangement and goal requirements. K-means clustering algorithm is used to classify users. Decision tree algorithm is combined to mine the association rules between user learning behavior and goal achievement. The profile is dynamically optimized by continuously updating user learning data so that the profile matches the actual situation of the user in real time. Learning Resource Library Module: Constructs a structured resource system by subject area, knowledge level, and resource type. Resource types include video courses, documents, exercises, and practical projects. Each resource is labeled with knowledge point tags, difficulty level, learning duration, and suitability level. A resource association graph is established to realize the association index between knowledge points. It supports fast resource retrieval and accurate matching based on user profile. The path planning engine module generates personalized learning paths based on user profiles and learning objectives, integrating reinforcement learning and collaborative filtering algorithms. The path includes phased learning objectives, recommended resource sequences, learning progress arrangements, and assessment nodes. Differentiated path generation strategies are set for different user types: users with weak foundations focus on consolidating basic knowledge, while users with strong abilities focus on expanding and improving their skills. At the same time, a path adjustment interface is reserved to support dynamic optimization. Learning process tracking module: Collects user learning behavior data in real time, including resource access records, learning duration, exercise answer status, knowledge mastery progress, etc. It uses time series data analysis algorithms to mine user learning patterns and potential problems, monitors the deviation between learning progress and plan through sliding window algorithm, marks abnormal learning behavior and triggers early warning; Dynamic assessment module: Establishes a multi-dimensional assessment index system, including knowledge mastery, learning ability improvement, learning efficiency, and goal achievement. It adopts fuzzy comprehensive evaluation method to quantitatively assess the user's learning effect, combines formative assessment and summative assessment, generates assessment reports in real time, and analyzes the learning progress trend and existing shortcomings by comparing the assessment results of users at different stages. Feedback and Adjustment Module: Based on learning process data and dynamic evaluation results, it intelligently generates adjustment suggestions, including resource replacement, path optimization, and learning method guidance. When the evaluation results show that the knowledge points are not up to standard, it automatically recommends targeted reinforcement resources. When the learning progress is lagging behind, it optimizes the subsequent learning plan and adjusts the task intensity. When it detects changes in the user's learning preferences, it updates the recommended resource types and path arrangements in a timely manner. Visualized Interactive Module: Displays personalized learning paths, learning progress, assessment results, and adjustment suggestions through a graphical interface. It supports users in viewing detailed learning data statistics and analysis reports, provides a path customization function, allowing users to adjust their learning plans according to their own situation, and supports multi-terminal adaptation on mobile and PC to ensure a consistent learning experience.

[0026] This invention also includes a comprehensive user capability assessment unit. This unit constructs a capability assessment model based on user learning data and calculates the user's comprehensive capability value through multi-dimensional weighted indicators. The assessment model expression is as follows: ,in This represents the user's overall ability score. To assess the number of indicators, For the first The weight coefficients of each indicator are determined using the analytic hierarchy process (AHP). For the first The raw scores for each indicator are derived from statistical analysis of the learning data. For the first The normalization coefficients of each indicator map the original scores to the interval between 0 and 1. Using the indicator sequence number, this model enables precise quantitative assessment of user capabilities, providing data support for personalized learning path planning and dynamic adjustment, and ensuring that path planning is highly adapted to user capability levels.

[0027] This invention also includes an intelligent learning resource adaptation module. This module constructs an adaptation calculation model based on user profiles and resource attributes, and evaluates the degree of adaptation between resources and users from three dimensions: knowledge point matching, difficulty adaptation, and preference matching. When the adaptation degree is lower than a preset threshold, alternative resources are automatically selected. At the same time, a dynamic resource update mechanism is established to regularly collect high-quality learning resources, label them, and evaluate their quality, while eliminating outdated and inefficient resources. The resource recommendation strategy is optimized based on user feedback on resources to ensure the timeliness, adaptability, and quality of recommended resources, thereby improving learning effectiveness and user experience.

