Intelligent learning management system and method based on learning terminal

Abstract

The invention discloses an intelligent learning management system and method based on a learning terminal, and belongs to the field of learning management systems, and the system comprises a user identity authentication module which is used for receiving user login information and carrying out identity authentication; the learning record storage module is used for storing historical learning data of the user; the exercise generation module responds to the passing of the identity authentication; the exercise evaluation module is used for judging whether the learning mastering ability reaches the standard or not; the content linkage analysis module is used for calculating the linkage degree S; the review mandatory control module is used for mandatory the user to complete review according to the linkage degree S; the learning state monitoring module is used for collecting video streams of the face and limbs in the learning process of the user; and the non-language behavior analysis engine is used for identifying attention abnormal behaviors. According to the method and the device, the forced review strategy can be adjusted according to the relevance between the last learning content assessment result and the new and old learning contents, so that the final learning effect of the user reaches the expectation.

Description

Technical Field

[0001] The present invention relates to the field of learning management systems, and in particular to an intelligent learning management system and method based on a learning terminal. Background Art

[0002] Smart learning is a process in which learners access learning resources on demand, flexibly conduct learning activities, and rapidly build knowledge and interpersonal networks in a smart environment. Its ultimate goal is to develop learners' learning intelligence and innovation capabilities. Smart learning based on learning terminals combines smart devices with advanced technologies to create a personalized, efficient, immersive, and continuous learning environment. Its core lies in its learner-centric approach, leveraging terminal data to drive teaching decisions, enabling precise resource delivery, interactive collaboration, and real-time feedback, ultimately improving learning outcomes and innovation. Typical applications include tablet-based junior high school mathematics instruction and the smart campus cloud platform, demonstrating the profound transformation technology is bringing to education models.

[0003] Existing patent 202311490140.7 A learning management method and system based on a smart learning terminal, including: borrowing a smart learning terminal from a smart base station cabin through face recognition; collecting the user's operation and usage data based on the smart learning terminal to generate a database file; after the student returns the smart learning terminal, the smart base station cabin synchronizes the structured data in the database file to the central control service center through a data processing tool; the central control service center sends the synchronized data to the server, and the supervisor performs corresponding approval processing on the server and issues new tasks and content; the server synchronizes the requested structured data and unstructured data to the central control service center according to the user's needs, and the central control service center synchronizes and sends them to the smart learning terminal through a data processing tool.

[0004] The existing technology has the following problems: before starting a new round of learning, users do not conduct targeted assessments on the content of the previous learning, and thus do not know the user's ability to master the content of the previous learning. Many adjacent knowledge units are related, which means that if the knowledge learned last time is not mastered, the quality and efficiency of the new round of learning will be affected. The existing review mechanism is rigid and cannot dynamically adjust the forced review strategy according to the correlation between new and old knowledge, which makes the user's final learning effect unable to meet expectations. Therefore, those skilled in the art provide a smart learning management system and method based on a learning terminal to solve the problems raised in the above background technology. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent learning management system and method based on a learning terminal, which can conduct targeted assessments on the previous learning content before the user starts a new round of learning, and then adjust the forced review strategy based on the assessment results and the correlation between the new and old learning content, so that the user's final learning effect reaches the expected level, thereby solving the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A smart learning management system based on a learning terminal, comprising:

[0008] User identity authentication module, used to receive user login information and perform identity authentication;

[0009] The learning record storage module is used to store the user's historical learning data, which includes the learning content identifier, learning timestamp, and the length of time spent on each knowledge unit;

[0010] The exercise question generation module, in response to the identity authentication being passed, retrieves the latest learning record corresponding to the user, identifies at least three time-consuming knowledge units based on the user's stay time in each knowledge unit in the learning record, and extracts a set of exercise questions from the question bank corresponding to the time-consuming knowledge units;

[0011] An exercise evaluation module is used to receive the user's answer data to the exercise set in real time, calculate the real-time score, and determine whether the learning and mastering ability meets the standard based on the comparison between the learning time interval and the preset standard-reaching rules;

[0012] A content linkage analysis module is activated when the practice assessment module determines that the learning ability does not meet the standard, and is used to analyze the relevance between the user's last learning content and the new content to be learned, and calculate the linkage degree S;

[0013] The review force control module forces the user to complete the review of t1, t2 or t3 according to the linkage level S, where t1 < t2 < t3;

[0014] The learning status monitoring module includes a camera deployed on the learning terminal, which is used to collect video streams of the user's face and body during the learning process;

[0015] a non-verbal behavior analysis engine, receiving the video stream and identifying abnormal attention behaviors through a computer vision algorithm;

[0016] A learning anomaly weight calculation module calculates the learning anomaly weight value V based on attention abnormal behavior data;

[0017] The learning effect correction module corrects the learning effect rating according to the V value.

