Learning path planning methods, devices, storage media and electronic equipment
By mapping target audiences, professions, and courses to a two-dimensional competency space, using a large language model to determine the coupling category of teaching methods, calculating the cognitive and coupling gaps, and recommending suitable decoupling or practical courses, the problem of assessment bias in existing technologies is solved, and personalized learning path planning is realized.
Patent Information
- Application Number
- CN202611054611.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-15
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies cannot effectively identify and assess the coupling structure between cognitive judgment and real-time sensorimotor feedback in tightly coupled occupations, leading to biased assessment of ability gaps and an inability to identify differences in teaching methods, resulting in the recommendation of inappropriate courses.
The target audience, profession, and courses are mapped to a two-dimensional capability space consisting of cognitive and coupling axes. The coupling category of teaching methods is determined by a large language model, the capability gap is calculated using cognitive and coupling vectors, and decoupling or practical courses are recommended.
It enables accurate competency assessment and personalized learning path planning for closely coupled professions, ensuring that course recommendations meet the actual needs of the target audience and improving the effectiveness of the recommendations.
Smart Images

Figure CN123117583A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of artificial intelligence technology, specifically to a learning path planning method, apparatus, storage medium, and electronic device. Background Technology
[0002] With industrial upgrading and the widespread application of artificial intelligence technology, the structural changes in the labor market are accelerating, leading to a growing demand for career development path planning and retraining among target groups. Competency development recommendation systems aim to recommend suitable careers and training courses to target individuals, helping them achieve their career goals. Existing technologies mainly include the following types of solutions.
[0003] The first type of approach is a recommendation method based on keyword and skill tag matching. This method performs string matching or TF-IDF similarity calculation between job descriptions and skill tags in the target candidate's resume, and recommends jobs and courses based on the matching degree. This method is simple to implement and computationally efficient, and was widely used in early recruitment platforms and online education platforms.
[0004] The second approach is a recommendation method based on cosine similarity of ability magnitudes. This method represents both occupations and target objects as multi-dimensional ability vectors, calculates the cosine similarity or Euclidean distance between the target object's ability vector and the occupation's ability requirement vector, recommends occupations based on similarity ranking, and recommends supplementary courses based on ability gaps. Compared to the first approach, this method introduces an ability-dimensional structure, resulting in improved recommendation accuracy.
[0005] The third type of approach is a knowledge graph-based recommendation method. This method constructs a career-skill-course knowledge graph and generates recommendations through graph embedding or path reasoning. While this method can leverage the topological structure of domain knowledge, it is costly in constructing and maintaining large-scale knowledge graphs.
[0006] However, the inventors of this disclosure, through long-term research, have discovered the following technical defects in the aforementioned prior art. First, the target audience, occupation, and course are not in the same capability space. In the prior art, occupational capability requirements are derived from capability profiles in occupational databases, target audience capabilities are derived from assessments or resume mapping, and course capability training objectives are derived from teaching syllabi or manual annotation. The definitions, dimensions, and reference systems of capability dimensions among the three are often inconsistent, resulting in a lack of comparable basis for calculating capability gaps. For example, problem-solving ability in occupational requirements and problem-solving ability in course training objectives may refer to different capability constructs, and the gap value obtained by directly subtracting them lacks physical meaning.
[0007] Second, existing technologies treat competence as an additive set of scalar quantities, neglecting the coupling structure between cognitive judgment and real-time sensorimotor feedback. For tightly coupled professions (such as surgery, precision machining, physical therapy, first aid, and live performance), professional competence is not a simple addition of cognitive and motor abilities, but rather is constituted and executed through a two-way interaction of cognitive judgment with real-time sensorimotor feedback from the environment. Existing recommendation methods fail to recognize this coupling structure, leading to systematic biases in the assessment of competence gaps in tightly coupled professions, and consequently recommending inappropriate courses—for example, recommending surgical lectures to surgeons instead of surgical internships, rendering the recommendations ineffective.
[0008] Third, existing technologies do not differentiate between courses that impart knowledge through lectures and those that build coupled capabilities that can be executed in real-time through physical and sensory feedback. This leads to the treatment of courses with the same theme but different teaching methods as the same, equating lectures on surgery with surgical internships. This results in the inability to identify gaps in specific skills (such as real-time eye-hand manipulation during surgery), which can only be closed through hands-on learning while lecture-based courses cannot, rendering recommendations ineffective. Summary of the Invention
[0009] In view of this, embodiments of the present disclosure provide a learning path planning method, apparatus, storage medium, and electronic device.
[0010] In a first aspect, this disclosure provides a learning path planning method, the method comprising the following steps: Obtain the target audience, target occupation, candidate courses, and pedagogical coupling categories of the candidate courses for the learning path to be planned; The common ability representations of the target object, target occupation, and candidate courses are determined respectively. These common ability representations include cognitive vectors and coupling vectors. The cognitive vectors of the target object, target occupation, and candidate courses respectively represent the cognitive level already possessed by the target object in a predefined cognitive dimension, the demand level of the target occupation in the predefined cognitive dimension, and the training level of the candidate courses in the predefined cognitive dimension. The coupling vectors of the target object, target occupation, and candidate courses respectively represent the coupling ability of the target object, the degree to which the target occupation requires cognitive judgment to be formed instantly through real-time sensorimotor feedback, and the degree to which the candidate courses can cultivate cognitive judgment to be formed instantly through real-time sensorimotor feedback. Based on the cognitive vector in the common ability representation, the cognitive gap between the target object and the target occupation is determined, and based on the coupling vector in the common ability representation, the coupling gap between the target object and the target occupation in different coupling channels is determined; If the cognitive gap is greater than the set cognitive threshold, then the candidate courses with the teaching method coupling category of decoupling are determined as the target courses for the target object on the learning path. If the coupling gap corresponding to at least one channel is greater than a set coupling threshold, then the candidate course whose teaching method coupling category is practical and whose dominant coupling channel is the same as the at least one coupling channel is determined as the target course of the target object on the learning path.
[0011] Furthermore, the coupling vector includes coupling strength, channel weight distribution, and bound capability dimension, wherein the bound capability dimension includes a pair of bound dominant cognitive dimension and dominant sensorimotor dimension.
[0012] Furthermore, the channel weight distribution of the target object includes channel weights on multiple predefined coupled channels, and the channel weights include the current coupling performance level of the target object on the corresponding channel.
[0013] Furthermore, the current coupling performance level of the target object is obtained through perturbation-adaptation paradigm measurement; wherein, during the perturbation-adaptation paradigm measurement process, the learnability coefficient of the target object is also measured, and the learnability coefficient is used to characterize the speed at which the target object recovers to a stable coupling state after being perturbed.
