Children English fuming content recommendation method and system based on multi-dimensional behavior analysis and forgetting curve

Abstract

The invention relates to the technical field of education, and discloses a children English smoked listening content recommendation method and system based on multi-dimensional behavior analysis and a forgetting curve, and the method comprises the following specific steps: S1, data collection: collecting passive behavior data of a user; according to the method, through multi-dimensional behavior analysis, passive behavior data (such as historical watching / listening content, single playing duration, repeated playing times and the like), active behavior data (such as advanced test results, user manual marking preferences and the like) and time sequence data (such as content initial contact time, last review time and the like) of a user are collected; according to the method, the learning habits, interests, preferences and English levels of each child can be deeply understood, and based on the data, the system can provide highly personalized fuming content recommendation for each child, so that a traditional'one-step 'recommendation mode is avoided, each child can receive the most suitable learning materials, and the learning effect is remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of educational technology, specifically to a method and system for recommending English listening content for children based on multi-dimensional behavioral analysis and forgetting curves. Background Technology

[0002] Traditional listening comprehension functions often simply repeat content based on a user's most recent viewing / listening history, such as looping yesterday's cartoon clips. This method has significant drawbacks: firstly, the monotonous content leads to a rapid decline in children's interest, making it difficult to maintain long-term learning motivation; secondly, it fails to incorporate scientific principles of memory (such as the forgetting curve) for optimization, resulting in a lack of scientific basis for the timing and frequency of repetition, leading to low memory retention rates. Furthermore, traditional recommendation systems have a single recommendation dimension, relying solely on basic user behavioral data (such as viewing history) and failing to integrate active learning behavior data (such as test results, interaction preferences, and learning progress) and individual difference information (such as memory ability and interest preferences), thus failing to achieve truly personalized recommendations.

[0003] While some systems attempt to introduce fixed-cycle content review mechanisms, such as repeating content every three days, this approach also has limitations. Fixed cycles cannot adapt to individual differences in memory abilities. For children with strong memories, the repetition cycle may be too long, leading to low learning efficiency; for children with weaker memories, the repetition cycle may be too short, resulting in an excessive learning burden. Furthermore, traditional systems do not dynamically link to children's actual learning progress (such as advancement test results) and cannot adjust recommendation strategies in a timely manner based on children's learning outcomes, resulting in a disconnect between recommended content and children's actual needs.

[0004] Therefore, existing technologies cannot effectively solve the problems of personalization and scientific rigor in recommending English listening content for children, and are insufficient to meet the actual needs of children's language learning. Based on the above problems, this invention proposes a method and system for recommending English listening content for children based on multi-dimensional behavioral analysis and the forgetting curve. It aims to achieve personalized and scientific content recommendation by expanding data collection dimensions, integrating active and passive behavioral data, dynamically adjusting the timing and frequency of content delivery based on the Ebbinghaus forgetting curve, and constructing a dynamically updated pool of listening content, thereby improving children's language learning efficiency and interest maintenance. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for recommending English listening content for children based on multi-dimensional behavioral analysis and forgetting curves, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for recommending English listening content for children based on multi-dimensional behavioral analysis and forgetting curves, comprising the following specific steps: S1. Data Acquisition Steps: Collect user passive behavior data, including but not limited to historical viewed / listened content, single playback duration, number of repeated playbacks, completion rate, and number of skips; Collect user-initiated behavioral data, including but not limited to advanced test results, user-manually labeled preferences, proactive search content, and interactive feedback; Collect time-series data, including but not limited to the time of first exposure to the content, the time of the last review, and the frequency of review; S2. Behavioral Analysis Steps: Based on proactive behavior data, we analyze users’ learning progress, knowledge mastery, and interest preferences, and mark content tags that need to be strengthened, interest weights, and learning difficulties. Based on passive behavior data, analyze users' content exposure frequency, preferences, and potential learning paths; S3. Steps for modeling memory decay: Based on the Ebbinghaus forgetting curve, and combined with individual test performance and learning progress, memory retention parameters are dynamically calibrated to construct a personalized forgetting curve model; through the formula... Recommendation priority = (Content difficulty, memory decay value, user interest weight, and relevance to learning progress) are used to calculate the push weight, where, A weighted function that comprehensively considers multiple factors; S4. Steps for building a dynamic content pool: The recommended content is divided into "Instant Review Zone", "Periodic Reinforcement Zone", "Interest Development Zone" and "Difficulty Breakthrough Zone" according to priority. The content pool composition is dynamically adjusted based on real-time behavioral data, learning progress, and changes in interests to achieve accurate content matching and updates. S5. Recommended generation steps: The optimal time to review content is calculated based on the memory decay model, and the timing of recommendations is adjusted in combination with the user's current behavior (such as active time and learning environment). The system matches content from the content pool that aligns with user interests, requires enhanced content tags, and is appropriate for learning difficulty, generating a personalized recommendation queue.