[0028] This invention also includes a learning objective decomposition unit, which decomposes the user-defined long-term learning objective into phased sub-objectives according to the time dimension and knowledge level. The decomposition model expression is as follows: ,in For the first Each stage of sub-goals, For users' long-term learning goals, For the first The time allocation coefficient for each stage is determined based on the total learning cycle. For the first The knowledge weight coefficients for each stage are determined based on the importance of the knowledge level. Each stage is numbered, and each stage's sub-goal clearly defines the corresponding knowledge points, learning resources, completion deadlines, and assessment standards. The gradual achievement of sub-goals drives the realization of long-term goals, while also supporting dynamic adjustments to sub-goals. When user evaluation results show that the improvement in ability exceeds expectations, the difficulty of subsequent sub-goals is appropriately increased; when evaluation results show that the improvement does not meet expectations, the progress and difficulty of sub-goals are adjusted to ensure the rationality and feasibility of the goal setting.

[0029] This invention also includes a learning behavior anomaly diagnosis module. This module constructs a normal learning behavior model and compares and analyzes the deviation between the user's real-time learning behavior data and the model. Deviation indicators include fluctuations in learning duration, abnormal resource access frequency, and sudden changes in the accuracy of answering exercises. Differentiated deviation thresholds are set to adapt to different learning stages and user types. When abnormal behavior is detected, the module analyzes the cause of the abnormality and generates targeted solutions. If the abnormality is caused by excessive learning difficulty, basic reinforcement resources are recommended; if the abnormality is caused by insufficient learning interest, related resources with strong interest are recommended; if the abnormality is caused by unreasonable time arrangement, the module optimizes the learning plan and time allocation suggestions to help users adjust their learning status in a timely manner and ensure the smooth progress of the learning process.

[0030] This invention also includes a collaborative learning interaction module. This module matches like-minded learning partners based on user profiles and learning goals, forming small learning groups that support interactive functions such as sharing learning progress, mutual resource assistance, and problem discussion within the group. It also sets group tasks and collective goals to enhance learning motivation and efficiency through teamwork. Furthermore, it establishes an interactive behavior evaluation mechanism to record user participation and contribution data in collaborative learning, incorporating the evaluation results into a comprehensive user assessment system as a reference for path adjustment and resource recommendation, thus enriching learning formats and enhancing the fun and initiative of learning.

[0031] This invention also includes a learning method recommendation module. Based on user learning behavior data and evaluation results, this module analyzes the strengths and weaknesses of the user's existing learning methods and recommends suitable learning methods based on the characteristics of different subjects and knowledge types. For example, it recommends associative memory and spaced repetition for memorization-based knowledge points, and mind mapping and logical reasoning for logic-based knowledge points. It also provides detailed guidance and practical examples of learning methods, tracks the changes in the user's performance after using new learning methods, and dynamically adjusts the recommendation strategy based on evaluation results to help users optimize their learning methods and improve learning efficiency and effectiveness.

[0032] This invention also includes a data security and privacy protection module. This module uses encryption technology to store and protect sensitive information such as user personal information and learning data. It sets up an access control mechanism, authorizing only users and system administrators to access relevant data. It establishes a data backup and recovery mechanism, regularly backing up user data and system data to prevent data loss. It follows the principle of data minimization, collecting only user data necessary for achieving system functions, clearly defining the scope and purpose of data use, and preventing the unauthorized use of user data for other purposes, thus ensuring user data security and privacy rights.

[0033] This invention also includes a system self-learning optimization module. Based on massive user learning data and feedback information, this module uses deep learning algorithms to continuously optimize core algorithms such as path planning models, evaluation index systems, and resource adaptation strategies. By analyzing learning effect data of different user types, it discovers the optimal path planning patterns and resource recommendation combinations; it regularly updates user profile construction algorithms and learning behavior analysis models to improve the system's understanding and responsiveness to user needs; and it supports online upgrades and iterations of algorithm models, enabling optimization without interrupting system operation, thus continuously improving system performance over time and adapting to ever-changing learning needs and scenarios.

