Logistics personalized learning path recommendation system based on AI

Abstract

The invention relates to the technical field of logistics learning systems, discloses an AI-based logistics personalized learning path recommendation system, and solves the problems that an existing logistics learning path recommendation system does not have the capabilities of precise personalization, real-time content, dynamic path and comprehensive evaluation, so that the learning efficiency and skill conversion effect are reduced, and the learning difficulty is reduced. The system comprises a logistics learner multi-dimensional portrait acquisition module, a logistics industry knowledge graph construction and real-time updating module, an AI intelligent demand matching and skill gap analysis engine, a personalized learning path generation and dynamic optimization module, and a scenarized learning content intelligent pushing module. The logistics learning path recommendation system has the capabilities of accurate individuation, real-time content, dynamic path and comprehensive evaluation, so that the learning efficiency and the skill conversion effect are improved, and the requirement of the logistics industry for specialized talent training is met.

Description

Technical Field

[0001] This invention belongs to the technical field of logistics learning systems, specifically an AI-based personalized learning path recommendation system for logistics. Background Technology

[0002] The AI-based personalized learning path recommendation system for logistics utilizes machine learning, natural language processing, and data mining technologies to deeply analyze learners' behavioral data, knowledge levels, and interest preferences, dynamically generating customized learning paths and resource recommendations. Its core lies in using user behavior tracking, knowledge point mastery modeling, and weak point identification, combined with collaborative filtering, content recommendation, and deep learning algorithms, to accurately push personalized courses, exercises, and practical projects. This system is widely used in logistics education, supporting teachers in reconstructing an "AI plus scenario-based" teaching model. Through intelligent simulation platforms, it simulates real business scenarios such as inventory optimization and transportation scheduling, enhancing students' practical abilities. Simultaneously, relying on an adaptive learning system, it dynamically adjusts recommendation strategies based on learning progress, addressing knowledge gaps or providing advanced content, thus helping to cultivate well-rounded logistics professionals with both professional competence and innovative capabilities. As the logistics industry transforms towards digitalization, intelligence, and globalization, sub-sectors such as e-commerce logistics, cross-border logistics, and cold chain logistics are developing rapidly, leading to increasingly refined and specialized skill requirements for logistics practitioners. Existing logistics learning systems mostly adopt a generalized training model, which has the following technical shortcomings: 1. Insufficient personalization: The existing system only collects basic job information of learners without deeply capturing the differences in job sub-scenarios, skill proficiency levels, learning preferences and other dimensions. As a result, the learning path recommendation is superficial and cannot meet the differentiated needs of learners at different levels and in different sub-fields. 2. Static knowledge graph: Logistics industry policies, regulations and technology applications are frequently updated, but the existing knowledge graphs are mostly statically constructed, failing to achieve dynamic adjustment of skill weights and the association between knowledge points and real-time cases. This results in a disconnect between learning content and actual industry needs, and a lack of specific annotations for violation risk points.

[0003] 3. Crude skill gap analysis: The existing system can only determine whether a skill "exists" or not, without quantifying the degree of the gap or determining the learning priority based on the learner's work task frequency. This results in a chaotic sorting of learning path nodes, with high-frequency and essential skills failing to be prioritized for improvement, leading to low learning efficiency. 4. Rigid learning paths: The learning paths of existing systems are mostly fixed processes that cannot be dynamically adjusted according to learners' real-time learning behavior and do not support learners' active intervention in the path, resulting in insufficient flexibility; at the same time, the learning content lacks scenario-based design and is detached from actual work scenarios. 5. One-sided evaluation of learning effectiveness: The existing system only relies on theoretical tests to evaluate learning effectiveness, without linking it to practical simulation data and work performance data. It cannot accurately judge the skill conversion effect, and the feedback and adjustment are delayed, making it difficult to form a closed loop of learning-evaluation-optimization. Summary of the Invention

[0004] In response to the above situation and to overcome the shortcomings of existing technologies, this invention provides an AI-based personalized learning path recommendation system for logistics. This system effectively solves the problem that existing logistics learning path recommendation systems lack the capabilities of accurate personalization, real-time content provision, dynamic path provision, and comprehensive evaluation, thereby reducing learning efficiency and skill conversion effectiveness, and failing to meet the logistics industry's demand for professional talent training.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an AI-based personalized learning path recommendation system for logistics, including a multi-dimensional profile collection module for logistics learners, a knowledge graph construction and real-time update module for the logistics industry, an AI intelligent demand matching and skills gap analysis engine, a personalized learning path generation and dynamic optimization module, a scenario-based intelligent learning content push module, a multi-dimensional evaluation and feedback module for learning effectiveness, and a system interaction and learning progress visualization module.