[0014] Furthermore, the coupling vector of the candidate course is determined based on the pedagogical coupling category of the candidate course.
[0015] Furthermore, the method also includes: Obtain multiple common capability representations of the target object obtained from multiple tests; Obtain the outer edge information of the point cloud in the common capability representation vector space of the multiple common capability representations; The predictive gain of the target course is determined based on the learnability coefficient of the target object corresponding to at least one of the target courses and the common ability representation. Based on the predicted gain of at least one of the target courses and the outer edge information of the point cloud, the improvement effect of the target object after learning the target course is obtained; the improvement effect includes the updated common ability representation of the target object after learning the target course, as well as the cognitive gap and coupling gap with the target occupation.
[0016] Furthermore, the method also includes: The system obtains the interest vectors of the target object and multiple target occupations in a set interest domain, and obtains the occupational exposure of the target occupations; the occupational exposure indicates whether the target occupation belongs to a high-substitution-risk occupation that is easily replaced by software artificial intelligence and embodied robots and is weakly coupled. Based on the interest vector, determine the interest matching results between the target object and the multiple target occupations; Based on the cognitive gap and coupling gap between the target professions and the target object, it is determined whether the target object has the capability to achieve the target professions. Based on the occupational exposure, interest matching results, and ability attainment results, a first set of occupations that match the target object's interests and are attainable but belong to high-substitution-risk occupations are selected from the multiple target occupations. A second set of occupations that match the target object's interests but are not attainable and do not belong to high-substitution-risk occupations are selected.
[0017] Secondly, embodiments of this disclosure provide a learning path planning apparatus, comprising: The acquisition module is configured to acquire the target object, target occupation, candidate courses, and pedagogical coupling category of the candidate courses for the learning path to be planned; The first determining module is configured to determine the common ability representations of the target object, the target occupation, and the candidate courses, respectively. The common ability representations include cognitive vectors and coupling vectors. The cognitive vectors of the target object, the target occupation, and the candidate courses respectively represent the cognitive level already possessed by the target object in a predefined cognitive dimension, the demand level of the target occupation in the predefined cognitive dimension, and the training level of the candidate courses in the predefined cognitive dimension. The coupling vectors of the target object, the target occupation, and the candidate courses respectively represent the coupling ability of the target object, the degree to which the target occupation requires cognitive judgments to be formed instantly through real-time sensorimotor feedback, and the degree to which the candidate courses can train cognitive judgments to be formed instantly through real-time sensorimotor feedback. The second determining module is configured to determine the cognitive gap between the target object and the target occupation based on the cognitive vector in the common ability representation, and to determine the coupling gap between the target object and the target occupation in different coupling channels based on the coupling vector in the common ability representation; the third determining module is configured to determine the candidate courses with the teaching method coupling category of decoupling as the target courses for the target object on the learning path if the cognitive gap is greater than a set cognitive threshold. The fourth determining module is configured to determine the candidate course of the target object on the learning path if the coupling gap corresponding to at least one channel is greater than a set coupling threshold.
[0018] Thirdly, embodiments of this disclosure provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods described in any of the above aspects.
[0019] Fourthly, embodiments of this disclosure provide an electronic device including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the method described in any of the above aspects.
[0020] The above-disclosed scheme maps the three aspects of people, occupations, and courses into a two-dimensional ability space composed of cognitive and coupling axes. It classifies courses according to the coupling attributes of teaching methods and generates personalized learning path planning schemes based on the ability gaps in channel decomposition.
[0021] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description
[0022] The above and other objects, features, and advantages of this disclosure will become clearer from the following description of embodiments of this disclosure with reference to the accompanying drawings, in which: Figure 1 A flowchart illustrating the learning path planning method proposed in an embodiment of this disclosure is shown. Figure 2 A structural block diagram of a learning path planning apparatus according to an embodiment of the present disclosure is shown; Figure 3 A simplified block diagram is shown for an electronic device 300 suitable for implementing embodiments of the present disclosure. Detailed Implementation
[0023] The present disclosure is described below based on embodiments, but it is not limited to these embodiments. In the detailed description of the present disclosure below, certain specific details are described in detail. However, those skilled in the art will fully understand the present disclosure without these details. To avoid obscuring the substance of the present disclosure, well-known methods, processes, flows, elements, and circuits are not described in detail.
[0024] Furthermore, those skilled in the art should understand that the accompanying drawings provided herein are for illustrative purposes only and are not necessarily drawn to scale.
[0025] Unless the context explicitly requires it, words such as "including" or "contains" throughout the application should be interpreted as including rather than exclusive or exhaustive; that is, meaning "including but not limited to".
[0026] In the description of this disclosure, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this disclosure, unless otherwise stated, "a plurality of" means two or more.
[0027] Based on the problems described in the background art, this disclosure provides a learning path planning method. This method achieves target object-career-course ability development through coupled channel routing, thereby generating a learning planning path for the target object. See also... Figure 1 As shown, the method includes the following steps: Step S1 involves acquiring the target audience, target occupation, and candidate courses. For each candidate course, the large language model pre-determines the pedagogical coupling category based on its textual description. In some embodiments, the pedagogical coupling category includes decoupling courses and practical courses. Decoupling courses are those primarily based on lectures or theoretical instruction, without establishing coupling capabilities through practical application. Practical courses establish coupling capabilities through real-time bodily-sensory feedback via hands-on practice, clinical settings, workshops, or on-site experience. The large language model also assigns a dominant coupling channel to practical courses. In some embodiments, the "LLM determination method requiring rule constraints" can be used to determine the pedagogical coupling category. The criterion is the pedagogy, not the course topic.
[0028] In some embodiments, the pedagogical coupling category is determined by a large language model based on the text description of candidate courses according to determination rules; the determination rules include: using whether students repeatedly practice through real-time physical or sensory feedback as the determination criterion rather than the course theme; operating a computer, keyboard input, or programming does not constitute sensorimotor coupling and is determined to be decoupled; courses with the same theme but different teaching methods are determined to be different pedagogical coupling categories.
[0029] For example, it would be as follows: Rule ① (Core Criterion): The criterion is "whether students practice repeatedly through real-time physical or sensory feedback," not the course title or theme. Distinguish between decoupling courses (primarily lecture-based or theoretical instruction, without establishing coupling ability through practical operation) and hands-on courses (establishing coupling ability through hands-on practice, clinical practice, workshops, or on-site practice, with real-time physical-sensory feedback).