[0034] This invention also includes a multilingual and multidisciplinary adaptation module, which supports switching between multiple languages ​​such as Chinese and English to adapt to learning needs in different language environments. It expands the learning resource library and evaluation indicator system by subject category, covering basic subjects such as Chinese, mathematics, English, physics, chemistry, biology, history, geography, and politics, as well as professional subjects such as computer science, finance, and engineering, providing targeted path planning and evaluation services for users of different disciplines. Furthermore, it optimizes path generation strategies and evaluation methods based on the knowledge characteristics and learning patterns of each subject; for example, science subjects emphasize logical reasoning and problem-solving exercises, while humanities subjects emphasize memorization, comprehension, and knowledge organization, ensuring that the system can provide high-quality personalized learning services in different subject scenarios.

[0035] The following two examples further illustrate specific embodiments of the present invention: Example 1: Personalized Mathematics Learning Scenarios for K-12 Middle School Students This embodiment is applied to personalized learning of mathematics for junior high school students. It is tailored to students in different grades from 7th to 9th grade, and adapts to knowledge modules such as algebra, geometry, and statistics. It needs to solve problems such as large differences in students' basic knowledge, insufficient learning interest, and difficulty in making up for knowledge gaps. Through the system, it can achieve accurate profiling, personalized path planning and dynamic optimization, and help students improve their math scores efficiently.

[0036] I. Implementation Details of Core Modules The user profile building module operates as follows: Upon initial login, students complete basic information entry, including grade, class, and a self-assessment of their math foundation. The system automatically pushes a diagnostic test covering various knowledge points, with question types including multiple choice, fill-in-the-blank, and problem-solving, comprehensively assessing students' knowledge mastery. Data collection includes diagnostic test scores, distribution of incorrect answers for each knowledge point, preferred study time (morning, evening, weekend), preferred resource type (video explanations, text / image materials, exercises), daily available study time, and semester goals (pass, good, excellent). The system uses K-means clustering to categorize users into three groups: weak, average, and excellent. Combined with decision tree algorithms, it uncovers the association rules between "error type - learning preference - goal achievement." For example, students with weak foundations who prefer videos will find a combination of short videos and basic exercises more efficient. The system updates user learning data every two weeks, dynamically optimizing profile tags to ensure real-time matching with students' actual situations.

[0037] The learning resource library is configured with modules categorized by grade level (7-9). Each grade level covers three main knowledge modules: algebra, geometry, and statistics. Resource types include micro-lessons (5-10 minutes each), illustrated explanations of knowledge points, tiered exercises (basic, advanced, and extended), practical inquiry projects, and mathematical modeling case studies. Each resource is tagged with knowledge point labels (e.g., solutions to quadratic equations), difficulty levels (1-5), learning duration, and suitability for basic, intermediate, and advanced learners. A resource association map is established; for example, learning the Pythagorean theorem automatically links to related resources on right-angled triangle applications. The resource library regularly adds high-quality teaching resources, which are evaluated by a team of mathematics teachers to eliminate outdated and inefficient resources, ensuring timeliness and quality.

[0038] The path planning engine module generates personalized learning paths based on user profiles and semester goals. Paths for students with weak foundations focus on consolidating basic knowledge; for example, a first-year junior high student might first learn a micro-lesson on the concept of rational numbers followed by basic exercises, then move on to rational number operations. Paths for average-level students balance consolidation and improvement, interspersed with basic and advanced questions. Paths for high-achieving students emphasize expansion and enhancement, incorporating past competition questions and mathematical modeling projects. The path includes weekly and monthly learning goals, recommended resource sequences, learning schedules, and assessment points such as unit tests. An interface for path adjustment is provided to support dynamic optimization based on learning outcomes.

[0039] The learning process tracking module collects student learning behavior data in real time, including resource access records, video viewing progress, number of repeated viewings, exercise answering status (answering time, accuracy rate, incorrect knowledge points), daily and weekly study time, focus level, and knowledge point mastery progress. It employs time-series data analysis algorithms to uncover learning patterns, such as finding that students have the highest accuracy rate when answering questions between 8 PM and 9 PM and the longest study time on weekends. A sliding window algorithm monitors deviations between learning progress and the plan; when the learning progress for a particular knowledge point lags behind the plan by more than 20%, it is marked as abnormal and an alert is triggered.