[0006] Preferably, the logistics learner multi-dimensional profile collection module is used to collect learner job segmentation information, skill proficiency level data, learning preferences, learning goals and work scenario feedback data, and generate a five-dimensional learner profile of job, skills, preferences, goals and work scenarios after cleaning and fusion.

[0007] Preferably, the logistics industry knowledge graph construction and real-time update module is used to connect with external industry dynamics and enterprise job standard data, construct and update a five-layer structured knowledge graph including job layer, skill layer, knowledge point layer, learning resource layer and risk warning layer in real time, and dynamically adjust skill weights and knowledge point relationships.

[0008] Preferably, the AI ​​intelligent demand matching and skills gap analysis engine is connected to the multi-dimensional profile collection module for logistics learners and the knowledge graph construction and real-time update module for the logistics industry, respectively. It associates learner profiles and knowledge graphs through semantic matching algorithms, calculates the quantitative value and type of skills gap based on the gap degree quantification model, maps gap priority by combining work task frequency, and achieves accurate matching of job positions, skills and learning content through a ternary matching model, and outputs a skills gap analysis report.

[0009] Preferably, the personalized learning path generation and dynamic optimization module is connected to the AI ​​intelligent demand matching and skills gap analysis engine. Based on the gap priority and learner's learning habits, it generates an initial learning path with a multi-branch structure. By capturing learner's real-time learning behavior data and combining dynamic adjustment rules and learner's active intervention requests, it achieves adaptive optimization of path nodes, sequence, and duration.

[0010] Preferably, the scenario-based learning content intelligent push module is connected to the personalized learning path generation and dynamic optimization module to build a scenario-based learning resource library. Based on the learning objectives and learner preferences of the path nodes, it filters and matches core resources, auxiliary resources and extended resources, and pushes them accurately according to the push strategy.

[0011] Preferably, the multi-dimensional learning effect evaluation and feedback module is connected to the scenario-based learning content intelligent push module, the personalized learning path generation and dynamic optimization module, and the logistics learner multi-dimensional profile collection module, respectively. Through theoretical evaluation, practical simulation evaluation, and work performance correlation analysis, it generates a multi-dimensional learning effect feedback report, providing a basis for path optimization and profile updates.

[0012] Preferably, the system interaction and learning progress visualization module is connected to the personalized learning path generation and dynamic optimization module and the multi-dimensional evaluation and feedback module for learning effects, respectively, to visualize the learning path, progress and skill improvement curve, provide interactive entry points such as path adjustment and question submission, and push message reminders to learners and enterprise administrators.

[0013] Compared with the prior art, the beneficial effects of the present invention are: Through five-dimensional learner profiling and quantitative gap analysis, combined with multi-branch path design, personalized recommendations are achieved for each individual with high personalization accuracy; the dynamic update mechanism of the knowledge graph ensures that the learning content keeps up with the latest industry regulations and technological innovations, and the customized scenario-based resources achieve deep integration of learning and work scenarios, making the content practical in real time; It supports two-way adjustment of dynamic system optimization and learner intervention, avoiding rigidity and ensuring high flexibility; it integrates three-dimensional evaluation of theory, practice, and work performance to accurately judge the effect of skill transformation and provide a comprehensive and objective evaluation. It covers multiple logistics sub-sectors and all job levels, and can meet the training needs of large, medium and small logistics enterprises, with a wide range of application scenarios; the modular design supports subsequent function expansion and knowledge graph expansion, extending the system life cycle, and its scalability is strong. Attached Figure Description