[0030] Rule ② (Exclusion of Counterexamples): Operating a computer, keyboard input, or programming does not constitute sensorimotor coupling, and therefore all courses are judged as decoupled courses.
[0031] Rule ③ (Same Name, Different Structures): Courses with the same theme but different teaching methods are classified into different categories. For example, a clinical kinesiology lecture is classified as decoupled (coupling strength is 0); dance training is classified as practical (coupling strength is 3, with proprioception and whole-body posture as the dominant coupling channels).
[0032] For example, while lectures on surgery and surgical internships cover the same topic, lectures do not require students to practice through real-time bodily feedback, and are therefore classified as decoupled. Surgical internships, on the other hand, require students to perform real-time procedures on the operating table, and are therefore classified as hands-on, with a vision-fine manipulation dominant coupling channel. Operating a computer, keyboard input, or programming does not constitute sensorimotor coupling, and is therefore classified as decoupled.
[0033] Step S2 involves decomposing the target object, target occupation, and candidate courses to be evaluated into common ability representations. These common ability representations include cognitive vectors and coupling vectors. The cognitive vectors of the target object, target occupation, and candidate courses respectively represent the target object's existing cognitive level in a predefined cognitive dimension, the target occupation's demand level in the predefined cognitive dimension, and the candidate courses' training level in the predefined cognitive dimension. The coupling vectors of the target object, target occupation, and candidate courses respectively represent the target object's coupling ability, the degree to which the target occupation requires cognitive judgments to be formed instantly through real-time sensorimotor feedback, and the degree to which the candidate courses can train cognitive judgments to be formed instantly through real-time sensorimotor feedback. The predefined cognitive dimensions include at least one of the following: verbal, fluid reasoning, quantity, spatial, working memory, speed attention, and creativity.
[0034] In some embodiments, coupling can characterize the bidirectional, instantaneous constitutive relationship between cognitive judgment and real-time sensorimotor feedback, i.e., cognitive judgment is not pre-completed or performed independently of the body, but rather relies on real-time sensorimotor feedback continuously acquired during task execution for online calibration and dynamic generation.
[0035] In some embodiments, the coupling capability can be characterized using coupling strength, channel weight distribution, and the bound capability dimension; therefore, the coupling vector includes coupling strength, channel weight distribution, and the bound capability dimension.
[0036] For the target occupation, the coupling strength characterizes the degree to which cognitive judgments for the corresponding task are formed instantly through real-time sensorimotor feedback; for the target object, the coupling strength represents the level of coupling ability already possessed by the target object; for the candidate course, the coupling strength represents the coupling training strength that the candidate course can establish. In some embodiments, the coupling strength can range from 0 to 3, where 0 represents that cognition and the body / senses occur separately at different times, with no coupling, and 3 represents that cognitive judgments are bidirectionally formed by real-time sensorimotor feedback, inseparable, and tightly coupled.
[0037] The channel weight distribution represents the channel weights of coupling capability across multiple predefined coupling channels. For a target profession, the channel weight can be understood as the coupling requirement weight of the target profession across multiple predefined coupling channels. For a target object, the channel weight can be understood as the normalized weight of the coupling performance level of the target object across multiple predefined coupling channels. For a candidate course, the channel weight can be understood as the coupling training weight of the candidate course across multiple predefined coupling channels.
[0038] In some embodiments, the predefined coupling channel includes multiple predefined sensory-motor channels, which include at least one of the following: visual-fine manipulation, tactile manipulation, proprioception and full-body posture, auditory judgment, real-time interpersonal interaction, and multi-sensory environmental monitoring. The different channels are distinguished based on the dominant sensorimotor mode that provides real-time feedback. For example, real-time hand-eye correction corresponds to visual-fine manipulation; force or texture feedback corresponds to tactile manipulation; body posture corresponds to proprioception and full-body posture; real-time sound judgment corresponds to auditory judgment; real-time response to people corresponds to real-time interpersonal interaction; and simultaneous monitoring of multiple environmental signals corresponds to multi-sensory environmental monitoring.
[0039] In some embodiments, the bound capability dimension includes a pair of bound dominant cognitive dimensions and a dominant sensorimotor dimension. The bound dominant cognitive dimension and dominant sensorimotor dimension are respectively the predetermined cognitive dimension and predetermined sensorimotor dimension corresponding to when cognitive judgment can be instantly constituted through real-time sensorimotor feedback; the dominant cognitive dimension is one of a plurality of predefined cognitive dimensions, and the dominant sensorimotor dimension is one of a plurality of predefined sensorimotor dimensions, the predefined sensorimotor dimensions including at least one of the following: fine motor skills, control precision, reaction and multi-limb coordination, strength, endurance and flexibility, gross coordination and balance, near vision, far vision, and hearing.
[0040] In some embodiments, for the target occupation, the bound dominant cognitive dimension and dominant sensorimotor dimension are the predetermined cognitive dimension and predetermined sensorimotor dimension bound in the task of the target occupation; for the target object, the bound dominant cognitive dimension and dominant sensorimotor dimension are the predetermined cognitive dimension and predetermined sensorimotor dimension that have actually been coupled in the coupling assessment of the target object; for the candidate course, the bound dominant cognitive dimension and dominant sensorimotor dimension are the predetermined cognitive dimension and predetermined sensorimotor dimension targeted in the coupling training.
[0041] The following example illustrates that the common ability representation of the target object, profession, and course is isomorphic, consisting of 14 components: p = [ C1…C7 (7 cognitive components)‖ s (coupling strength)‖ w1…w6 (6 channel weight components) ]. Here, the cognitive vector corresponds to 7 components, and the coupling vector is a concatenation of 1 coupling strength component and 6 channel weight components.
[0042] The following diagram illustrates the physical meaning of each part of the common abilities of the target audience, profession, and course:
[0043] The following describes how the target object, target occupation, and candidate courses are broken down into common capability representations.
[0044] (I) Occupational side Obtain multiple task-level natural language description texts of the target profession (e.g.) (Task statements or main duties of the Canadian NOC). Using a large language model, the task-level natural language description text of the target occupation is analyzed to extract its cognitive level values on a predefined cognitive dimension, obtaining the occupational cognitive component. For each task-level natural language description text, multiple large language models are integrated to output coupling strength, channel weight distribution, and bound capability dimension, obtaining the occupational coupling vector.
