An artificial intelligence-based learning resource pushing method

Abstract

The application belongs to the field of course recommendation, and provides a learning resource pushing method based on artificial intelligence, which comprises the following steps: a learning system acquires the age of a user, determines a first brain development degree evaluation result according to the age of the user; the first brain development degree evaluation result is input into a first model, the first model recommends a first learning resource; the first brain development degree evaluation result and the first learning resource are input into a second model, the second model generates a cognitive ability test question according to the age of the user and the first learning resource; the cognitive ability test question is inserted into the teaching process of the first learning resource; the age of the user and the test result of the cognitive ability test question are sent to a third model, the third model outputs a brain development degree evaluation correction result of the user; the brain development degree evaluation correction result is sent to the first model, and the first model recommends a second learning resource according to the brain development degree evaluation correction result.

Description

Technical Field

[0001] This invention belongs to the field of course recommendation, and specifically relates to a learning resource recommendation method based on artificial intelligence. Background Technology

[0002] Traditional learning resource recommendation systems typically push resources based on users' interests or historical learning records. However, this approach ignores, to some extent, the actual differences in users' cognitive abilities and brain development levels, making it difficult to accurately match the learning needs of different users.

[0003] Existing research indicates that the level of human brain development varies significantly across different age groups, directly impacting learners' cognitive abilities, information processing speed, and learning outcomes. Therefore, incorporating a user's brain development level into the core parameters of learning resource recommendations can better match their actual learning needs, thereby improving learning efficiency. However, assessing brain development involves multiple complex dimensions of cognitive ability (such as memory, logical reasoning, and language comprehension), and the assessment results may deviate from the user's actual cognitive level. Dynamically correcting and optimizing recommendation strategies remains a significant technical challenge. Summary of the Invention

[0004] To address the problems in the existing technology, this invention provides a learning resource recommendation method based on artificial intelligence, comprising the following steps:

[0005] Step S10: The learning system obtains the user's age, and the learning system determines the first brain development level assessment result based on the user's age;

[0006] Step S20: The learning system inputs the first brain development level assessment result into the first model, and the first model recommends the first learning resource;

[0007] Step S30: The learning system inputs the first brain development assessment result and the first learning resource into the second model, and the second model generates cognitive ability test questions based on the user's age and the first learning resource.

[0008] Step S40: The learning system inserts the cognitive ability test questions into the teaching process of the first learning resource;

[0009] Step S50: The learning system sends the user's age and the results of the cognitive ability test to the third model, and the third model outputs the user's brain development assessment and correction results.

[0010] In step S60, the learning system sends the brain development level assessment correction result to the first model, and the first model recommends a second learning resource based on the brain development level assessment correction result.

[0011] Furthermore, the first brain development assessment result is an average developmental level reference value derived from statistical data analysis by age distribution.

[0012] Furthermore, the first model is a rule-based model, a collaborative filtering model, or a neural network model.

[0013] Furthermore, the collaborative filtering model mines the learning resource preference data of users whose developmental deviation is within a preset value to generate learning resource recommendations corresponding to the user evaluation results.

[0014] Furthermore, the second model is either a rule-based adaptive test item generation model or a deep learning-based dynamic evaluation model.

[0015] Furthermore, the cognitive ability test questions are of the same type as the first learning resource topic.

[0016] Furthermore, the learning system dynamically selects suitable teaching nodes as test question insertion points based on monitoring the user's learning status of the first learning resource.

[0017] Furthermore, the third model extracts features from the received user age and cognitive ability test results, and normalizes the test results using preset feature weighting rules.

[0018] Furthermore, the third model, based on a preset brain development assessment rule base, compares the user's test results with statistical data of users of the same age group to calculate the deviation between the user's current ability and the average brain development level.

[0019] Furthermore, the score weights of each capability indicator in the first model's analysis and correction results are used to prioritize recommending learning resources that can improve the user's weak capability areas.