[0014] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0015] In the attached diagram: Figure 1 This is a diagram of the AI-based personalized learning path recommendation system for logistics according to the present invention. Figure 2 This is a system architecture diagram of the multi-dimensional profile collection module for logistics learners in this invention; Figure 3 This is a system architecture diagram of the logistics industry knowledge graph construction and real-time update module of the present invention; Figure 4 This is a system architecture diagram of the AI ​​intelligent demand matching and skills gap analysis engine of this invention; Figure 5 This is a system architecture diagram of the personalized learning path generation and dynamic optimization module of this invention; Figure 6 This is a system architecture diagram of the multi-dimensional evaluation and feedback module for learning effectiveness in this invention; Figure 7 This is a system architecture diagram of the scenario-based learning content intelligent push module of the present invention; Figure 8 This is a system architecture diagram of the interactive and learning progress visualization module of the present invention; Detailed Implementation

[0016] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0017] Example 1, by Figures 1 to 8 The present invention includes a multi-dimensional profile collection module for logistics learners, a knowledge graph construction and real-time update module for the logistics industry, an AI intelligent demand matching and skills gap analysis engine, a personalized learning path generation and dynamic optimization module, a scenario-based intelligent learning content push module, a multi-dimensional evaluation and feedback module for learning effectiveness, and a system interaction and learning progress visualization module. The logistics learner multi-dimensional profile collection module is used to collect learner job segmentation information, skill proficiency level data, learning preferences, learning goals and work scenario feedback data. After cleaning and fusion, it generates a five-dimensional learner profile based on job, skills, preferences, goals and work scenario. The knowledge graph construction and real-time update module for the logistics industry is used to connect with external industry dynamics and enterprise job standard data to build and update a five-layer structured knowledge graph that includes job layer, skill layer, knowledge point layer, learning resource layer and risk warning layer, and dynamically adjust skill weights and knowledge point relationships. The AI-powered demand matching and skills gap analysis engine is connected to the multi-dimensional profile collection module for logistics learners and the knowledge graph construction and real-time update module for the logistics industry. It associates learner profiles and knowledge graphs through semantic matching algorithms, calculates the quantitative value and type of skills gaps based on the gap degree quantification model, maps gap priority by combining the frequency of work tasks, and achieves accurate matching of job positions, skills and learning content through a ternary matching model, and outputs a skills gap analysis report. The personalized learning path generation and dynamic optimization module is connected to the AI ​​intelligent demand matching and skills gap analysis engine. Based on the gap priority and learner's learning habits, it generates an initial learning path with a multi-branch structure. By capturing learner's real-time learning behavior data and combining dynamic adjustment rules and learner's active intervention requests, it achieves adaptive optimization of path nodes, sequence, and duration. The scenario-based learning content intelligent push module is connected with the personalized learning path generation and dynamic optimization module to build a scenario-based learning resource library. Based on the learning objectives and learner preferences of the path nodes, it filters and matches core resources, auxiliary resources and extended resources, and pushes them accurately according to the push strategy. The multi-dimensional learning effect evaluation and feedback module is connected to the scenario-based learning content intelligent push module, the personalized learning path generation and dynamic optimization module, and the logistics learner multi-dimensional profile collection module. Through theoretical assessment, practical simulation evaluation, and work performance correlation analysis, it generates a multi-dimensional learning effect feedback report, providing a basis for path optimization and profile updates. The system interaction and learning progress visualization module is connected to the personalized learning path generation and dynamic optimization module and the multi-dimensional evaluation and feedback module for learning outcomes, respectively. It visualizes the learning path, progress and skill improvement curve, provides interactive entry points such as path adjustment and question submission, and pushes message reminders to learners and enterprise administrators.