[0045] It should be noted that multiple task-level natural language description texts corresponding to the target occupation are input into multiple large language models. The integrated output of these multiple large language models includes cognitive vectors, coupling strength, channel weight distribution, and bound capability dimensions. In some embodiments, each of the multiple large language models obtains its own output according to a predefined scoring specification. Then, the outputs of the large language models are integrated using a weighted fusion method to obtain the final output. Specifically, for a pair of bound capability dimensions, namely the bound dominant cognitive dimension and dominant sensorimotor dimension, each language model can directly identify and determine them according to the corresponding content in the predefined scoring specification. The final determination result is the integrated result of the multiple language models. For example, the large language model can directly determine which of the multiple predefined cognitive dimensions is bound to which of the multiple predefined sensorimotor dimensions in real time within the real-time coupling loop of the task (i.e., the coupling of cognitive judgment and real-time sensorimotor function required for the target occupation under this task). The cognitive judgment participating in the task immediately constitutes this effect through real-time sensorimotor feedback.
[0046] The following section defines coupling strength, channel weight distribution, and the binding capability dimension.
[0047] (a) Coupling strength, used to characterize the degree to which cognitive judgments for a given task are formed instantly through real-time sensorimotor feedback, can take integer values from 0 to 3. A coupling strength of 0 indicates no coupling, meaning the cognitive judgments for that task are not formed instantly through real-time sensorimotor feedback; that is, no cognitive dimension of the target occupation under that task is formed instantly through any real-time sensorimotor feedback. If the coupling strength is the highest, such as 3, then the cognitive judgments for that task are formed instantly through real-time sensorimotor feedback; that is, all cognitive dimensions of the target occupation under that task are formed instantly through the corresponding real-time sensorimotor feedback. In other words, coupling strength is determined using a "real-time bidirectional integration" approach, rather than a demand co-occurrence approach: only when cognitive judgments rely on feedback from any dimension such as the body or senses to be formed or corrected instantly is it considered high coupling strength (coupling strength = 3, such as surgical suturing or troubleshooting living circuits); if the cognitive judgments for the target occupation under that task and the demand for any sensorimotor dimension such as physical strength or senses both exist but occur at different times (such as designing before delivering code, moving building materials during the day, and studying construction drawings at night), it is considered low coupling strength or no coupling. It should be noted that when the coupling strength is zero, the channel weight distribution is empty.
[0048] (b) Channel weight distribution, including the channel weights corresponding to the task on a predefined plurality of coupled channels. In some embodiments, the channel weight distribution includes the channel weights of one dominant coupled channel and at most two secondary channels selected from the predefined plurality of coupled channels. The channel weights of each channel in the channel weight distribution are located in the interval [0,1], and the sum of the channel weights is 1. In some embodiments, the predefined coupled channels include six channels: visual-fine manipulation, tactile manipulation, proprioception and full-body posture, auditory judgment, real-time interpersonal interaction, and multi-sensory environment monitoring. The distinction between different channels is based on the dominant modality (visual, auditory, tactile, interpersonal, proprioception, etc.) that provides real-time feedback. For example, real-time hand-eye correction is visual-fine manipulation; force or texture feedback is tactile manipulation; body posture is proprioception and full-body posture; real-time sound judgment is auditory judgment; real-time reaction to the target object is real-time interpersonal interaction; and simultaneous monitoring of multiple environmental signals is multi-sensory environment monitoring. Channel weights are determined by integrating multiple large language models according to the corresponding content in the predefined scoring specifications: which sensorimotor dimension constitutes the cognitive judgment of the task in real time; the dominant coupled channel corresponding to the dominant sensorimotor dimension is given the highest weight, and the secondary channel corresponding to the secondary sensorimotor dimension is given a secondary weight.
[0049] (c) A pair of bound dominant cognitive dimensions and dominant sensorimotor dimensions. The bound dominant cognitive dimension and dominant sensorimotor dimension are the predetermined cognitive dimension and predetermined sensorimotor dimension corresponding to the cognitive judgment of the corresponding task, respectively, when the cognitive judgment can be instantly constructed through real-time sensorimotor feedback. Specifically, the bound sensorimotor dimension is the sensory or motor pathway that provides real-time feedback, and the bound cognitive dimension is the type of cognitive judgment that is instantly constructed or modified by that real-time feedback. In some embodiments, the dominant cognitive dimension may be selected from seven cognitive dimensions, including verbal, fluid reasoning, quantity, spatial, working memory, speed attention, and creativity; the dominant sensorimotor dimension may be selected from sensorimotor dimensions such as fine motor skills, control precision, reaction and multi-limb coordination, strength, endurance and flexibility, gross coordination and balance, near vision, far vision, and hearing. For example, the bound dominant cognitive dimension for surgical suturing is spatial, and the bound dominant sensorimotor dimension is fine manipulation; the bound dominant cognitive dimension for on-site circuit troubleshooting is fluid reasoning, and the bound dominant sensorimotor dimension is fine manipulation. In some embodiments, the multiple large language models can also integrate and output judgment reasons for subsequent verification.
[0050] It should be noted that the predefined scoring specifications in this disclosure include a set of cognitive dimensions, a set of sensorimotor dimensions, a set of channels, coupling strength anchor points, judgment criteria, and output patterns. These can be designed and fixed in advance according to actual needs, and the large language model cannot be changed.
[0051] (ii) Target Object Obtain textual descriptions of the target object, including its profile, historical assessment results, or perturbation-adaptation paradigm measurement data. Utilize a large language model to parse these textual descriptions, extracting the target object's cognitive level across predefined cognitive dimensions to obtain cognitive components. The cognitive vector is composed of the cognitive components corresponding to the predefined cognitive dimensions. The definition and acquisition methods of the target object's cognitive vector and bound ability dimensions are similar to those of the target occupation; specific details can be found in the description of the target occupation and will not be repeated here.
[0052] In some embodiments, the coupling vector of the target object is measured using a perturbation-adaptation paradigm. This measurement outputs an adaptation endpoint characterizing the target object's current coupling performance level on the current channel and an adaptation rate as a learnability coefficient. The adaptation endpoint can be used to determine the channel weight distribution of the target object's coupling vector. The perturbation-adaptation paradigm measures the adaptation endpoint and adaptation rate of the target object through real-time tracking and perturbation feedback. The adaptation rate is the learnability coefficient, representing the speed at which the target object recovers to a stable coupling state after being perturbed. A higher adaptation rate indicates a larger learnability coefficient, meaning a greater potential to improve coupling capability through training. The adaptation rate is used to predict the rate at which the coupling gap narrows after learning. The learnability coefficient of the target object can be used to characterize the rate at which the target object's tracking error decays with the number of practice iterations, i.e., the speed at which the target object recovers to a stable coupling state after being perturbed.