[0018] As a further solution of the present invention, the specific process of extracting the exercise set by the exercise generation module is as follows:

[0019] (1) Extract the dwell time Ti of each knowledge unit, i=1···n, where n is a positive integer;

[0020] (2) Calculate the average length of stay ;

[0021] (3) Screening Ti>k· The knowledge unit is regarded as a high-time-consuming knowledge unit, where k is the difficulty coefficient adjustment factor and k ≥ 1.5;

[0022] (4) Sort by the length of stay Ti from large to small, and select the first m time-consuming knowledge units as the selected units, where m is a positive integer and m ≥ 3;

[0023] (5) For each selected unit, randomly select q questions from the question bank to form a subset of practice questions, where q is a positive integer;

[0024] (6) Aggregate all subsets of exercises to form the final set of exercises.

[0025] As a further solution of the present invention: the preset achievement rules in the practice evaluation module are specifically:

[0026] Preset score threshold P th The functional relationship with the time interval Δt is:

[0027] ;

[0028] Among them, P max is the highest threshold value when Δt=0, 90%≤ and P max ≤100%, λ is the attenuation coefficient, and 0<λ<0.05, the unit of Δt is day;

[0029] If the user's real-time score P 实时 ≥P th , it means that the user's learning ability has reached the standard. If the user's real-time score P 实时 <P th , it means that the user's learning and mastering ability is not up to standard.

[0030] As a further solution of the present invention: the specific process of the content linkage analysis module calculating the linkage degree S is as follows:

[0031] (a) Construct a knowledge graph, where nodes represent knowledge units and edges represent prerequisite relationships;

[0032] (b) Locating the last learned content node set A and the to-be-learned content node set B;

[0033] (c) Calculate the set of shortest path lengths from A to B {L1, L2, ···, Lp};

[0034] (d) According to the formula Calculate the degree of linkage, where p is the number of elements in the above shortest path length set.

[0035] As a further solution of the present invention: the specific process of the forced review of the review forced control module is:

[0036] When S≤30%, forced review t1=5-10 minutes;

[0037] When 30%<S≤60%, forced review t2=11-20 minutes;

[0038] When S>60%, mandatory review t3=21-30 minutes;

[0039] Lock access to new content during review periods.

[0040] As a further aspect of the present invention, the non-verbal behavior analysis engine identifies the following abnormal attention behaviors:

[0041] (S1) The pupil diameter change rate is lower than the threshold or the eye movement frequency is lower than the preset value;

[0042] (S2) Blinking frequency is not within the preset range;

[0043] (S3) single fixation lasting more than 10 seconds without blinking;

[0044] (S4) The time spent looking away from the learning interface accounts for ≥30%;

[0045] (S5) The tension of the glabellar muscles decreases more than the baseline value or the trunk backward angle continues to increase;

[0046] (S6) The frequency of non-learning movements exceeds the standard (pen spinning > 5 times / min or leg shaking > 8 times / min);

[0047] (S7) Any combination of behaviors from (S1) to (S6) appears for 3 consecutive minutes.

[0048] As a further solution of the present invention: the learning abnormality weight calculation module calculates the V value by the formula, specifically:

[0049] ;

[0050] Where N is the number of abnormalities, T is the cumulative abnormal duration (minutes), and when V ≥ 0.7, it is determined to be a significant learning abnormality.

[0051] As a further solution of the present invention, the present invention further includes a feedback adjustment module for:

[0052] Predict the optimal learning time based on the historical V value sequence;

[0053] When the V value is ≥0.7 for three consecutive times, the single learning period is shortened by 10%;

[0054] Correlate abnormal behavior to error rate analysis of practice questions;

[0055] When V≥0.8, relaxation training is promoted; when 0.7≤V<0.8, the difficulty of the exercises is reduced.