[0018] The multi-dimensional profile collection module for logistics learners includes: a refined job information collection unit, used to collect learners' industry sub-fields, job levels, core work tasks, and high-frequency work scenarios, generating over 100 detailed job tags; a skill level grading assessment unit, including theoretical knowledge testing, practical simulation assessment, and work case analysis sub-units, outputting a 0-10 score for skill proficiency and a five-level grading result; a learning preference and goal collection unit, which obtains learners' learning format preferences, duration habits, and short- and long-term learning goals through questionnaires and behavioral back-inference, generating a goal priority matrix; a work scenario feedback collection unit, which connects to enterprise WMS / ERP / TMS systems to collect data on frequently reported errors and difficult tasks, supporting learners to proactively submit work feedback; and a profile data cleaning and fusion unit, which performs deduplication, error correction, and standardization on the raw data, merging it to generate a five-dimensional learner profile and storing it in the profile database.

[0019] Through units such as refined collection of job information, skill level assessment, collection of learning preferences and goals, and collection of work scenario feedback, learner data from multiple aspects is obtained, and after cleaning and integration, a five-dimensional learner profile is generated and stored.

[0020] The logistics industry knowledge graph construction and real-time update module includes: an external data access unit, which connects to authoritative industry channels, internal enterprise data sources, and technological innovation data sources to obtain policy and regulatory information, job standards, and technological dynamics; a core knowledge graph construction unit, which constructs a five-layer structured knowledge graph consisting of job positions, skills, knowledge points, learning resources, and risk warnings, and labels skill types, knowledge point relationships, and violation risk points; a dynamic skill weight adjustment unit, which adjusts the weight values ​​of each skill in the knowledge graph in real time based on enterprise performance data and recruitment needs; a real-time knowledge point-case association unit, which binds real-world industry cases with knowledge points using NLP technology and labels case scenario types; and a knowledge graph update scheduling unit, which triggers updates according to preset rules (real-time updates for policies and monthly updates for job requirements), and ensures the logical coherence of the graph through a knowledge consistency verification unit.

[0021] By connecting with authoritative external channels and internal enterprise data sources, a five-layer structured knowledge graph is constructed, including job positions, skills, and knowledge points. Skill weights are dynamically adjusted, real-time cases are linked, and the graph is updated regularly, taking into account industry dynamics and enterprise needs.

[0022] The AI-powered intelligent demand matching and skills gap analysis engine includes: a profile-knowledge graph association unit, which uses a semantic matching algorithm to associate a five-dimensional learner profile with the job and skill levels of the knowledge graph; a skills gap quantification analysis unit, which, based on a gap degree quantification model, compares skill proficiency scores with job qualification thresholds and outputs gap quantification values ​​and three gap types: basic deficiency, advanced improvement, and optimization; a task frequency-gap priority mapping unit, which combines work task frequency with gap type to generate S / A / B three-level gap priority; a learning goal-skill association unit, which uses a goal decomposition algorithm to break down learners' short- and long-term goals into corresponding skill sets; and a ternary matching model operation unit, which, based on a deep learning model with a CNN and Transformer architecture, takes profile data, knowledge graph data, and gap quantification data as input and outputs suggestions on learning content type and difficulty.

[0023] A semantic matching algorithm is used to link learner profiles and knowledge graphs. A gap quantification model is used to calculate the quantitative value and type of skill gaps. The priority of gaps is determined by combining the frequency of work tasks. Then, a ternary matching model is used to achieve accurate matching of job positions, skills and learning content, and output a skill gap analysis report.

[0024] The personalized learning path generation and dynamic optimization module includes: an initial path planning unit, which uses a path node arrangement algorithm to determine path nodes, sort node order, allocate learning time, and mark node relationships; a path branching adaptive unit, which designs three types of multi-branch paths—basic branches, in-depth branches, and cross-boundary branches—for advanced learning gaps; a real-time behavior capture unit, which captures learners' path node completion progress, learning interaction data, pause / skip behaviors, and feedback; a dynamic adjustment decision unit, which executes three types of rules—progress adjustment, content adjustment, and sequence adjustment—based on real-time behavior data and learning effect evaluation results; a learner proactive intervention response unit, which receives and verifies learners' path adjustment requests and simultaneously optimizes the path; and a path version management unit, which records path adjustment history and generates a path version library to support backtracking.

[0025] An initial learning path with multiple branches is generated based on gap priority and learner habits. Learner learning behavior data is captured in real time. Combined with dynamic adjustment rules and learner intervention requests, the path nodes, order and duration are adaptively optimized, and the adjustment history is recorded.