[0053] In some embodiments, the target object currently possesses a coupling performance level w in channel C. By adapting to the endpoint Obtained through normalization. It can be used to calculate the gap between the current state of the target career and the current state of the target; the target corresponds to the learnability coefficient η of the candidate course in channel C. It can be used to predict the gap reduction rate after learning. Therefore, the coupling vector of the target object can characterize the current ability level of the target object to make judgments on each predefined coupling channel.
[0054] In some embodiments, the perturbation-adaptation paradigm measurement includes: Present a tracking task with real-time sensory feedback to the target object on the display screen; apply a controlled perturbation to the real-time sensory feedback mapping of the target object after a predefined number of tests; collect the tracking error sequence for each tracking test and fit a decay function; characterize the coupling performance level of the target object in each channel with the fitted asymptotic error and characterize the learnability coefficient of the target object in each channel with the decay rate.
[0055] The perturbation-adaptation paradigm measurement process can be summarized as follows: tracking task → applying controlled perturbation (visual stimulation, tactile stimulation, or posture perturbation) → fitting the decay curve of error during successive tests → adaptation endpoint = asymptotic error e∞ (the lower the error, the stronger the ability), adaptation rate = 1 / τ.
[0056] Taking the visual-fine channel test on a tablet as an example: the tracking task presents a target moving along a preset trajectory on the screen, and the subject, i.e., the target object, tracks it in real time with a stylus; the screen displays the subject's own tracking handwriting and / or cursor position in real time (the subject can instantly visually review the results of their actions), forming a real-time closed loop of "action → seeing deviation → correction"; after a set number of tracking tests, the system applies a controlled perturbation to the handwriting-cursor mapping (such as rotation by 30°, translation offset, or gain change, i.e., the classic visual motion rotation paradigm), and the subject recalibrates online. The tracking error sequence (e1, e2, …, e_K, representing the average distance between the target and the cursor in each test) is collected for each test. An exponential decay function is fitted to this tracking error sequence, which includes fitting the asymptotic error and the decay rate; the fitted asymptotic error represents the current coupling performance level of the channel (normalized to the current coupling performance level of the target object in this channel), and the fitted decay rate represents the learnability coefficient η of the channel. The current coupling performance level is used to calculate the current gap; the learnability coefficient is used to predict the gap reduction rate after learning. In this embodiment, no specific threshold is limited; each threshold is determined by the training data / outcome data.
[0057] In some embodiments, the exponential decay function is represented as follows:
[0058] in, For the first The tracking error of each test (e.g., the average distance between the cursor and the target). To adapt to the endpoint, that is, when the number of tests... The asymptotic error at time represents the current coupling performance level of the target object under this channel (the lower the error, the stronger the current coupling capability). The initial error, that is Error in time. For the test number sequence ( ). The time constant represents how quickly the error decays. To adapt the rate, characterize the learnability of coupling ( The larger the value, the faster the error decays and the stronger the learnability.
[0059] (III) Curriculum Side The process involves obtaining textual descriptions and syllabi of candidate courses from the candidate course database. A large language model is used to decompose the course texts, determine their pedagogical coupling categories, and assign corresponding coupling vectors. Simultaneously, based on the course syllabus, the large language model evaluates the course's training level on predefined cognitive dimensions, thus obtaining the course-side cognitive vector. This determination process requires strict rule constraints. The method for obtaining the course-side cognitive vector is similar to that for the target profession; details can be found in the description of the target profession and will not be repeated here. The pedagogical coupling categories include decoupling and practical types.
[0060] The cognitive vector is determined by LLM according to predefined specifications from the course text, indicating the degree of development of each cognitive dimension (such as the 7 cognitive dimensions described above) in the course (0–3). The coupling vector is constructed by the pedagogical coupling category: if the candidate course is determined to be decoupled, then the coupling strength s=0, the channel weight distribution is empty, and the bound ability dimension is empty; if the candidate course is determined to be practical, then the coupling strength s= is rated according to the depth of practical experience, such as 1–3. The channel weight is determined by what kind of physical or sensory real-time feedback practice the student uses, and the channel weight corresponding to the physical or sensory experience obtained from the determination is set to a larger value, while the weights of other channels are set to a smaller value or 0. For example, clinical clerkship is conducted through interpersonal and tactile real-time feedback practice, dance is conducted through proprioceptive real-time feedback practice, and microscopy experiments are conducted through visual-fine real-time feedback practice.
[0061] It should be noted that the candidate courses include internships or practical courses, which are classified as hands-on. That is, the teaching methodology classification result of the internship / practice course is hands-on, and its channel weight distribution is predicted based on the task-level natural language description text of the internship / practice course through a large language model. The prediction process is directly decomposed in the same way as the occupation described above—because an internship itself is a real job.
[0062] In some embodiments, the large language model also assigns a dominant coupling channel to practical courses. For courses identified as practical, the large language model further determines from the course text "which coupling channel's real-time feedback is used for the practical exercise," thereby assigning a channel weight distribution, with the channel with the highest weight being the dominant coupling channel. For example, "microbiology experiments" mainly rely on tactile sensation and micro-management, so the dominant coupling channel is determined to be tactile manipulation; "on-site troubleshooting practice" requires real-time communication and response with on-site personnel, so the dominant coupling channel is determined to be real-time interpersonal interaction.
[0063] Step S3: Calculate the gap between the target object and the target occupation in terms of their shared ability representations. This gap includes a coupling gap decomposed by channel and a cognitive gap. The cognitive gap is the difference in ability levels between the target object and the occupation across each cognitive dimension, i.e., the difference in cognitive vectors. The coupling gap is the difference in coupling vectors between the target object and the occupation across each channel.
[0064] Step S4: For coupling gaps greater than a set coupling threshold, courses or internships that are classified as practical and whose dominant coupling channel is the same as the corresponding channel are recommended. For target objects with cognitive gaps greater than a set cognitive threshold, decoupling courses are recommended. This step ensures that coupling gaps can only be closed by practical learning, avoiding invalid recommendations that close coupling gaps with lectures. The candidate courses include internships or practical courses, which are classified as practical, and their channel weight distribution is obtained by decomposing their task-level natural language text descriptions according to a career-side method.
[0065] In some embodiments, the coupling threshold and the cognitive threshold can be set by component, such as ≥0.5 level. For channel components with coupling gaps exceeding the set coupling threshold: practical courses / internships in the same channel are recommended; for dimensional components with cognitive gaps exceeding the set cognitive threshold: courses that cultivate that dimension (including decoupling courses) are recommended; no recommendations are generated for components below the set coupling threshold and the set cognitive threshold.