[0007] Preferably, the data acquisition step further includes: The client collects user behavior data through embedded sensors, user interface, and background logs, and uploads it to the server in encryption. The server receives and stores user behavior data, and performs data cleaning, preprocessing, and outlier detection to ensure data quality.

[0008] Preferably, the behavior analysis step further includes: Historical data is processed using a time decay algorithm, while recent behavioral data is given higher weight to more accurately reflect the user's current learning status and interest preferences. By combining machine learning algorithms (such as collaborative filtering and deep learning) to deeply mine user behavior data, potential learning needs and preference patterns can be discovered.

[0009] Preferably, the dynamic content pool construction step further includes: Based on the number of times, duration, and feedback from users who actively skip content, the priority of related content in the content pool is dynamically adjusted or the content is removed. A content quality assessment mechanism is introduced to dynamically rate content based on user feedback, expert review, and market performance, prioritizing and recommending high-quality content.

[0010] Another technical problem to be solved by this invention is to provide a system for recommending English listening content for children based on the multi-dimensional behavioral analysis and forgetting curve described above, comprising: Client: Used to collect user behavior data, including viewing / listening records, interaction feedback, etc., and upload it to the server; It provides a user interface that displays recommended content, learning reports, user settings options, and personalized learning suggestions.

[0011] Servers include: Data storage module: used to store user behavior data, memory retention parameters, content pool information, and user profiles; Behavioral analysis engine: Used to process historical data using time decay algorithms and machine learning algorithms to analyze users' learning progress, interests, preferences, and potential needs; Recommendation engine: Used to call memory models and content quality assessment mechanisms to generate content sequences, calculate recommendation priorities, and adjust recommendation timing based on the user's current behavior state; Content scheduler: Used to push content to the client according to priority, user's current behavior status and learning environment, so as to achieve precise content delivery.

[0012] Preferably, the client further includes: Intelligent reminder module: Based on the user's learning progress, memory decay model and current time, it intelligently generates learning reminders to ensure that the user reviews on time; Parental monitoring module: Provides a parent interface that displays the child's learning reports, behavioral data, and recommended content, supporting parents to remotely monitor and guide their child's learning.

[0013] Preferably, the server further includes: Expert review module: Used for professional review of content to ensure its accuracy, suitability, and educational value; Market Feedback Module: Used to collect user feedback, market performance, and competitor information to provide a basis for content updates and recommendation strategy adjustments.

[0014] This invention provides a method and system for recommending English listening content for children based on multi-dimensional behavioral analysis and forgetting curves. It has the following beneficial effects: (1) This invention uses multi-dimensional behavioral analysis, including collecting passive behavior data (such as historical viewing / listening content, single playback duration, number of repeated playbacks, etc.), active behavior data (such as advanced test results, user manual marking preferences, etc.), and time series data (such as the time of first exposure to content, the time of last review, etc.), to gain a deep understanding of each child's learning habits, interests, and English proficiency. Based on this data, the system can provide highly personalized listening content recommendations for each child, avoiding the traditional "one-size-fits-all" recommendation method, ensuring that each child can receive the most suitable learning materials for themselves, thereby significantly improving learning effectiveness.

[0015] (2) This invention combines the principle of the Ebbinghaus forgetting curve to scientifically plan the timing and frequency of content delivery. The forgetting curve reveals the pattern of human memory forgetting. The system dynamically calibrates memory retention parameters based on the child's individual test performance and learning progress, and constructs a personalized forgetting curve model. By reasonably arranging the review and delivery of content, the system can effectively consolidate the child's learning results, reduce forgetting, and improve learning efficiency. This scientific content delivery method is more in line with the child's learning needs than the traditional random or fixed frequency delivery method, and helps to improve language learning efficiency.

[0016] (3) This invention not only provides an advanced recommendation algorithm, but also constructs a complete system support platform, which includes a client and a server. The client is responsible for collecting user behavior data and displaying recommended content, while the server is responsible for data storage, behavior analysis, recommendation generation, and content scheduling. The system has functions such as user management, content management, and push management, and can record and analyze children's learning behavior data, providing a basis for continuous optimization of the recommendation algorithm. At the same time, the system also focuses on user experience, providing a simple and clear interface design and smooth operation process, as well as rich feedback mechanisms, so that children can understand their learning progress and results in a timely manner. This comprehensive system support makes this invention more stable and reliable in practical applications, providing children with a good learning experience. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the method steps of the present invention; Figure 2 This is a system architecture view of the present invention; Figure 3A schematic view of the dynamic content pool construction of this invention. Detailed Implementation

[0018] 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.