[0020] This invention provides an artificial intelligence-based learning resource recommendation method. By incorporating an assessment of the user's brain development level, combined with cognitive ability test results and a dynamic correction mechanism, it solves the problems of insufficient assessment accuracy, weak resource recommendation targeting, and disconnect between assessment and recommendation in existing technologies. Its beneficial effects are mainly reflected in the following aspects:

[0021] This invention combines multiple indicators, including user age and brain development level, to preliminarily assess a user's cognitive ability. It then uses the results of cognitive ability tests to dynamically correct the assessment model, ensuring that the assessment results accurately reflect the user's actual cognitive level and avoiding the bias problems caused by single-dimensional assessment in traditional methods.

[0022] This invention seamlessly integrates cognitive ability test questions with learning resources during the teaching process, achieving a close combination of learning and assessment. This not only ensures the user's learning experience but also captures changes in the user's cognitive abilities in real time, further optimizing assessment and recommendation strategies.

[0023] This invention dynamically adjusts the recommendation strategy for learning resources based on the revised brain development assessment results, prioritizing learning resources for users' weak areas while also considering the expansion of their strengths. This ensures that the difficulty and content of the learning resources are highly matched with the user's current cognitive level and learning needs, effectively improving learning outcomes. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0026] The invention will now be described in preferred form with reference to the accompanying drawings and specific embodiments.

[0027] This embodiment solves the above problems through the following steps:

[0028] In one embodiment, reference Figure 1 This invention provides a learning resource recommendation method based on artificial intelligence, which involves analyzing user characteristics through intelligent models and accurately matching personalized learning content to meet the different cognitive levels and learning needs of users.

[0029] Specifically, the method includes the following steps:

[0030] Step S10: The learning system obtains the user's age, and the learning system determines the first brain development assessment result based on the user's age.

[0031] The assessment results are average developmental level reference values ​​derived from the analysis of large-scale statistical data distributed by age. Specifically, the first brain development level assessment result is determined through a pre-built model. This model quantifies the assessment based on the distribution range of core indicators such as attention, memory, language comprehension, and executive function, using statistical data on the brain cognitive abilities of users in different age groups. However, due to individual differences, the first brain development level assessment result is a general reference value for an age group, and the actual developmental level of an individual may deviate from the statistical mean. The assessment results provide a preliminary basis for subsequent delivery of learning resources and dynamic adjustments.

[0032] Optionally, brain development scores can be mapped to user age. For example, at age 10, the corresponding memory score is 80 / 100 and the logic score is 60 / 100. Similarly, the average brain development score can be determined for other age groups based on statistical data.

[0033] In step S20, the learning system inputs the first brain development level assessment result into the first model, and the first model recommends the first learning resource.

[0034] The first model can adopt a rule-based model to determine a preset set of learning resources based on the results of the first brain development assessment, and recommend the preset set of learning resources to the user.

[0035] For example, a set of learning resources corresponding to different developmental levels can be set in the database. For instance, a preset number of learning resources can be set when the logical reasoning ability score is 60 / 100-70 / 100. When the user's first brain development level assessment result meets the set rules, the corresponding learning resources are recommended.

[0036] The first model can employ a collaborative filtering model, which mines learning resource preference data of users with the same or similar developmental levels to generate learning resource recommendations corresponding to the user's assessment results; or it can employ a content-based recommendation model, which combines specific cognitive indicators (such as attention level, language ability, etc.) in the user's first brain development assessment results with the matching degree of the content characteristics of the learning resources to output the most suitable first learning resource for the user.

[0037] Furthermore, the collaborative filtering model is implemented in the following way:

[0038] Each user's feature vector is expanded to the following form:

[0039] f u =[a u memory u ,logic u ,language u,p u ]

[0040] in:

[0041] f u Represents the user's feature vector;

[0042] a u Indicates the user's age;

[0043] memory u This indicates the user's memory score;

[0044] logic u This represents a score indicating the user's logical reasoning ability.

[0045] language u This represents the user's language comprehension score;

[0046] p u This represents a user preference score, which can be calculated from the user's historical behavior.