[0056] As a further solution of the present invention: the specific process of the learning effect correction module correcting the learning effect rating according to the V value is as follows:

[0057] Generate learning diagnostic reports including heat maps of abnormal behavior;

[0058] When V≥0.8, learning is suspended and the preset fatigue relief program is started;

[0059] When 0.7≤V<0.8, dynamically adjust the cognitive load intensity of subsequent learning content;

[0060] Map the V value to a learning efficiency index of 0-100 and store it in the user portrait.

[0061] This application also discloses a smart learning management method based on a learning terminal, which adopts a smart learning management system based on a learning terminal, including the following steps:

[0062] Store users' historical learning data;

[0063] Receive user login information and perform authentication;

[0064] In response to the identity authentication being passed, the latest learning record corresponding to the user is retrieved, at least three time-consuming knowledge units are identified based on the user's stay time in each knowledge unit in the learning record, and a set of practice questions is extracted from the question bank corresponding to the time-consuming knowledge units;

[0065] Receive the user's answer data on the set of exercises in real time, calculate the real-time score, and determine whether the learning and mastering ability meets the standard based on the comparison between the learning time interval and the preset standard-reaching rules;

[0066] When the practice evaluation module determines that the learning ability does not meet the standard, it is activated to analyze the relevance between the user's last learning content and the new content to be learned, and calculate the linkage degree S;

[0067] According to the linkage level S, users are forced to complete the review of time t1, t2 or t3, where t1 < t2 < t3;

[0068] Through the camera deployed on the learning terminal, the video stream of the user's face and body is collected during the learning process;

[0069] receiving the video stream and identifying abnormal attention behavior using a computer vision algorithm;

[0070] Calculate the learning abnormality weight value V based on the attention abnormal behavior data;

[0071] Correct the learning effect rating according to the V value.

[0072] Compared with the prior art, the present invention has the following beneficial effects:

[0073] The present invention significantly improves review efficiency and avoids ineffective practice by dynamically identifying time-consuming knowledge units and generating targeted exercises; a time-attenuated threshold model is used to quantitatively apply the forgetting curve, making the judgment of mastery more in line with cognitive laws; an innovative three-level mandatory review mechanism with a linkage degree S is designed to effectively block the spread of knowledge gaps; a non-verbal behavior analysis engine is used to accurately detect multiple types of attention anomalies, and the number of anomalies and duration are combined to calculate the learning anomaly weight value V, breaking through the limitation of traditional learning systems that rely solely on answering data; when V ≥ 0.7, graded intervention (reducing difficulty / shortening duration / pausing learning) is automatically triggered, combined with the dynamic optimization of cognitive load in the feedback adjustment module to improve the learning efficiency index; and finally a closed-loop system of "exercise evaluation-content linkage-status monitoring-effect correction" is constructed, which has been verified by actual measurements to reduce the review time cost, reduce the ineffective learning time, and improve the accuracy of knowledge relevance recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 This is a structural block diagram of an intelligent learning management system based on learning terminals. DETAILED DESCRIPTION

[0075] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0076] As mentioned in the background technology of this application, research has found that before starting a new round of learning, users do not conduct targeted assessments on the content of the previous learning, and thus do not know the user's ability to control the content of the previous learning. Many adjacent knowledge units are related to each other, which means that if the knowledge learned last time is not mastered, it will affect the quality and efficiency of the new round of learning. The existing review mechanism is rigid and cannot dynamically adjust the forced review strategy according to the correlation between new and old knowledge, which makes the user's final learning effect unable to meet expectations, and there are certain defects.

[0077] In order to solve the above-mentioned defects, the present application discloses an intelligent learning management system and method based on a learning terminal, which can conduct targeted assessments on the previous learning content before the user starts a new round of learning, and then adjust the forced review strategy based on the assessment results and the correlation between the new and old learning content, so that the user's final learning effect can meet the expectations.

[0078] The following will describe in detail how the solution of this application solves the above technical problems with reference to the accompanying drawings.