[0026] The scenario-based learning content intelligent push module includes: a learning resource library management unit, which builds a scenario-based learning resource library containing various types of resources such as video courses, text and image materials, practical simulations, and case studies, and categorizes and tags them; a resource-path node matching unit, which uses a resource adaptation algorithm to filter core resources, auxiliary resources, and extended resource combinations for path nodes; a scenario-based resource customization unit, which customizes exclusive resources including real tasks, frequently asked questions, and the latest policies for learners' high-frequency work scenarios; a push strategy execution unit, which executes pushes based on learner preferences and path progress, according to preference adaptation, progress synchronization, and frequency control rules; and a resource feedback collection unit, which collects learners' ratings of the usability of the pushed resources and their optimization suggestions.

[0027] Based on a scenario-based learning resource library, core, auxiliary, and extended resource combinations are selected according to the learning objectives of the path nodes and learner preferences. These resources are then precisely pushed to learners using strategies such as preference adaptation and progress synchronization. At the same time, learner feedback on the resources is collected.

[0028] The multi-dimensional assessment and feedback module for learning outcomes includes: a theoretical knowledge assessment unit that automatically generates personalized test papers, supports real-time and phased assessments, and marks weak knowledge points; a practical simulation assessment unit that evaluates task completion, operational standardization, efficiency indicators, and error types through a virtual simulation system; a work performance correlation unit that connects to the enterprise's HR system and logistics management system to obtain learners' work practice data after learning; a multi-dimensional assessment integration unit that uses a weighted scoring algorithm (30% theory, 40% practical skills, and 30% work performance) to generate a comprehensive score and skills improvement curve; a feedback information generation unit that compiles and generates a learning outcome feedback report containing detailed scores, weaknesses, improvement suggestions, and path adjustment suggestions; and a feedback push unit that pushes the feedback report to the path optimization module, profile collection module, and system interaction module.

[0029] Through theoretical assessments, practical simulation evaluations, and work performance correlation analysis, a weighted scoring algorithm is used to generate comprehensive scores, skill improvement curves, and learning effect feedback reports, which are then pushed to relevant modules to provide a basis for path optimization and profile updates.

[0030] The system's interaction and learning progress visualization module includes: a learning path visualization unit, which displays the status of path nodes, their relationships, and branch selection entry points in the form of a node graph and timeline; a learning progress tracking unit, which calculates and displays the overall progress and sub-item progress in real time, supporting multi-dimensional progress statistics; a skills improvement visualization unit, which displays skills changes and gap-filling progress in the form of radar charts and line graphs; an interactive operation entry unit, which provides functions for path adjustment applications, resource feedback, question submission, and learning plan export; an enterprise administrator visualization unit, which displays team skills distribution, training progress, gap summary, and training effectiveness analysis data; and a message reminder unit, which sends progress reminders, assessment reminders, feedback reminders, and adjustment reminders through multiple channels.

[0031] Using node graphs, timelines, radar charts, and other formats, it visually displays learning paths, progress, and skill improvement status to learners and enterprise administrators, provides interactive entry points for path adjustment and question submission, and sends various message reminders.