[0047] The feature vector of each learning resource is defined as:

[0048] f r =[d r memory r ,logic r ,language r ]

[0049] in,

[0050] f r The feature vector representing the learning resource;

[0051] d r Indicates the difficulty of obtaining the resource;

[0052] memory r This indicates the degree to which resources are needed to enhance memory capacity;

[0053] logic r This indicates the degree to which resources require reasoning ability;

[0054] language r This indicates the degree to which resources require language comprehension skills.

[0055] Based on user feature vectors and resource feature vectors, a matching function between users and resources is defined:

[0056]

[0057] in:

[0058] wi This is a weighting parameter used to balance the importance of different cognitive dimensions (such as memory ability, logical ability, etc.).

[0059] f u,i and f r,i These are the i-th dimensions in the user and resource feature vectors, respectively.

[0060] A matching function is calculated for the users and resources to obtain an m×n user-resource interaction matrix M, where M is the element of the user-resource interaction matrix. i,j This represents the matching degree between the i-th user and the j-th resource.

[0061] Construct a user set U = {u1, u2, ..., u} m}, where m represents the number of users, u m This represents the m-th user.

[0062] Construct a learning resource set R = {r1, r2, ..., r} n}, where n represents the number of resources, r n This represents the nth resource.

[0063] For any user u, determine a set N(u) of users whose similarity to it is greater than a preset threshold, and determine the collaborative matching degree between user u and resource i based on the behavior of these users.

[0064]

[0065] Where sim(u,v) represents the similarity between users u and v, which can be calculated using Pearson correlation coefficient, cosine similarity, etc. v,i This represents the degree of matching between user v and resource i.

[0066] Traditional collaborative filtering models make recommendations based solely on users' historical behavior and rating data, ignoring differences in users' cognitive abilities. The improved model incorporates users' brain development assessment results (such as memory, logical reasoning, and language comprehension), enabling it to more accurately capture users' learning needs. For example, if a user's initial brain development assessment shows a high memory score (e.g., 85 / 100) but a low logical reasoning score (e.g., 40 / 100), the first model compares data from other users with similar brain development assessment results in a large-scale learning database, extracting learning resources that these users frequently use and find effective, such as engaging math puzzles and spatial logic training games to improve logical reasoning skills.

[0067] In addition, the first model can adopt a neural network model, wherein the input layer receives the first brain development level assessment result, the hidden layer processes the input data through a multi-dimensional weight matrix constructed from the training data, and the output layer generates a list of learning resources suitable for the user's current cognitive level.

[0068] Specifically, the structure of the first model is as follows:

[0069] Input layer:

[0070] The input layer receives the user's brain development assessment results and related characteristics, including the following information:

[0071] User age: Indicates the user's basic age range.

[0072] Memory ability score: A quantitative assessment of a user's memory ability.

[0073] Logical reasoning ability score: A score of the user's logical reasoning ability.

[0074] Language comprehension score: A score of the user's language ability.

[0075] User preference characteristics: such as users' preferences for certain subjects or resource types (e.g., mathematics, language learning, etc.).

[0076] These features constitute the user's feature vector, which serves as the input to the model.

[0077] Hidden layer:

[0078] The hidden layer performs multi-layer non-linear processing on the input user features:

[0079] The first hidden layer performs preliminary processing on the basic features of the user input and extracts important feature relationships.

[0080] For example, based on the combination of memory and logical abilities, users' potential needs in different learning areas can be extracted.

[0081] The second hidden layer combines resource tag characteristics with user features to learn more complex interaction relationships.

[0082] For example, matching users with low logical ability scores with the logical training needs of resources can reveal potential resource recommendation priorities.

[0083] The third hidden layer performs dimensionality reduction on the features to generate matching scores between users and resources, providing support for the output layer.

[0084] Output layer:

[0085] The output layer generates a probability distribution of learning resource recommendations, where the recommendation probability of each learning resource represents the likelihood that a user is suited to that resource. The model outputs a list of recommended learning resources, sorted by priority. For example, the input layer receives the results of a first-order brain development assessment, including specific indicators such as an attention score of 75 / 100 and a language comprehension ability of 90 / 100. The hidden layer processes the input features using a weight matrix generated from the training data, outputting the learning resources that best match these indicators, such as challenging reading comprehension exercises, interactive learning animations requiring high concentration, and memory-based word memorization challenges.