[0066] Step S5 involves superimposing the predicted gain of the recommended courses or internships on the target candidate's ability representation, and outputting the updated ability representation and the reduction in the gap with the target profession. This step achieves closed-loop feedback in ability development recommendations, enabling the target candidate to intuitively understand the expected effect of the recommended scheme.
[0067] Furthermore, the common ability representation of the target object can be represented as the shape of a point cloud in a vector space with cognition and coupling as coordinate axes. The prediction gain of the recommended course is represented as the extension of this point cloud shape along the corresponding coordinate axis. Here, the prediction gain g = η⊙v_course, where η represents the learnability coefficient, v_course represents the common ability representation of the candidate course, and the cognition vector is multiplied by the corresponding cognitive learnability coefficient, and the channel weight distribution is multiplied by the channel learnability coefficient η. Corresponding multiplication.
[0068] In some embodiments, the method further includes: Obtain multiple common capability representations of the target object obtained from multiple tests; Obtain the outer edge information of the point cloud in the common capability representation vector space of the multiple common capability representations; The predictive gain of the target course is determined based on the learnability coefficient of the target object corresponding to at least one of the target courses and the common ability representation. Based on the predicted gain of at least one of the target courses and the outer edge information of the point cloud, the improvement effect of the target object after learning the target course is obtained; the improvement effect includes the updated common ability representation of the target object after learning the target course, as well as the cognitive gap and coupling gap with the target occupation.
[0069] In this embodiment, the common ability representation of the target object is not a single mean vector. Each completed assessment task / work / internship performance can generate a corresponding common ability representation. Each common ability representation can be represented as a point in the common ability representation vector space. The points corresponding to the performance of all tests constitute a point cloud. The outer edge shape of the point cloud (such as the quantile envelope of each axis) can retain more of the fluctuation range and instability of the target object's ability, as well as the upper and lower limits that the target object has actually reached, than a single vector.
[0070] The extension amount represents the predictive improvement effect of a course, which is equal to the distance the outer edge of the point cloud of the corresponding target object is pushed outward along the axis trained by the course (the axis of the cognitive dimension for cognitive courses; the axis of the channel for practical courses). The value is the predictive gain component g. The direction of the corresponding coordinate axis can be understood as the direction of the axis trained by the course. By adding the predictive gain vectors of each course in the recommendation list component by component (taking the upper limit of saturation for the same component), and adding it to the current common representation of the target object, we can obtain the updated representation p′ after completing the entire recommendation combination and the gap reduction amount. The gap reduction amount includes the reduction amount of cognitive gap and coupling gap.
[0071] This embodiment further filters candidate occupations by superimposing an interest axis and a resilience axis. Multiple basic interest domains are pre-defined for both the target object and the occupation. Each component of the target object's interest vector represents the target object's preference for that basic interest domain, which can be obtained through corresponding assessments. Each component of the occupation's interest vector represents the degree to which the occupation's daily work exposes it to that interest domain. The interest matching result is the matching result between the target object's interest vector and the occupation's interest vector, such as whether the similarity or cosine value of each component exceeds a preset interest threshold.
[0072] Resilience can be understood as a profession's resilience against automation; high resilience indicates that the profession is not easily automated under foreseeable circumstances. For example: Profession resilience = 1 Occupational exposure score. This occupational automation exposure score can be pre-calculated using any known occupational automation exposure assessment method, such as publicly available assessment methods based on a comparison of occupational task composition and technological benchmarks (e.g., occupational computerization probability index, artificial intelligence occupational exposure index, etc.).
[0073] Capability attainability indicates whether the gap between the target object's shared capabilities and occupational requirements is below a set gap threshold, which includes cognitive and coupling gaps. In some embodiments, capability attainability is indicated if the gap is currently below the set gap threshold, or if it is predicted, after recommended learning, that the gap will fall below the set gap threshold; the set gap threshold includes a set cognitive threshold and a set coupling threshold.
[0074] Occupations falling into a predefined high automation exposure zone are considered high-substitution-risk occupations. These include occupations with an automation exposure score higher than a preset exposure threshold, and whose coupling strength in the common ability representation is lower than a preset coupling threshold. The ability requirements of these high-substitution-risk occupations are covered by the capability frontier of software artificial intelligence and embodied robots, and these occupations are weakly coupled, meaning they can be separated from the executor and handed over to the machine.
[0075] The following describes some possible implementation details of the above-described scheme in conjunction with exemplary embodiments.
[0076] Example 1: Overall System Flow The solution provided in this embodiment includes the following steps. First, the system receives input target object information, target occupation information, and candidate course library information. The target object information includes resume, assessment results, or perturbation-adaptation paradigm measurement data; the target occupation information includes occupation description and task list; and the course library includes course text descriptions and syllabi. The system decomposes the three pieces of information into a common ability representation, which includes a cognitive vector and a coupling vector.
[0077] Secondly, the system uses a large language model to determine the pedagogical coupling category of the text descriptions for each candidate course. The large language model determines whether a course is decoupled or pedagogically based on whether the course text contains keywords such as "practical," "clinical," "studio," or "on-site practice," and whether it requires students to practice through real-time physical or sensory feedback. For pedagogically-based courses, the system further determines its dominant coupling channel, i.e., which coupling channel the course primarily trains.
[0078] The system then calculates the gap in channel decomposition capabilities between the target individual and the target profession. The cognitive gap is the difference in ability values between the target individual and the profession across each cognitive dimension, while the coupling gap is the difference in coupling strength between the target individual and the profession across each channel. For channel components where the coupling gap exceeds a set coupling threshold, the system only recommends practical or internship courses within the same channel; for dimension components where the cognitive gap exceeds a set cognitive threshold, the system recommends decoupling courses to develop that dimension. Components below the threshold are not recommended.
[0079] Finally, the system superimposes the predicted gains of the recommended courses or internships on the target candidate's ability representation, onto the current ability representation, outputting the updated ability representation and the reduction in the gap with the target profession, and visualizing it in the form of an ability point cloud. The calculation of the predicted gain takes into account the target candidate's adaptation rate; the higher the adaptation rate of a channel, the larger its predicted gain value, indicating that the coupling gap of that channel can be learned and closed more quickly.
[0080] Example 2: Determining the Coupling Category of Curriculum Teaching Methods The course teaching method coupling category determination in this embodiment uses a large language model to analyze the course text. Specifically, the system inputs the course text into the large language model, which scores the course according to predefined scoring criteria. The scoring criteria include the following criteria: whether the course requires students to practice through real-time physical or sensory feedback, whether the course includes hands-on, clinical, workshop, or on-site practice components, and whether the course is mainly lecture-based or theoretically imparted.