[0019] Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the invention, and should not be construed as limiting the invention.

[0020] A preferred embodiment of the children's English listening content recommendation method and system based on multi-dimensional behavioral analysis and forgetting curve provided by this invention is as follows: Figure 1-3 As shown: The method for recommending English listening content for children based on multi-dimensional behavioral analysis and the forgetting curve includes the following specific steps: S1. Data Acquisition Steps: Collect user passive behavior data, including but not limited to historical viewed / listened content, single playback duration, number of repeated playbacks, completion rate, and number of skips; Collect user-initiated behavioral data, including but not limited to advanced test results, user-manually labeled preferences, proactive search content, and interactive feedback; Collect time-series data, including but not limited to the time of first exposure to the content, the time of the last review, and the frequency of review; The data collection steps also include: The client collects user behavior data through embedded sensors, user interface, and background logs, and uploads it to the server in encryption. The server receives and stores user behavior data, and performs data cleaning, preprocessing and outlier detection to ensure data quality. S2. Behavioral Analysis Steps: Based on proactive behavior data, we analyze users’ learning progress, knowledge mastery, and interest preferences, and mark content tags that need to be strengthened, interest weights, and learning difficulties. Based on passive behavior data, analyze users' content exposure frequency, preferences, and potential learning paths; The behavioral analysis steps also include: Historical data is processed using a time decay algorithm, while recent behavioral data is given higher weight to more accurately reflect the user's current learning status and interest preferences. By combining machine learning algorithms (such as collaborative filtering and deep learning) to deeply mine user behavior data, potential learning needs and preference patterns can be discovered. S3. Steps for modeling memory decay: Based on the Ebbinghaus forgetting curve, and combined with individual test performance and learning progress, memory retention parameters are dynamically calibrated to construct a personalized forgetting curve model; through the formula... Recommendation priority = (Content difficulty, memory decay value, user interest weight, and relevance to learning progress) are used to calculate the push weight, where, A weighted function that comprehensively considers multiple factors; The steps for building a dynamic content pool also include: Based on the number of times, duration, and feedback from users who actively skip content, the priority of related content in the content pool is dynamically adjusted or the content is removed. Introduce a content quality assessment mechanism to dynamically rate content based on user feedback, expert review, and market performance, and prioritize the recommendation of high-quality content; S4. Steps for building a dynamic content pool: The recommended content is divided into "Instant Review Zone", "Periodic Reinforcement Zone", "Interest Development Zone" and "Difficulty Breakthrough Zone" according to priority. The content pool composition is dynamically adjusted based on real-time behavioral data, learning progress, and changes in interests to achieve accurate content matching and updates. S5. Recommended generation steps: The optimal time to review content is calculated based on the memory decay model, and the timing of recommendations is adjusted in combination with the user's current behavior (such as active time and learning environment). The system matches content from the content pool that aligns with user interests, requires enhanced content tags, and is appropriate for learning difficulty, generating a personalized recommendation queue.

[0021] Another technical problem this invention aims to solve is to provide a children's English listening content recommendation system based on multi-dimensional behavioral analysis and forgetting curves, including: Client: Used to collect user behavior data, including viewing / listening records, interaction feedback, etc., and upload it to the server; It provides a user interface that displays recommended content, learning reports, user settings options, and personalized learning suggestions.

[0022] The client also includes: Intelligent reminder module: Based on the user's learning progress, memory decay model and current time, it intelligently generates learning reminders to ensure that the user reviews on time; Parental monitoring module: Provides a parent interface that displays the child's learning reports, behavioral data, and recommended content, supporting parents to remotely monitor and guide their child's learning.

[0023] Servers include: Data storage module: used to store user behavior data, memory retention parameters, content pool information, and user profiles; Behavioral analysis engine: Used to process historical data using time decay algorithms and machine learning algorithms to analyze users' learning progress, interests, preferences, and potential needs; Recommendation engine: Used to call memory models and content quality assessment mechanisms to generate content sequences, calculate recommendation priorities, and adjust recommendation timing based on the user's current behavior state; Content scheduler: Used to push content to the client according to priority, user's current behavior status and learning environment, so as to achieve precise content delivery.

[0024] The server also includes: Expert review module: Used for professional review of content to ensure its accuracy, suitability, and educational value; Market Feedback Module: Used to collect user feedback, market performance, and competitor information to provide a basis for content updates and recommendation strategy adjustments.