[0086] In this way, the first model can recommend the first learning resource that matches the user's cognitive ability based on the user's specific brain development characteristics, ensuring the personalization and relevance of the learning content.

[0087] First-level learning resources refer to personalized learning tutorials recommended based on the user's brain development assessment results. These tutorials are presented in the form of specific courses, covering learning content of varying difficulty and themes. First-level learning resources can include mathematics courses, language courses, science courses, professional skills courses, etc., aiming to meet the user's current cognitive abilities and learning needs, and to maximize learning effectiveness through highly adaptable content design.

[0088] For example, user characteristics: Age: 10 years old; Brain development assessment: Weak logical reasoning ability, high memory ability score.

[0089] Learning objective: Improve logical reasoning ability

[0090] Recommended primary learning resource:

[0091] Course type: Elementary mathematics course. Step S30: The learning system inputs the first brain development assessment result and the first learning resource into the second model, and the second model generates cognitive ability test questions based on the first brain development assessment result and the first learning resource.

[0092] In this step, the cognitive ability test questions are used to measure whether the first brain development assessment results match the user's actual ability; the test questions are designed to be combined with the first learning resource content and integrated into the learning process, so as to achieve dynamic assessment of the user's cognitive ability while the user is learning; the second model includes a rule-based adaptive test question generation model or a deep learning-based dynamic assessment model.

[0093] The second model can be a rule-based adaptive test generation model. This model generates targeted cognitive ability tests based on the cognitive ability score range (such as memory, attention, logical reasoning ability, etc.) in the first brain development assessment results and the content characteristics of the first learning resource. The model measures whether the user's actual cognitive ability matches the assessment results by analyzing the user's accuracy, reaction time, and answering patterns. By combining the user's brain development assessment results, test questions are generated to evaluate the user's cognitive ability, verifying the match between the learning resource and the user's ability. Based on the content of the first learning resource, test questions are designed to be naturally embedded in the learning process, avoiding obvious assessment behaviors perceived by the user.

[0094] For example, if the first brain development assessment results show that the user's logical reasoning ability score is 60 / 100, and the first resource is a math lesson, the test questions generated by the model may be "find the increasing pattern in a series of numbers" or "determine whether two mathematical statements contradict each other". By analyzing the user's answer performance, it can be verified whether their logical reasoning ability meets the assessment expectations.

[0095] The second model can be a deep learning-based dynamic evaluation model. The model compares the user's real-time answer data (such as accuracy, reaction speed, error patterns, etc.) with a pre-trained ability evaluation network to dynamically generate subsequent test questions to more accurately verify the user's actual ability.

[0096] In one implementation, the dynamic assessment model may include a generator and a discriminator. In the dynamic assessment model, the generator generates personalized cognitive ability test questions, and the discriminator determines whether the generated test questions meet the user's ability characteristics and learning resource requirements, thereby achieving high-quality test question generation.

[0097] Generator:

[0098] The generator is a neural network that accepts the following input data:

[0099] User features (embedding vectors).

[0100] Learn resource features (embedding vectors).

[0101] Random noise vector (used to introduce test item diversity).

[0102] The generator uses a multi-layered neural network to map the input to a test item characteristic space, including the question content, question type, and difficulty level. For example:

[0103] Input: [User features + learning resource features + noise].

[0104] Output: [Problem Description: "Calculate the next number in the following sequence"; Question Type: Multiple Choice; Target Ability: Logical Reasoning; Difficulty: Medium].

[0105] The generator's output characteristics are used by the decoding module to generate specific test questions that users can answer.

[0106] Discriminator:

[0107] The discriminator is a binary classification neural network used to evaluate whether the generated test questions match the user's ability level and the requirements of the learning resources. Its inputs include:

[0108] Characteristics of the generated test questions (such as question type, difficulty, and target ability).

[0109] User's true characteristics (such as brain development assessment results).