[0079] See also Figure 1In an embodiment of the present invention, a smart learning management system based on a learning terminal includes: a user identity authentication module for receiving user login information and performing identity authentication; a learning record storage module for storing the user's historical learning data, the historical learning data including a learning content identifier, a learning timestamp, and a length of time spent on each knowledge unit; an exercise question generation module, in response to identity authentication being passed, retrieving the latest learning record corresponding to the user, identifying at least three time-consuming knowledge units based on the length of time the user spent on each knowledge unit in the learning record, and extracting an exercise question set from a question bank corresponding to the time-consuming knowledge unit; an exercise evaluation module for receiving the user's answer data on the exercise question set in real time, calculating a real-time score, and making a judgment based on a comparison between the learning time interval and a preset standard-reaching rule. Whether the learning and mastering ability meets the standards; the content linkage analysis module is activated when the practice assessment module determines that the learning and mastering ability does not meet the standards, and is used to analyze the correlation between the user's last learning content and the new content to be learned, and calculate the linkage degree S; the review force control module forces the user to complete the review of t1, t2 or t3 duration according to the linkage degree S, where t1 < t2 < t3; the learning status monitoring module includes a camera deployed on the learning terminal, which is used to collect video streams of the face and limbs during the user's learning process; the non-verbal behavior analysis engine receives the video stream and identifies abnormal attention behavior through computer vision algorithms; the learning abnormality weight calculation module calculates the learning abnormality weight value V based on the abnormal attention behavior data; the learning effect correction module corrects the learning effect rating according to the V value. This application constructs an intelligent learning management framework based on multi-module collaboration to achieve a closed loop of the entire process of "identity authentication → learning record analysis → dynamic exercise question generation → mastering ability assessment → content linkage decision-making → learning status monitoring".

[0080] In this embodiment, the specific process of extracting the exercise set by the exercise generation module is as follows: (1) extracting the dwell time Ti of each knowledge unit, i=1···n, where n is a positive integer; (2) calculating the average dwell time ;(3)Screen Ti>k· The knowledge units are regarded as high-time-consuming knowledge units, where k is the difficulty coefficient adjustment factor and k≥1.5; (4) Sort by the length of stay Ti from large to small, and select the first m high-time-consuming knowledge units as selected units, where m is a positive integer and m≥3; (5) For each selected unit, randomly select q questions from the question bank to form a subset of practice questions, where q is a positive integer; (6) Aggregate all the subsets of practice questions to form the final set of practice questions. Through the dynamic threshold (Ti>k· ) Accurately quantify "time-consuming knowledge units" to ensure that exercises are generated based on the user's real weaknesses. This setting solves the problem of traditional systems blindly extracting exercises.

[0081] In this embodiment, the preset standard-reaching rule in the practice evaluation module is specifically: the preset standard-reaching score threshold P th The functional relationship with the time interval Δt is: Among them, P max is the highest threshold value when Δt=0, 90%≤ and P max ≤100%, λ is the attenuation coefficient, and 0<λ<0.05, the unit of Δt is day; if the user's real-time score P 实时 ≥P th , it means that the user's learning ability has reached the standard. If the user's real-time score P 实时 <P th , it means that the user's learning ability is not up to standard. This setting uses an exponential decay function Dynamically adjust the achievement standards to comply with the law of the forgetting curve.

[0082] In this embodiment, the specific process of the content linkage analysis module to calculate the linkage degree S is as follows: (a) construct a knowledge graph, where nodes represent knowledge units and edges represent prerequisite relationships; (b) locate the last learned content node set A and the to-be-learned content node set B; (c) calculate the shortest path length set {L1, L2, ···, Lp} from A to B; (d) calculate the shortest path length set {L1, L2, ···, Lp} from A to B according to the formula Calculate the degree of linkage. This setting calculates the degree of linkage S based on the topological relationship of the knowledge graph (the inverse mean of the shortest path), achieving an objective measurement of content relevance and breaking through the limitations of traditional systems that rely solely on manual annotation of relevance.

[0083] In this embodiment, the forced review control module implements the following mandatory review process: when S ≤ 30%, mandatory review t1 = 5-10 minutes; when 30% < S ≤ 60%, mandatory review t2 = 11-20 minutes; when S > 60%, mandatory review t3 = 21-30 minutes; access to new content is blocked during the review period. This setting triggers differentiated mandatory review durations (t1 / t2 / t3) based on the degree of interaction S, balancing learning efficiency and knowledge coherence (longer review is required when S > 60%). The time threshold design (5-30 minutes) meets the attention span.