Claims

1. An AI-based personalized learning path recommendation system for logistics, including a multi-dimensional profile collection module for logistics learners, a knowledge graph construction and real-time update module for the logistics industry, an AI intelligent demand matching and skills gap analysis engine, a personalized learning path generation and dynamic optimization module, a scenario-based intelligent push module for learning content, a multi-dimensional evaluation and feedback module for learning effectiveness, and a system interaction and learning progress visualization module. The logistics learner multi-dimensional profile collection module is used to collect learner job segmentation information, skill proficiency level data, learning preferences, learning goals and work scenario feedback data. After cleaning and fusion, it generates a five-dimensional learner profile based on job, skills, preferences, goals and work scenario. The knowledge graph construction and real-time update module for the logistics industry is used to connect with external industry dynamics and enterprise job standard data to build and update a five-layer structured knowledge graph that includes job layer, skill layer, knowledge point layer, learning resource layer and risk warning layer, and dynamically adjust skill weights and knowledge point relationships. The AI-powered demand matching and skills gap analysis engine is connected to the multi-dimensional profile collection module for logistics learners and the knowledge graph construction and real-time update module for the logistics industry. It associates learner profiles and knowledge graphs through semantic matching algorithms, calculates the quantitative value and type of skills gaps based on the gap degree quantification model, maps gap priority by combining the frequency of work tasks, and achieves accurate matching of job positions, skills and learning content through a ternary matching model, and outputs a skills gap analysis report. The personalized learning path generation and dynamic optimization module is connected to the AI ​​intelligent demand matching and skills gap analysis engine. Based on the gap priority and learner's learning habits, it generates an initial learning path with a multi-branch structure. By capturing learner's real-time learning behavior data and combining dynamic adjustment rules and learner's active intervention requests, it achieves adaptive optimization of path nodes, sequence, and duration. The scenario-based learning content intelligent push module is connected with the personalized learning path generation and dynamic optimization module to build a scenario-based learning resource library. Based on the learning objectives and learner preferences of the path nodes, it filters and matches core resources, auxiliary resources and extended resources, and pushes them accurately according to the push strategy. The multi-dimensional learning effect evaluation and feedback module is connected to the scenario-based learning content intelligent push module, the personalized learning path generation and dynamic optimization module, and the logistics learner multi-dimensional profile collection module. Through theoretical assessment, practical simulation evaluation, and work performance correlation analysis, it generates a multi-dimensional learning effect feedback report, providing a basis for path optimization and profile updates. The system interaction and learning progress visualization module is connected to the personalized learning path generation and dynamic optimization module and the multi-dimensional evaluation and feedback module for learning outcomes, respectively. It visualizes the learning path, progress and skill improvement curve, provides interactive entry points such as path adjustment and question submission, and pushes message reminders to learners and enterprise administrators.

2. The AI-based personalized learning path recommendation system for logistics according to claim 1, characterized in that: The multi-dimensional profile collection module for logistics learners includes: a refined job information collection unit, used to collect learners' industry sub-fields, job levels, core work tasks, and high-frequency work scenarios, generating over 100 subdivided job tags; a skill level grading assessment unit, including theoretical knowledge testing sub-units, practical simulation assessment sub-units, and work case analysis sub-units, outputting a 0-10 score for skill proficiency and a five-level grading result; a learning preference and goal collection unit, which obtains learners' learning format preferences, duration habits, and short- and long-term learning goals through questionnaire surveys and behavioral back-inference, generating a goal priority matrix; a work scenario feedback collection unit, which connects to the enterprise's WMS / ERP / TMS system to collect data on frequently reported errors and difficult tasks, supporting learners to proactively submit work feedback; and a profile data cleaning and fusion unit, which performs deduplication, error correction, and standardization on the raw data, fusion to generate a five-dimensional learner profile and stores it in the profile database.

3. The AI-based personalized learning path recommendation system for logistics according to claim 1, characterized in that: The logistics industry knowledge graph construction and real-time update module includes: an external data access unit, which connects to authoritative industry channels, internal enterprise data sources, and technological innovation data sources to obtain policy and regulation data, job standards, and technological dynamic data; a knowledge graph core construction unit, which constructs a five-layer structured knowledge graph of job positions, skills, knowledge points, learning resources, and risk warnings, and labels skill types, knowledge point relationships, and violation risk points; a skill weight dynamic adjustment unit, which adjusts the weight values ​​of each skill in the knowledge graph in real time based on enterprise performance data and recruitment needs; a knowledge point-case real-time association unit, which binds real industry cases with knowledge points through NLP technology and labels case scenario types; and a knowledge graph update scheduling unit, which triggers updates according to preset rules and ensures the logical coherence of the graph through a knowledge consistency verification unit.