[0110] Characteristics of learning resources (such as knowledge point categories and content difficulty).

[0111] The discriminator outputs a binary probability, indicating whether the generated test question meets the requirements, which can be determined through statistical features or manually.

[0112] 0: The test questions are not suitable for the user or learning resources.

[0113] 1. The test questions meet the requirements. When the user's first brain development assessment result shows that their language comprehension ability is 80 / 100, and the first learning resource is language and literature, the model initially generates test questions in the language and literature category, such as "Read a short passage and answer questions". If the user's accuracy rate is significantly higher than expected, the model will dynamically adjust and generate more complex test questions (such as "Analyze the causal relationship between sentences in a given paragraph") to further verify the accuracy of the assessment results.

[0114] The cognitive ability test questions generated by the second model are highly integrated with the first learning resource and have the same theme type. The learning system assesses the user's cognitive ability while providing teaching content, ensuring the continuity of the learning process and the authenticity of the test questions, and providing efficient support for subsequent learning resource recommendations and ability assessment corrections.

[0115] Step S40: The learning system inserts the cognitive ability test questions into the teaching process of the first learning resource.

[0116] Specifically, the learning system dynamically selects suitable teaching nodes as test insertion points based on monitoring the user's learning status (including learning duration, interactive operations, completion progress, etc.) of the first learning resource. For example, when the first learning resource is mathematical calculation skills training, the learning system inserts a logic reasoning question after the user completes a certain number of basic calculation tasks to assess the user's coherence of thought and comprehension ability.

[0117] Furthermore, the learning system can adjust the presentation of cognitive ability test questions based on their characteristics (such as difficulty, time required, and assessment objectives). For example, for questions that require a longer time commitment, the learning system can insert them at key stages or intervals within the teaching module; for simple, real-time assessment questions, the learning system can present them instantly as the user learns key knowledge points via pop-ups, thus capturing the user's cognitive performance in a timely manner during the teaching process.

[0118] Through the above insertion strategy, the learning system can achieve seamless integration of cognitive ability test questions and teaching content, ensuring the user's learning experience while dynamically assessing the user's cognitive level, and providing data support for the optimized delivery of subsequent learning resources.

[0119] In step S50, the learning system sends the user's age and the results of the cognitive ability test to the third model. The third model outputs a corrected assessment result of the user's brain development level. This corrected result is used to optimize the delivery of subsequent learning resources. The third model achieves a comprehensive assessment and correction of the user's cognitive abilities by integrating a rule-based model and a deep learning model.

[0120] The third model extracts features from the received test results of user age and cognitive ability, including indicators such as the user's answer accuracy, reaction time, error distribution, and answer logic.

[0121] For users of different age groups, the model normalizes the test results through preset feature weight rules, making the test results of different users comparable.

[0122] The third model, based on a pre-defined brain development assessment rule base, compares the user's test results with statistical data of users in the same age group to calculate the deviation between the user's current ability and the average brain development level. For example, if a 10-year-old user's memory score is significantly higher than the average level, but their attention score is lower, the rule model quantifies the deviation as an assessment correction factor. In this example, the memory score can be increased, while the attention score can be decreased.

[0123] In step S60, the learning system sends the brain development level assessment correction result to the first model, and the first model recommends a second learning resource based on the brain development level assessment correction result.

[0124] The first model receives the brain development assessment correction results, including the user's specific scores and deviation information on multiple cognitive ability indicators (such as memory ability, attention, logical reasoning ability, language comprehension ability, etc.).

[0125] The model's analytical correction results prioritize recommending learning resources that can improve the user's weak areas, while retaining learning content that can further enhance the user's strong areas.

[0126] The first model filters resources in the learning resource database and selects the learning resources most suitable for the user's current ability level by combining resource tags (such as knowledge point classification, difficulty level, interaction form, etc.) with the cognitive ability needs in the evaluation and correction results.

[0127] For example, if a user's logical reasoning ability score is 65 / 100 and their language comprehension ability is 90 / 100, the model prioritizes matching interactive learning resources that include logical reasoning training, while also adding an appropriate amount of language comprehension challenge content.