[0084] In this embodiment, the non-verbal behavior analysis engine identifies the following abnormal attention behaviors: (S1) The pupil diameter change rate is lower than the threshold or the eye movement frequency is lower than the preset value. The pupil tracking algorithm (such as Ellipse Fitting) is used to calculate the pupil diameter change rate per unit time. Under normal cognitive activities, the pupil will contract and expand periodically (the average change rate is ≥0.5mm / s). When the change rate is continuously <0.2mm / s, it is judged as "sluggish". The eye movement frequency is calculated by the displacement vector of the iris feature point. During normal reading, the eye scan frequency is 2-4 times / second. If it is <0.5 times / second, an alarm is triggered; (S2) The blinking frequency is not within the preset range. The normal physiological range is: 10-25 times / minute; (S3) A single gaze lasts for more than 10 seconds without blinking. If it lasts for 10 seconds, it is judged as "false concentration"; (S4) The time of sight away from the learning interface accounts for ≥30%. At this time, the knowledge absorption rate is greatly reduced; (S5) The tension of the glabellar muscle decreases more than the baseline value or the trunk backward angle continues to increase. Through FACS (Facial The system uses a motion coding system (SMC) to calculate the rate of change in glabellar wrinkle depth. During normal thinking, tension fluctuates within a ±15% range; a decrease >25% is considered relaxation. Skeletal key points detect the angle θ between the shoulder-hip line and the vertical axis. If θ consistently exceeds 15° and increases at a rate >0.5° / s, it is considered "slumped." Biomechanics (trunk angle) and micro-expression analysis (brow muscles) are integrated to verify negative states using dual signals. (S6) Excessive frequency of non-learning movements (e.g., pen spinning >5 times / minute or leg shaking >8 times / minute) and (S7) Three consecutive minutes of any combination of behaviors from (S1) to (S6) are present. This setup transforms subjective teaching experiences (e.g., "daze" or "drowsiness") into machine-detectable indicators. The combined conditional design of S7 significantly reduces the false alarm rate (abnormality must persist for three minutes).

[0085] In this embodiment, the learning anomaly weight calculation module calculates the V value through the formula, specifically: Where N is the number of abnormalities, and T is the cumulative abnormal duration (in minutes). A significant learning abnormality is determined when V ≥ 0.7. A weighted formula combines abnormality frequency and duration to output a standardized assessment value. The weight coefficient (0.4 / 0.6) reflects the cognitive principle that duration is more important than number of times. The threshold of V ≥ 0.7 provides a unified intervention trigger point.

[0086] This embodiment also includes a feedback adjustment module that: predicts the optimal study time based on historical V-value sequences; shortens a single study session by 10% when V is ≥ 0.7 three times in a row; correlates abnormal behavior with exercise error rate analysis; promotes relaxation training when V ≥ 0.8, and reduces exercise difficulty when 0.7 ≤ V < 0.8. This setting dynamically optimizes learning parameters (duration / difficulty) based on historical V-value data and correlates abnormal behavior with error analysis, forming a self-evolving cycle of "monitoring → diagnosis → adjustment." When V ≥ 0.7 three times in a row, the study time is shortened by 10% to prevent cumulative fatigue.

[0087] In this embodiment, the learning effect correction module modifies the learning effect rating based on the V value as follows: generating a learning diagnostic report including a heat map of abnormal behavior; pausing learning and initiating a preset fatigue reduction program when V ≥ 0.8; dynamically adjusting the cognitive load intensity of subsequent learning content when 0.7 ≤ V < 0.8; and mapping the V value to a learning efficiency index on a scale of 0-100 and storing it in the user profile. The V value is mapped to actionable instructions (pause learning / reduce load / generate report) and the learning efficiency index is output. Heat map reports visualize learning obstacles (improve user stickiness), while the learning efficiency index quantifies the user profile (laying the foundation for personalized recommendations).