[0081] Taking surgery courses as an example, "Surgery Lectures" primarily impart knowledge through lectures and do not require students to practice using real-time bodily feedback; therefore, it is classified as a decoupled course with a coupling strength of 0. "Surgery Internship" requires students to perform real-time operations on the operating table, coordinating hand-eye movements to complete suturing, cutting, and other procedures; therefore, it is classified as a practical course with a coupling strength of 3, with the dominant coupling channel being vision-fine manipulation. Although "Computer Programming Experiments" require students to operate on computers, keyboard input does not constitute sensorimotor coupling; therefore, according to the criteria of this disclosure, it is classified as a decoupled course.
[0082] Similarly, dance training courses were classified as hands-on, with proprioception and full-body posture as the dominant coupling channels. On-site troubleshooting practice was also classified as hands-on, with real-time interpersonal interaction as the dominant coupling channel. Clinical psychology lectures were classified as decoupled, with a coupling strength of 0. These classifications enable the system to correctly distinguish between lectures and hands-on practice, avoiding closed coupling gaps in lecture sessions.
[0083] Example 3: Perturbation-Adaptive Measurement of Target Object Coupling Vector In this embodiment, the coupling vector of the target object is obtained through a perturbation-adaptation paradigm. The perturbation-adaptation paradigm measures the current coupling and adaptation rate of the target object through real-time tracking and perturbation feedback. The adaptation rate is the learnability coefficient; a higher adaptation rate indicates a greater potential to improve the coupling capability through training. Except for the vision-fine manipulation channel, the adaptation rate algorithms for the other channels are under design; this embodiment only describes the method without providing specific thresholds. The adaptation rate is used to predict the rate at which the coupling gap narrows after learning, and is then used for gain prediction.
[0084] Example 4: Visualization of Capability Point Cloud This embodiment projects the target object's capabilities as a point cloud shape in a two-dimensional space with cognition and coupling as the coordinate axes. The horizontal axis is the coupling axis, and the vertical axis is the cognition axis. Current capabilities are displayed in point cloud form, and the target occupation is displayed as marked points. Channel decomposition gaps are marked with arrows of different colors, and the extension direction and magnitude of recommended courses on the capability point cloud are marked with arrows. This visualization method makes the capability development path intuitive and understandable. The target object can clearly see the gap between its current capabilities and the target occupation, as well as in which directions recommended courses can provide benefits.
[0085] Example 5: Occupational Recommendation and Automated Exposure Filtering This embodiment further filters candidate occupations by overlaying an interest axis and a resilience axis. The system first performs an initial screening of candidate occupations based on the target object's interest matching degree and capability attainability. Interest matching is defined as the similarity between the target object's interest vector and the occupation's interest exposure vector exceeding a preset threshold. Capability attainability is defined as the gap between the target object's capability representation and the occupation's requirements currently falling below a threshold, or being predictable within budget by the recommended learning path. For occupations that are interest-matched and capability attainable but fall into a predefined high automation exposure zone, i.e., occupations with high automation exposure scores, the system marks them and outputs alternative occupation recommendations. This automation exposure score can be pre-calculated using any known occupation automation exposure assessment method, such as publicly available assessment methods based on occupational task composition and technical benchmark comparisons (e.g., occupational computerization probability index, artificial intelligence occupational exposure index, etc.). The resilience axis is defined as the occupation's resilience against automation; resilience equals 1 minus the occupation's automation exposure score. The marked occupations are those that are interest-matched, capability attainable, and have an automation exposure score higher than a preset threshold. The alternative occupations subsequently output by the system must still satisfy the conditions of interest matching and high resilience (low automation exposure score).
[0086] Figure 2 The diagram illustrates a structural block diagram of a learning path planning apparatus according to an embodiment of the present disclosure. This apparatus can be implemented as part or all of an electronic device through software, hardware, or a combination of both. Figure 2 As shown, the learning path planning device includes: The acquisition module 201 is configured to acquire the target object, target occupation, candidate courses, and pedagogical coupling category of the candidate courses for the learning path to be planned; The first determining module 202 is configured to determine the common ability representation of the target object, the target occupation, and the candidate courses, respectively. The common ability representation includes a cognitive vector and a coupling vector. The cognitive vectors of the target object, the target occupation, and the candidate courses respectively represent the cognitive level already possessed by the target object in a predefined cognitive dimension, the demand level of the target occupation in the predefined cognitive dimension, and the training level of the candidate courses in the predefined cognitive dimension. The coupling vectors of the target object, the target occupation, and the candidate courses respectively represent the coupling ability of the target object, the degree to which the target occupation requires cognitive judgment to be formed instantly through real-time sensorimotor feedback, and the degree to which the candidate courses can train cognitive judgment to be formed instantly through real-time sensorimotor feedback. The second determining module 203 is configured to determine the cognitive gap between the target object and the target occupation based on the cognitive vector in the common ability representation, and to determine the coupling gap between the target object and the target occupation based on the coupling vector in the common ability representation; The third determining module 204 is configured to determine the candidate courses with the teaching method coupling category of decoupling as the target courses for the target object on the learning path if the cognitive gap is greater than a set cognitive threshold. The fourth determining module 205 is configured to determine the candidate courses with the teaching method coupling category of practical type as the target courses of the target object on the learning path if the coupling gap is greater than a set coupling threshold.
[0087] The technical features and their corresponding explanations and descriptions involved in the above-described device embodiments are the same as, corresponding to, or similar to the technical features and their corresponding explanations and descriptions involved in the above-described method embodiments. For the technical features and their corresponding explanations and descriptions involved in the above-described device embodiments, please refer to the technical features and their corresponding explanations and descriptions involved in the above-described method embodiments. This disclosure will not repeat them here.
[0088] Figure 3 A simplified block diagram is shown for an electronic device 300 suitable for implementing embodiments of the present disclosure. Figure 3 As shown, the electronic device 300 includes a processing unit 301, which can be implemented as a CPU, GPU, FPGA, NPU, or other processing unit. The processing unit 301 can execute various processes according to any of the above-described embodiments of the methods disclosed herein, based on a program stored in the read-only memory (ROM) 302 or a program loaded from the storage portion 308 into the random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.
[0089] The following components are connected to I / O interface 305: an input section 306 including a keyboard, mouse, etc.; an output section 307 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN card, modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.