4. The AI-based personalized learning path recommendation system for logistics according to claim 1, characterized in that: The AI-powered intelligent demand matching and skills gap analysis engine includes: a profile-knowledge graph association unit, which uses a semantic matching algorithm to associate the five-dimensional learner profile with the job level and skill level of the knowledge graph; a skills gap quantification analysis unit, which, based on a gap degree quantification model, compares skill proficiency scores with job qualification thresholds and outputs gap quantification values ​​and three gap types: basic deficiency, advanced improvement, and optimization; a task frequency-gap priority mapping unit, which combines work task frequency with gap type to generate S / A / B three-level gap priority; a learning goal-skill association unit, which uses a goal decomposition algorithm to decompose learners' short-term and long-term goals into corresponding skill sets; and a ternary matching model operation unit, which, based on a deep learning model with a CNN plus Transformer architecture, takes profile data, knowledge graph data, and gap quantification data as input and outputs suggestions on learning content type and difficulty.

5. The AI-based personalized learning path recommendation system for logistics according to claim 1, characterized in that: The personalized learning path generation and dynamic optimization module includes: an initial path planning unit, which uses a path node arrangement algorithm to determine path nodes, sort node order, allocate learning time, and mark node relationships; a path branch adaptive unit, which designs three types of multi-branch paths—basic branches, in-depth branches, and cross-boundary branches—for advanced learning gaps; a real-time behavior capture unit, which captures learners' path node completion progress, learning interaction data, pause / skip behaviors, and feedback; a dynamic adjustment decision unit, which executes three types of rules—progress adjustment, content adjustment, and sequence adjustment—based on real-time behavior data and learning effect evaluation results; a learner proactive intervention response unit, which receives and verifies learners' path adjustment requests and simultaneously optimizes the path; and a path version management unit, which records path adjustment history and generates a path version library to support backtracking.

6. The AI-based personalized learning path recommendation system for logistics according to claim 1, characterized in that: The scenario-based learning content intelligent push module includes: a learning resource library management unit, which constructs a scenario-based learning resource library containing various types of resources such as video courses, text and image materials, practical simulations, and case studies, and categorizes and labels them; a resource-path node matching unit, which uses a resource adaptation algorithm to filter core resources, auxiliary resources, and extended resource combinations for path nodes; a scenario-based resource customization unit, which customizes exclusive resources including real tasks, frequently asked questions, and the latest policies for learners' high-frequency work scenarios; a push strategy execution unit, which executes pushes based on learner preferences and path progress, according to preference adaptation, progress synchronization, and frequency control rules; and a resource feedback collection unit, which collects learners' ratings of the usability of the pushed resources and their optimization suggestions.

7. The AI-based personalized learning path recommendation system for logistics according to claim 1, characterized in that: The multi-dimensional learning outcome assessment and feedback module includes: a theoretical knowledge assessment unit that automatically generates personalized test papers, supports real-time and phased assessments, and marks weak knowledge points; a practical simulation assessment unit that evaluates task completion, operational standardization, efficiency indicators, and error types through a virtual simulation system; a work performance correlation unit that connects to the enterprise's HR system and logistics management system to obtain learners' post-learning work practice data; a multi-dimensional assessment fusion unit that uses a weighted scoring algorithm to generate a comprehensive score and skill improvement curve; a feedback information generation unit that compiles and generates a learning outcome feedback report containing score details, weak points, improvement suggestions, and path adjustment suggestions; and a feedback push unit that pushes the feedback report to the path optimization module, the profile collection module, and the system interaction module.

8. The AI-based personalized learning path recommendation system for logistics according to claim 1, characterized in that: The system's interaction and learning progress visualization module includes: a learning path visualization unit, which displays the status of path nodes, their relationships, and branch selection entry points in the form of a node graph and timeline; a learning progress tracking unit, which calculates and displays the overall progress and sub-item progress in real time, supporting multi-dimensional progress statistics; a skills improvement visualization unit, which displays skills changes and gap-filling progress in the form of radar charts and line graphs; an interactive operation entry unit, which provides functions for path adjustment applications, resource feedback, question submission, and learning plan export; an enterprise administrator visualization unit, which displays team skills distribution, training progress, gap summary, and training effect analysis data; and a message reminder unit, which sends progress reminders, assessment reminders, feedback reminders, and adjustment reminders through multiple channels.

Images