[0088] The present application also discloses a smart learning management method based on a learning terminal, which adopts a smart learning management system based on a learning terminal, including the following steps: storing a user's historical learning data; receiving user login information and performing identity authentication; in response to successful identity authentication, retrieving the latest learning record corresponding to the user, identifying at least three time-consuming knowledge units based on the user's stay time in each knowledge unit in the learning record, and extracting a set of practice questions from the question bank corresponding to the time-consuming knowledge units; receiving the user's answer data to the practice question set in real time, calculating a real-time score, and judging whether the learning mastery ability meets the standard based on the comparison of the learning time interval and the preset standard-reaching rules; when the practice evaluation module determines that the learning mastery ability does not meet the standard, it is activated to analyze the correlation between the user's last learning content and the new content to be learned, and calculate the linkage degree S; forcing the user to complete a review of t1, t2, or t3 according to the linkage degree S, where t1 < t2 < t3; collecting a video stream of the user's face and body during the learning process through a camera deployed on the learning terminal; receiving the video stream, identifying abnormal attention behavior through a computer vision algorithm; calculating a learning abnormality weight value V based on the abnormal attention behavior data; and correcting the learning effect rating according to the V value.

[0089] The present invention significantly improves review efficiency and avoids ineffective practice by dynamically identifying time-consuming knowledge units and generating targeted exercises; a time-attenuated threshold model is used to quantitatively apply the forgetting curve, making the judgment of mastery more in line with cognitive laws; an innovative three-level mandatory review mechanism with a linkage degree S is designed to effectively block the spread of knowledge gaps; a non-verbal behavior analysis engine is used to accurately detect multiple types of attention anomalies, and the number of anomalies and duration are combined to calculate the learning anomaly weight value V, breaking through the limitation of traditional learning systems that rely solely on answering data; when V ≥ 0.7, graded intervention (reducing difficulty / shortening duration / pausing learning) is automatically triggered, combined with the dynamic optimization of cognitive load in the feedback adjustment module to improve the learning efficiency index; and finally a closed-loop system of "exercise evaluation-content linkage-status monitoring-effect correction" is constructed, which has been verified by actual measurements to reduce the review time cost, reduce the ineffective learning time, and improve the accuracy of knowledge relevance recognition.

[0090] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

[0091] The above is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solutions and inventive concepts of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A smart learning management system based on learning terminals, characterized by: include: User identity authentication module, used to receive user login information and perform identity authentication; The learning record storage module is used to store the user's historical learning data, which includes the learning content identifier, learning timestamp, and the length of time spent on each knowledge unit; The exercise question generation module, in response to the identity authentication being passed, retrieves the latest learning record corresponding to the user, identifies at least three time-consuming knowledge units based on the user's stay time in each knowledge unit in the learning record, and extracts a set of exercise questions from the question bank corresponding to the time-consuming knowledge units; An exercise evaluation module is used to receive the user's answer data to the exercise set in real time, calculate the real-time score, and determine whether the learning and mastering ability meets the standard based on the comparison between the learning time interval and the preset standard-reaching rules; A content linkage analysis module is activated when the practice assessment module determines that the learning ability does not meet the standard, and is used to analyze the relevance between the user's last learning content and the new content to be learned, and calculate the linkage degree S; The review force control module forces the user to complete the review of t1, t2 or t3 according to the linkage level S, where t1 < t2 < t3; The learning status monitoring module includes a camera deployed on the learning terminal, which is used to collect video streams of the user's face and body during the learning process; a non-verbal behavior analysis engine, receiving the video stream and identifying abnormal attention behaviors through a computer vision algorithm; A learning anomaly weight calculation module calculates the learning anomaly weight value V based on attention anomaly behavior data; The learning effect correction module corrects the learning effect rating according to the V value.

2. The intelligent learning management system based on a learning terminal according to claim 1 is characterized in that: The specific process of extracting the exercise set by the exercise generation module is as follows: (1) Extract the dwell time Ti of each knowledge unit, i=1···n, where n is a positive integer; (2) Calculate the average length of stay ; (3) Screening Ti>k· The knowledge unit is regarded as a high-time-consuming knowledge unit, where k is the difficulty coefficient adjustment factor and k ≥ 1.5; (4) Sort by the length of stay Ti from large to small, and select the first m time-consuming knowledge units as the selected units, where m is a positive integer and m ≥ 3; (5) For each selected unit, randomly select q questions from the question bank to form a subset of practice questions, where q is a positive integer; (6) Aggregate all subsets of exercises to form the final set of exercises.