[0040] The dynamic assessment module works by establishing a multi-dimensional assessment indicator system, including knowledge mastery (unit test accuracy rate), error review rate, improvement in learning ability (problem-solving speed), difficulty level tackling rate, learning efficiency (knowledge points mastered per unit time), and the gap between target achievement and semester goals. A fuzzy comprehensive evaluation method is used for quantitative assessment, combining formative assessment (classroom exercises, homework completion) with summative assessment (unit tests, mid-term and final exams) to generate assessment reports in real time. By comparing students' assessment results at different stages, progress trends are analyzed. For example, a student's accuracy rate on quadratic equations improved from 60% to 85%, while also highlighting a weakness in the logical clarity of geometric proofs.

[0041] The feedback and adjustment module responds by generating adjustment suggestions based on learning process data and assessment results. When the assessment results show that the accuracy rate for a knowledge point is below 70%, it automatically recommends targeted reinforcement resources, such as a micro-lesson on "Detailed Explanation of Mistakes in Quadratic Equations" plus specialized exercises. When learning progress lags behind, it optimizes subsequent learning plans, appropriately reducing the number of extension questions and increasing basic consolidation tasks. When it detects a change in student learning preferences from a preference for videos to a preference for text and images, it promptly updates the recommended resource types. It also provides guidance on learning methods; for example, for students weak in geometry proofs, it recommends the "mind mapping for logical reasoning" learning method.

[0042] Additional modules include: a comprehensive user ability assessment unit that calculates overall ability values ​​based on multi-dimensional indicators to provide data support for path planning; an intelligent learning resource matching module that assesses resource suitability from three dimensions: knowledge point matching, difficulty matching, and preference matching, and automatically filters alternative resources; a learning goal decomposition unit that breaks down semester goals into monthly sub-goals, clarifying the knowledge points, resources, and assessment standards corresponding to each sub-goal, and appropriately increasing the difficulty of subsequent sub-goals when student assessment results show that ability improvement exceeds expectations; and a learning behavior anomaly diagnosis module that detects a sharp drop in a student's answer accuracy rate for three consecutive days, analyzes it as excessive learning difficulty, and recommends basic reinforcement. Resources; the collaborative learning interaction module matches students with similar interests to form learning groups, supporting discussions of incorrect answers and resource sharing within the group; the learning method recommendation module recommends spaced repetition for memorization-based knowledge points and mind mapping for logic-based knowledge points; the visualization interaction module displays learning paths, progress, and assessment results on PC and mobile devices, allowing students to customize and adjust their learning plans; the data security and privacy protection module encrypts and stores students' personal information and learning data, and sets access permissions; the system self-learning optimization module optimizes the path planning model based on massive student data; the multi-subject adaptation module can be expanded to subjects such as Chinese and English in the future.

[0043] Table 1: Comparison of Personalized Mathematics Learning Performance for Junior High School Students Table 1 clearly demonstrates the advantages of this invention in junior high school mathematics learning. Traditional standardized learning systems use the same learning paths and resources, resulting in a knowledge point mastery rate of only 65% ​​and a target achievement rate of 58%, failing to meet the individual needs of students. This invention, through precise user profiling and personalized path planning, increases the knowledge point mastery rate to 88%; combined with dynamic assessment and targeted feedback, learning efficiency is improved by 35%, and the semester target achievement rate reaches 85%. Student satisfaction increases from 62% to 90%, and the error correction rate increases from 45% to 82%, effectively addressing the problem of difficult-to-make up knowledge gaps. The system's collaborative learning and learning method guidance functions further enhance students' learning initiative and efficiency, providing comprehensive and personalized support for junior high school mathematics learning.