[0025] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0026] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for recommending English listening content for children based on multi-dimensional behavioral analysis and forgetting curves, characterized in that, The specific steps include the following: S1. Data Acquisition Steps: Collect user passive behavior data, including but not limited to historical viewed / listened content, single playback duration, number of repeated playbacks, completion rate, and number of skips; Collect user-initiated behavioral data, including but not limited to advanced test results, user-manually labeled preferences, proactive search content, and interactive feedback; Collect time-series data, including but not limited to the time of first exposure to the content, the time of the last review, and the frequency of review; S2. Behavioral Analysis Steps: Based on proactive behavior data, we analyze users’ learning progress, knowledge mastery, and interest preferences, and mark content tags that need to be strengthened, interest weights, and learning difficulties. Based on passive behavior data, analyze users' content exposure frequency, preferences, and potential learning paths; S3. Steps for modeling memory decay: Based on the Ebbinghaus forgetting curve, and combined with individual test performance and learning progress, memory retention parameters are dynamically calibrated to construct a personalized forgetting curve model; through the formula... Recommendation priority = (Content difficulty, memory decay value, user interest weight, and relevance to learning progress) are used to calculate the push weight, where, A weighted function that comprehensively considers multiple factors; S4. Steps for building a dynamic content pool: The recommended content is divided into "Instant Review Zone", "Periodic Reinforcement Zone", "Interest Development Zone" and "Difficulty Breakthrough Zone" according to priority; The content pool composition is dynamically adjusted based on real-time behavioral data, learning progress, and changes in interests to achieve accurate content matching and updates. S5. Recommended generation steps: The optimal time to review content is calculated based on the memory decay model, and the timing of recommendations is adjusted in combination with the user's current behavior (such as active time and learning environment). The system matches content from the content pool that aligns with user interests, requires enhanced content tags, and is appropriate for learning difficulty, generating a personalized recommendation queue.

2. The method for recommending English listening content for children based on multi-dimensional behavioral analysis and forgetting curves according to claim 1, characterized in that: The data acquisition steps also include: The client collects user behavior data through embedded sensors, user interface, and background logs, and uploads it to the server in encryption. The server receives and stores user behavior data, and performs data cleaning, preprocessing, and outlier detection to ensure data quality.

3. The method for recommending English listening content for children based on multi-dimensional behavioral analysis and forgetting curves according to claim 1, characterized in that: The behavior analysis steps also include: Historical data is processed using a time decay algorithm, while recent behavioral data is given higher weight to more accurately reflect the user's current learning status and interest preferences. By combining machine learning algorithms (such as collaborative filtering and deep learning) to deeply mine user behavior data, potential learning needs and preference patterns can be discovered.

4. The method for recommending English listening content for children based on multi-dimensional behavioral analysis and forgetting curves according to claim 1, characterized in that: The dynamic content pool construction steps also include: Based on the number of times, duration, and feedback from users who actively skip content, the priority of related content in the content pool is dynamically adjusted or the content is removed. A content quality assessment mechanism is introduced to dynamically rate content based on user feedback, expert review, and market performance, prioritizing and recommending high-quality content.

5. A system for recommending English listening content for children based on multi-dimensional behavioral analysis and forgetting curves as described in claims 1-4, characterized in that, include: Client: Used to collect user behavior data, including viewing / listening records, interaction feedback, etc., and upload it to the server; It provides a user interface that displays recommended content, learning reports, user settings options, and personalized learning suggestions. Servers include: Data storage module: used to store user behavior data, memory retention parameters, content pool information, and user profiles; Behavioral analysis engine: Used to process historical data using time decay algorithms and machine learning algorithms to analyze users' learning progress, interests, preferences, and potential needs; Recommendation engine: Used to call memory models and content quality assessment mechanisms to generate content sequences, calculate recommendation priorities, and adjust recommendation timing based on the user's current behavior state; Content scheduler: Used to push content to the client according to priority, user's current behavior status and learning environment, so as to achieve precise content delivery.

6. The children's English listening comprehension content recommendation system based on multi-dimensional behavioral analysis and forgetting curve as described in claim 5, characterized in that: The client also includes: Intelligent reminder module: Based on the user's learning progress, memory decay model and current time, it intelligently generates learning reminders to ensure that the user reviews on time; Parental monitoring module: Provides a parent interface that displays the child's learning reports, behavioral data, and recommended content, supporting parents to remotely monitor and guide their child's learning.

7. The children's English listening comprehension content recommendation system based on multi-dimensional behavioral analysis and forgetting curve as described in claim 5, characterized in that: The server also includes: Expert review module: Used for professional review of content to ensure its accuracy, suitability, and educational value; Market Feedback Module: Used to collect user feedback, market performance, and competitor information to provide a basis for content updates and recommendation strategy adjustments.

Images