3. The intelligent learning management system based on a learning terminal according to claim 2 is characterized in that: The preset achievement rules in the exercise assessment module are as follows: Preset score threshold P th The functional relationship with the time interval Δt is: ； Among them, P max is the highest threshold value when Δt=0, 90%≤ and P max ≤100%, λ is the attenuation coefficient, and 0<λ<0.05, the unit of Δt is day; If the user's real-time score P 实时 ≥P th , it means that the user's learning ability has reached the standard. If the user's real-time score P 实时 <P th , it means that the user's learning and mastering ability is not up to standard.

4. The intelligent learning management system based on a learning terminal according to claim 3 is characterized in that: The specific process of calculating the linkage degree S by the content linkage analysis module is as follows: (a) Construct a knowledge graph, where nodes represent knowledge units and edges represent prerequisite relationships; (b) Locating the last learned content node set A and the to-be-learned content node set B; (c) Calculate the set of shortest path lengths from A to B {L1, L2, ···, Lp}; (d) According to the formula Calculate the degree of linkage.

5. The intelligent learning management system based on a learning terminal according to claim 4 is characterized in that: The specific process of the forced review of the review forced control module is as follows: When S≤30%, forced review t1=5-10 minutes; When 30%<S≤60%, forced review t2=11-20 minutes; When S>60%, mandatory review t3=21-30 minutes; Lock access to new content during review periods.

6. The intelligent learning management system based on a learning terminal according to claim 5 is characterized in that: The nonverbal behavior analysis engine identifies the following attentional abnormalities: (S1) The pupil diameter change rate is lower than the threshold or the eye movement frequency is lower than the preset value; (S2) Blinking frequency is not within the preset range; (S3) single fixation lasting more than 10 seconds without blinking; (S4) The time spent looking away from the learning interface accounts for ≥30%; (S5) The tension of the glabellar muscles decreases more than the baseline value or the trunk backward angle continues to increase; (S6) The frequency of non-learning actions exceeds the standard; (S7) Any combination of behaviors from (S1) to (S6) appears for 3 consecutive minutes.

7. The intelligent learning management system based on a learning terminal according to claim 6 is characterized in that: The learning anomaly weight calculation module calculates the V value through the formula, specifically: ； Where N is the number of abnormalities, T is the cumulative abnormal duration, and when V ≥ 0.7, it is determined to be a significant learning abnormality.

8. The intelligent learning management system based on a learning terminal according to claim 7 is characterized in that: It also includes a feedback regulation module for: Predict the optimal learning time based on the historical V value sequence; When the V value is ≥0.7 for three consecutive times, the single learning period is shortened by 10%; Correlate abnormal behavior to error rate analysis of practice questions; When V≥0.8, relaxation training is promoted; when 0.7≤V<0.8, the difficulty of the exercises is reduced.

9. The intelligent learning management system based on a learning terminal according to claim 8, characterized in that: The specific process of the learning effect correction module correcting the learning effect rating according to the V value is as follows: Generate learning diagnostic reports including heat maps of abnormal behavior; When V≥0.8, learning is suspended and the preset fatigue relief program is started; When 0.7≤V<0.8, dynamically adjust the cognitive load intensity of subsequent learning content; Map the V value to a learning efficiency index of 0-100 and store it in the user portrait.

10. A smart learning management method based on a learning terminal, characterized in that: The intelligent learning management system based on the learning terminal according to any one of claims 1 to 9 is adopted, comprising the following steps: Store users' historical learning data; Receive user login information and perform authentication; In response to the identity authentication being passed, the latest learning record corresponding to the user is retrieved, at least three time-consuming knowledge units are identified based on the user's stay time in each knowledge unit in the learning record, and a set of practice questions is extracted from the question bank corresponding to the time-consuming knowledge units; Receive the user's answer data on the set of exercises in real time, calculate the real-time score, and determine whether the learning and mastering ability meets the standard based on the comparison between the learning time interval and the preset standard-reaching rules; When the practice evaluation module determines that the learning ability does not meet the standard, it is activated to analyze the relevance between the user's last learning content and the new content to be learned, and calculate the linkage degree S; According to the linkage level S, users are forced to complete the review of time t1, t2 or t3, where t1 < t2 < t3; Through the camera deployed on the learning terminal, the video stream of the user's face and body is collected during the learning process; receiving the video stream and identifying abnormal attention behavior using a computer vision algorithm; Calculate the learning abnormality weight value V based on the attention abnormal behavior data; Correct the learning effect rating according to the V value.

Images