[0044] Example 2: Personalized Learning Scenario for Adult Python Programming Professional Skills This embodiment is applied to adult Python programming professional skills learning, and is suitable for learners with different needs such as zero-based beginners, those with programming experience advancing, and those seeking to improve workplace skills. It needs to solve problems such as fragmented learning time, strong goal orientation, and high demand for practical operation. Through the system, it realizes personalized path planning, practical guidance, and dynamic evaluation, helping learners to quickly master Python skills and apply them to their work.

[0045] I. Implementation Details of Core Modules The user profile building module operates as follows: After registration, learners complete basic information entry, including whether they have programming experience, their previous professional background, learning goals, career applications, career transition interests, daily available study time, the degree of fragmentation in their study time (whether they can study continuously), and their preferred learning methods (online live streaming, recorded videos, self-study). The system pushes basic programming tests to assess mastery of core knowledge points such as variables, loops, and functions, collecting test scores, error distribution, and study time arrangements (e.g., only able to utilize commuting time and evenings after get off work). The system uses K-means clustering to categorize users into three groups: beginners, those with some basic knowledge, and those seeking career advancement. It also uses decision tree algorithms to uncover association rules between "fragmented study time - resource type - skill improvement." For example, users with fragmented learning strategies benefit more from a combination of short practical videos and small programming tasks. The system updates user learning data monthly, dynamically optimizing profile tags to adapt to learners' skill improvement and changing goals.

[0046] The learning resource library is configured with modules categorized by skill level: beginner, intermediate, and advanced, covering Python basics, data processing, web development, web scraping, and machine learning. Resource types include pre-recorded courses (10-15 minutes each), text tutorials, practical examples with online programming environments, project-based e-commerce data analysis, web scraping development, and workplace application scenarios. Each resource is tagged with knowledge point labels (e.g., Pandas data processing), difficulty levels (1-5), learning duration, suitable skill level, and workplace application scenarios (data analysis and development). A resource association graph is established; for example, learning "Python functions" automatically links to practical examples of "functions in data analysis." The resource library supports multilingual support, providing both Chinese and English resources to meet the needs of different language environments.

[0047] The path planning engine module generates personalized learning paths based on user profiles and learning goals. For learners with no prior Python experience, the path starts with basic Python syntax, beginning with micro-lessons on "variables and data types" followed by simple practical tasks, then progressing to "loops and conditional statements." For learners with some experience, the path skips basic syntax and directly covers "data processing" and "practical projects." For learners seeking career advancement, the path is tailored to their target job requirements; for example, learners aiming for data analysis roles will focus on Pandas, Matplotlib resources, and data analysis projects. The path includes phased learning objectives (mastering one to two core skills at each stage), recommended resource sequences, a learning schedule, and assessment points such as practical project evaluation. It supports dynamic adjustments based on learning progress.

[0048] The learning process tracking module collects learner behavior data in real time, including resource access records, video viewing progress, completion status of practical cases, coding task responses, code submission counts, execution accuracy, debugging time, learning duration (fragmented learning time, continuous learning time), and skill mastery progress. It employs time-series data analysis algorithms to uncover learning patterns; for example, it discovers that fragmented learners are better suited to completing small practical tasks during their commute and to advancing project practice during continuous weekend time. A sliding window algorithm monitors deviations between learning progress and the plan; when the learning progress of a skill lags behind the plan by more than 30%, it is marked as abnormal and an alert is triggered.

[0049] The dynamic evaluation module works by establishing a multi-dimensional evaluation index system, including skill mastery, practical task accuracy, project completion quality, learning ability improvement, code debugging efficiency, problem-solving rate, learning efficiency, skill mastery per unit time, and the gap between goal achievement and workplace application goals. A fuzzy comprehensive evaluation method is used for quantitative assessment, combined with formative assessment of practical task completion, code review, and summative assessment of project acceptance, generating evaluation reports in real time. By comparing evaluation results at different stages, progress trends are analyzed; for example, a learner progresses from being unable to independently complete a web scraping project to being able to independently develop a simple e-commerce web scraper, while also identifying shortcomings in code optimization capabilities.