[0090] In particular, according to embodiments of this disclosure, any of the methods described above in the embodiments of this disclosure can be implemented as a computer software program. For example, embodiments of this disclosure include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program containing program code for performing any of the methods in the embodiments of this disclosure. In such an embodiment, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311.
[0091] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, apparatus (devices), or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0092] This disclosure is described with reference to flowchart illustrations of methods, apparatus (devices), and computer program products according to embodiments of this disclosure. It should be understood that each step in the flowchart can be implemented by computer program instructions.
[0093] These computer program instructions may be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction means, the implementation process of which is described in the instruction means. Figure 1 The function specified in one or more processes.
[0094] These computer program instructions may also be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing device, produce instructions for implementing processes. Figure 1 A device for a function specified in one or more processes.
[0095] Another embodiment of this disclosure relates to a non-volatile storage medium for storing a computer-readable program for use by a computer to execute some or all of the above-described method embodiments.
[0096] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by specifying relevant hardware through a program. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the embodiments of this disclosure. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
[0097] The above description is merely a preferred embodiment of this disclosure and is not intended to limit this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the protection scope of this disclosure.
Claims
1. A learning path planning method, characterized in that, The method includes the following steps: Obtain the target audience, target occupation, candidate courses, and pedagogical coupling categories of the candidate courses for the learning path to be planned; The common ability representations of the target object, target occupation, and candidate courses are determined respectively. These common ability representations include cognitive vectors and coupling vectors. The cognitive vectors of the target object, target occupation, and candidate courses respectively represent the cognitive level already possessed by the target object in a predefined cognitive dimension, the demand level of the target occupation in the predefined cognitive dimension, and the training level of the candidate courses in the predefined cognitive dimension. The coupling vectors of the target object, target occupation, and candidate courses respectively represent the coupling ability of the target object, the degree to which the target occupation requires cognitive judgment to be formed instantly through real-time sensorimotor feedback, and the degree to which the candidate courses can cultivate cognitive judgment to be formed instantly through real-time sensorimotor feedback. Based on the cognitive vector in the common ability representation, the cognitive gap between the target object and the target occupation is determined, and based on the coupling vector in the common ability representation, the coupling gap between the target object and the target occupation in different coupling channels is determined; If the cognitive gap is greater than the set cognitive threshold, then the candidate courses with the teaching method coupling category of decoupling are determined as the target courses for the target object on the learning path. If the coupling gap corresponding to at least one channel is greater than a set coupling threshold, then the candidate course whose teaching method coupling category is practical and whose dominant coupling channel is the same as the at least one coupling channel is determined as the target course of the target object on the learning path.
2. The method according to claim 1, characterized in that, The coupling vector includes coupling strength, channel weight distribution, and bound capability dimension, wherein the bound capability dimension includes a pair of bound dominant cognitive dimension and dominant sensorimotor dimension.
3. The method according to claim 2, characterized in that, The channel weight distribution of the target object includes channel weights on multiple predefined coupling channels, and the channel weights include the current coupling performance level of the target object on the corresponding channel.
4. The method according to claim 3, characterized in that, The current coupling performance level of the target object is obtained by perturbation-adaptation paradigm measurement; wherein, during the perturbation-adaptation paradigm measurement process, the learnability coefficient of the target object is also measured, and the learnability coefficient is used to characterize the speed at which the target object recovers to a stable coupling state after being perturbed.
5. The method according to any one of claims 1-4, characterized in that, The coupling vector of the candidate course is determined based on the pedagogical coupling category of the candidate course.
6. The method according to any one of claims 1-4, characterized in that, The method further includes: Obtain multiple common capability representations of the target object obtained from multiple tests; Obtain the outer edge information of the point cloud in the common capability representation vector space of the multiple common capability representations; The predictive gain of the target course is determined based on the learnability coefficient of the target object corresponding to at least one of the target courses and the common ability representation. Based on the predicted gain of at least one of the target courses and the outer edge information of the point cloud, the improvement effect of the target object after learning the target course is obtained; the improvement effect includes the updated common ability representation of the target object after learning the target course, as well as the cognitive gap and coupling gap with the target occupation.
7. The method according to any one of claims 1-4, characterized in that, The method further includes: The system obtains the interest vectors of the target object and multiple target occupations in a set interest domain, and obtains the occupational exposure of the target occupations; the occupational exposure indicates whether the target occupation belongs to a high-substitution-risk occupation that is easily replaced by software artificial intelligence and embodied robots and is weakly coupled. Based on the interest vector, determine the interest matching results between the target object and the multiple target occupations; Based on the cognitive gap and coupling gap between the target professions and the target object, it is determined whether the target object has the capability to achieve the target professions. Based on the occupational exposure, interest matching results, and ability attainment results, a first set of occupations that match the target object's interests and are attainable but belong to high-substitution-risk occupations are selected from the multiple target occupations. A second set of occupations that match the target object's interests but are not attainable and do not belong to high-substitution-risk occupations are selected.
8. A learning path planning device, characterized in that, The device includes: The acquisition module is configured to acquire the target object, target occupation, candidate courses, and pedagogical coupling category of the candidate courses for the learning path to be planned; The first determining module is configured to determine the common ability representations of the target object, the target occupation, and the candidate courses, respectively. The common ability representations include cognitive vectors and coupling vectors. The cognitive vectors of the target object, the target occupation, and the candidate courses respectively represent the cognitive level already possessed by the target object in a predefined cognitive dimension, the demand level of the target occupation in the predefined cognitive dimension, and the training level of the candidate courses in the predefined cognitive dimension. The coupling vectors of the target object, the target occupation, and the candidate courses respectively represent the coupling ability of the target object, the degree to which the target occupation requires cognitive judgments to be formed instantly through real-time sensorimotor feedback, and the degree to which the candidate courses can cultivate cognitive judgments to be formed instantly through real-time sensorimotor feedback. The second determining module is configured to determine the cognitive gap between the target object and the target occupation based on the cognitive vector in the common ability representation, and to determine the coupling gap between the target object and the target occupation in different coupling channels based on the coupling vector in the common ability representation; the third determining module is configured to determine the candidate courses with the teaching method coupling category of decoupling as the target courses for the target object on the learning path if the cognitive gap is greater than a set cognitive threshold. The fourth determining module is configured to determine the candidate course of the target object on the learning path if the coupling gap corresponding to at least one channel is greater than a set coupling threshold.
9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1-7.
10. An electronic device comprising a memory and a processor, characterized in that, When the processor executes the computer program in the memory, it implements the steps of the method according to any one of claims 1-7.