[0128] Furthermore, the first model dynamically adjusts the recommendation strategy based on historical learning data and current evaluation correction results, including a gradual increase in resource difficulty, a diversification of learning task types, and a focus on resource allocation.

[0129] For example, for users with weak logical reasoning skills and whose learning goals are in mathematics, the model prioritizes recommending basic mathematical problems related to logical reasoning, such as "predicting the pattern of a sequence" or "simple logical relationship diagrams," while appropriately adding highly engaging interactive content to enhance learning interest.

[0130] For any module structures not specifically defined in this invention, the existing technical descriptions shall prevail. The prior art mentioned in the foregoing background and specific embodiments sections can be considered part of this invention and used to understand the meaning of certain technical features or parameters.

Claims

1. A learning resource recommendation method based on artificial intelligence, characterized in that, The method includes the following steps: Step S10: The learning system obtains the user's age, and the learning system determines the first brain development level assessment result based on the user's age; Step S20: The learning system inputs the first brain development level assessment result into the first model, and the first model recommends the first learning resource; Step S30: The learning system inputs the first brain development assessment result and the first learning resource into the second model, and the second model generates cognitive ability test questions based on the first brain development assessment result and the first learning resource. The cognitive ability test questions are used to measure whether the results of the first brain development assessment are consistent with those obtained using... Matching the user's actual capabilities; The test questions are designed to be integrated with the content of the first learning resource and incorporated into the learning process, so as to achieve dynamic assessment of the user's cognitive ability while the user is learning; Step S40: The learning system inserts the cognitive ability test questions into the teaching process of the first learning resource; Step S50: The learning system sends the user's age and the results of the cognitive ability test to the third model, and the third model outputs the user's brain development assessment and correction results. The third model extracts features from the received test results of user age and cognitive ability, including the user's answer accuracy, reaction time, error distribution, and answer logic; For users of different age groups, the model normalizes the test results through preset feature weight rules to make the test results of different users comparable. The third model is based on a preset brain development assessment rule base. It compares the user's test results with the statistical data of users of the same age group and calculates the deviation between the user's current ability and the average brain development level. Step S60: The learning system sends the brain development level assessment correction result to the first model, and the first model recommends a second learning resource based on the brain development level assessment correction result. The first model receives the brain development assessment and correction results, including the user's specific scores and deviation information on multiple cognitive ability indicators; The second model is either a rule-based adaptive test item generation model or a deep learning-based dynamic evaluation model; The third model extracts features from the received user age and cognitive ability test results and normalizes the test results using a preset feature weighting rule. The model's analytical correction results use the score weights of each capability indicator to prioritize recommending learning resources that can improve the user's weak areas, while retaining learning content that can further enhance the user's strong capabilities. The first model filters resources in the learning resource database and selects the learning resources most suitable for the user's current ability level by combining resource tags with cognitive ability needs in the assessment and correction results. The first model dynamically adjusts its recommendation strategy based on historical learning data and current evaluation and correction results, including a gradual increase in resource difficulty, a diversification of learning task types, and a focus on resource allocation.

2. The learning resource recommendation method based on artificial intelligence according to claim 1, characterized in that, The first brain development assessment result is an average developmental level reference value derived from statistical data analysis by age distribution.

3. The learning resource recommendation method based on artificial intelligence according to claim 1, characterized in that, The first model is a rule-based model, a collaborative filtering model, or a neural network model.

4. The learning resource recommendation method based on artificial intelligence according to claim 3, characterized in that, The collaborative filtering model mines the learning resource preference data of users whose developmental deviation is within a preset value, and generates learning resource recommendations corresponding to the user's evaluation results.

5. The learning resource recommendation method based on artificial intelligence according to claim 1, characterized in that, The cognitive ability test questions are of the same type as the first learning resource topic.

6. The learning resource recommendation method based on artificial intelligence according to claim 1, characterized in that, The learning system dynamically selects suitable teaching nodes as test question insertion points based on monitoring the user's learning status of the first learning resource.

Images