[0050] The feedback and adjustment module responds by generating adjustment suggestions based on learning process data and assessment results. When assessment results show that skills are not up to standard and the accuracy rate of practical tasks is below 60%, it automatically recommends targeted reinforcement resources, such as a micro-course on "Python Web Scraping Debugging Techniques" plus a specific practical task. When learning progress lags behind, it optimizes subsequent learning plans, breaking down large projects into smaller tasks to suit fragmented learning. When it detects a change in learner preferences from pre-recorded videos to live interactive sessions, it promptly updates the recommended resource types. It also provides workplace application guidance; for example, for learners in data analysis roles, it recommends practical case studies on "Python Applications in Excel Data Processing."

[0051] Additional modules include: a comprehensive user ability assessment unit that calculates learners' overall ability scores to provide data support for path planning; an intelligent learning resource matching module that assesses resource suitability and automatically filters alternative resources; a learning goal decomposition unit that breaks down workplace skill goals into phased sub-goals, clarifying the skills, resources, and acceptance criteria for each sub-goal, and appropriately increasing the difficulty of subsequent sub-goals when assessment results show that ability improvement exceeds expectations; a learning behavior anomaly diagnosis module that detects learners have not submitted programming tasks for a week, analyzing that the task difficulty is too high, breaking down the task and providing step-by-step guidance; a collaborative learning interaction module that matches learners with similar target positions to form project teams, supporting project collaboration and code review within the team; a learning method recommendation module that recommends methods such as "learning by doing" and "reviewing mistakes" for programming learning; a visual interaction module that displays learning paths, progress, and assessment results, supports customized adjustments to learning plans, and is compatible with mobile and PC platforms; a data security and privacy protection module that encrypts and stores learner information and learning data; a system self-learning optimization module that optimizes the path planning model based on massive user data; and a multi-language and multi-disciplinary adaptation module that supports switching between Chinese and English, and can be expanded to other programming disciplines such as Java and C++ in the future.

[0052] Table 2: Comparison of Personalized Learning Performance of Python Programming for Adults Table 2 data highlights the application value of this invention in adult Python programming learning scenarios. Traditional programming learning platforms use standardized resources and paths, resulting in only 60% skill mastery, 55% project completion quality, and a low learning persistence rate of 40%, making it difficult to adapt to the fragmented learning needs and career goals of adults. This invention, through precise profiling and personalized path planning, improves skill mastery to 85% and project completion quality to 82%; its path design and task breakdown adapted to fragmented learning increase the learning persistence rate to 78%; and its resource recommendation and goal breakdown combined with workplace application scenarios achieve 88% workplace application adaptability. User satisfaction increases from 58% to 91%, effectively addressing the needs of adults with fragmented learning time and strong goal orientation, providing efficient and personalized support for workplace skill improvement.

[0053] Reference Figure 2This chart visually demonstrates the advantages of this system in different subjects. Traditional methods resulted in an average improvement of 10% in mathematics, 15% in English, 8% in programming, 12% in physics, and 18% in history. This system, through personalized path planning and dynamic assessment, improved scores to 25% in mathematics, 30% in English, 20% in programming, 28% in physics, and 35% in history. History saw the largest improvement because the system provides personalized learning resources and paths tailored to the characteristics of history, helping students better understand and memorize historical knowledge. The improvement in programming was relatively smaller, but still reached 20%, likely because programming learning requires more practice, and this system provides students with more hands-on opportunities and guidance in this regard. Overall, this system demonstrated good adaptability and effectiveness across different subjects, meeting the learning needs of students in various disciplines.

[0054] Reference Figure 3 This graph illustrates how the average difficulty level of recommended resources changes over time during the Python programming fundamentals learning process. In the early stages, the system recommends resources with lower difficulty levels, primarily to help students build a solid foundation. As learning progresses, the difficulty gradually increases to meet students' evolving learning needs. In weeks 4 and 7, the difficulty decreases, possibly because the system detected less than ideal learning outcomes or difficulties encountered by students at these stages, thus reducing the difficulty to provide more opportunities for consolidation and practice. Overall, this system dynamically adjusts the learning path and resource difficulty based on students' learning progress, ensuring that students receive effective challenges without experiencing frustration due to excessive difficulty, thereby improving learning efficiency and effectiveness.

[0055] Reference Figure 4 This chart illustrates the frequency of different types of learning resources used by students while using this system. Video courses are used most frequently, accounting for 40%, because their intuitive and vivid nature helps students better understand and master knowledge. Documents are used next most frequently, accounting for 25%, as they provide students with more detailed and systematic explanations. Exercises and practical projects are used relatively less frequently, accounting for 20% and 15% respectively. Exercises help students consolidate their knowledge, while practical projects allow them to apply their knowledge in real-world situations. Overall, this system provides a variety of learning resources to meet the diverse learning needs and styles of students. Students can choose appropriate resources based on their own circumstances to improve their learning outcomes.

[0056] Reference Figure 5This chart presents a multi-dimensional assessment of a user's learning ability. The user excels most in learning interest, scoring 9.0, indicating high enthusiasm and motivation. Knowledge mastery is also high at 8.5, demonstrating good understanding of the learned material. Learning ability and habits are at 7.8 and 7.0 respectively, placing the user in the upper-middle range. Learning efficiency is relatively low at 6.5, which may be an area for improvement. Overall, the radar chart provides a comprehensive understanding of a user's learning ability. The system can offer personalized learning suggestions and support based on the assessment results, helping users further improve their learning abilities and outcomes.

[0057] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A personalized learning path intelligent planning and evaluation system, characterized in that, Includes the following modules: User profile building module: Collects data covering learning foundation, knowledge mastery, learning ability, learning preferences, time arrangement and goal requirements, uses K-means clustering algorithm to classify users, and combines decision tree algorithm to mine the association rules between user learning behavior and goal achievement, and continuously updates data to dynamically optimize profiles; Learning Resource Library Module: Constructs a structured resource system, labels each resource with knowledge point tags, difficulty level, learning time and suitability level, and establishes a resource association graph to realize the association index between knowledge points; Path planning engine module: Based on user profiles and learning objectives, it integrates reinforcement learning and collaborative filtering algorithms to generate personalized learning paths, sets differentiated path generation strategies for different user types, and reserves a path adjustment interface to support dynamic optimization; Learning process tracking module: Collects user learning behavior data in real time, uncovers user learning patterns and potential problems, monitors deviations between learning progress and plan through a sliding window algorithm, marks abnormal learning behaviors and triggers alerts; Dynamic evaluation module: Establish a multi-dimensional evaluation index system, use fuzzy comprehensive evaluation method to quantitatively evaluate the user's learning effect, generate evaluation reports in real time, and analyze the learning progress trend and existing shortcomings by comparing the evaluation results of users at different stages. Feedback and Adjustment Module: This module can intelligently generate adjustment suggestions. When the evaluation results show that the knowledge points are not up to standard, it automatically recommends targeted reinforcement resources. When the learning progress is lagging behind, it optimizes the subsequent learning plan and adjusts the task intensity. When changes in user learning preferences are detected, the recommended resource types and path arrangements should be updated promptly. Visualized Interactive Module: Displays personalized learning paths, learning progress, evaluation results, and adjustment suggestions through a graphical interface. It supports users in viewing learning data statistics and analysis reports, provides a path customization and modification function, and supports multi-terminal adaptation for mobile and PC.

2. The personalized learning path intelligent planning and evaluation system according to claim 1, characterized in that, It also includes a comprehensive user capability assessment unit. This unit builds a capability assessment model based on user learning data and calculates the user's comprehensive capability value through multi-dimensional weighted indicators. The assessment model expression is as follows: ,in This represents the user's overall ability score. To assess the number of indicators, For the first The weighting coefficients of each indicator For the first The raw scores of each indicator For the first The normalization coefficient of each indicator, This is the indicator number, and the model enables accurate quantitative assessment of user capabilities.

3. The personalized learning path intelligent planning and evaluation system according to claim 1, characterized in that, It also includes a learning resource intelligent adaptation module, which builds an adaptation calculation model based on user profiles and resource attributes. It evaluates the degree of adaptation between resources and users from three dimensions: knowledge point matching, difficulty adaptation, and preference matching. It establishes a dynamic resource update mechanism, regularly collects high-quality learning resources, and performs tagging and quality assessment, eliminates outdated and inefficient resources, and optimizes resource recommendation strategies based on user feedback on resources.

4. The personalized learning path intelligent planning and evaluation system according to claim 1, characterized in that, It also includes a learning objective decomposition unit, which breaks down the user's long-term learning objectives into phased sub-objectives based on time and knowledge level. The decomposition model expression is as follows: ,in For the first Each stage of sub-goals, For users' long-term learning goals, For the first The time percentage coefficient for each stage For the first Knowledge weight coefficients for each stage Each stage is numbered, and each stage's sub-goal clearly defines the corresponding knowledge points, learning resources, completion deadlines, and assessment standards. The gradual achievement of sub-goals promotes the realization of long-term goals, while also supporting the dynamic adjustment of sub-goals.

5. The personalized learning path intelligent planning and evaluation system according to claim 1, characterized in that, It also includes a learning behavior anomaly diagnosis module. This module constructs a normal learning behavior model, compares and analyzes the deviation between the user's real-time learning behavior data and the model, sets differentiated deviation thresholds to adapt to different learning stages and user types; when abnormal behavior is detected, it analyzes the cause of the anomaly and generates targeted solutions. If the anomaly is caused by excessive learning difficulty, it recommends basic reinforcement resources. If the abnormality is due to a lack of interest in learning, we recommend related resources that are more engaging; if the abnormality is due to an unreasonable time arrangement, we suggest optimizing the learning plan and time allocation.

6. The personalized learning path intelligent planning and evaluation system according to claim 1, characterized in that, It also includes a collaborative learning interaction module, which matches like-minded learning partners based on user profiles and learning goals to build small learning groups; at the same time, it sets group tasks and collective goals, establishes an interactive behavior evaluation mechanism, and incorporates the evaluation results into the user's comprehensive evaluation system as a reference for path adjustment and resource recommendation.

7. The personalized learning path intelligent planning and evaluation system according to claim 1, characterized in that, It also includes a learning method recommendation module, which analyzes the strengths and weaknesses of users' existing learning methods based on user learning behavior data and evaluation results, and recommends suitable learning methods in combination with the characteristics of different subjects and knowledge types. At the same time, it provides guidance and practical cases on learning methods, tracks the changes in the effect of users after using new learning methods, and dynamically adjusts the recommendation strategy according to the evaluation results.

8. The personalized learning path intelligent planning and evaluation system according to claim 1, characterized in that, It also includes a data security and privacy protection module, which uses encryption technology to protect the storage and transmission of sensitive information, sets up an access control mechanism to authorize only users and system administrators to access relevant data, establishes a data backup and recovery mechanism to regularly back up user data and system data, and follows the principle of data minimization to collect only the user data necessary to achieve system functions.

9. The personalized learning path intelligent planning and evaluation system according to claim 1, characterized in that, It also includes a system self-learning optimization module, which uses deep learning algorithms to continuously optimize the core algorithm. By analyzing the learning effect data of different user types, it can discover the optimal path planning pattern and resource recommendation combination; regularly update the user profile construction algorithm and learning behavior analysis model; and support online upgrades and iterations of the algorithm model, which can be optimized without interrupting system operation.

10. The personalized learning path intelligent planning and evaluation system according to claim 1, characterized in that, It also includes a multilingual and multidisciplinary adaptation module, which supports multiple language switching, expands the learning resource library and evaluation index system by subject category, covering basic and professional subjects; and optimizes path generation strategies and evaluation methods according to the knowledge characteristics and learning patterns of each subject.

Images