An ai-based personalized learning path planning management method and system

Abstract

The application discloses an AI-based personalized learning path planning management method and system, relates to the field of enterprise training, and is applied to skill improvement of employees of an enterprise.The method comprises the following steps: collecting operation behavior information of employees on a work platform; analyzing the operation behavior information, identifying skill characteristics reflecting the technical level of the employees, and updating the ability portrait of the employees according to the skill characteristics; in response to the update of the ability portrait, adjusting the learning path of the employees in combination with preset post skill demand information; and matching learning resources associated with the adjusted learning path from a dynamically updated learning resource library, and pushing the learning resources to the employees.The application can realize real-time, detailed and operation-based evaluation of the skills of the employees, ensure high personalization and dynamic adaptability of the learning path, can respond to changes in ability and evolution of demand in real time, and realize accurate matching and efficient utilization of learning resources through a knowledge graph.

Description

Technical Field

[0001] This application relates to the field of corporate training, and in particular to an AI-based personalized learning path planning and management method and system. Background Technology

[0002] In the field of corporate training, traditional learning path planning methods rely on fixed templates, making it difficult to adapt to the dynamic changes in employee capabilities and job requirements. This leads to a mismatch between training resources and actual needs, and delayed learning feedback. Existing solutions have low-frequency and coarse-grained assessments of employee capabilities, failing to accurately measure their practical skills. Furthermore, the rapid and continuous updates to job skill requirements mean that learning resource allocation is often based on outdated information, resulting in wasted resources and negatively impacting learning motivation. Simultaneously, the diverse backgrounds and learning preferences of employees mean that a uniform training model cannot meet individual needs, and the lag in learning effectiveness evaluation hinders timely adjustments to learning plans. A deeper problem lies in the inconsistent data interface standards and lack of real-time information exchange among various internal management systems (such as HR, performance, and training systems), resulting in a lack of accurate and timely data support for personalized planning. In addition, the vast amount of learning materials lacks unified knowledge point markings and connections to the skills system, making it difficult for employees to effectively search and match materials to their own skill gaps, further reducing resource utilization efficiency. Summary of the Invention

[0003] This application proposes an AI-based personalized learning path planning and management method and system, aiming to solve the technical problems in the field of corporate training. Traditional methods rely on fixed templates, which are difficult to adapt to differences in employee abilities and dynamic changes in positions, resulting in resource and demand mismatch, delayed feedback, and crude and outdated evaluation. Furthermore, the lack of data support for personalized planning and low efficiency in resource retrieval and matching are caused by data silos in the system and the lack of structured labeling of resources.

[0004] Firstly, this application provides an AI-based personalized learning path planning and management method for improving the skills of employees in an enterprise. The method includes the following steps: Collect information on employees' operational behavior on the work platform; The operational behavior information is analyzed to identify skill characteristics that reflect the employee's technical level, and the employee's competency profile is updated based on these skill characteristics. In response to the update of the competency profile and in combination with the preset job skill requirements information, the learning path of the employee is adjusted. From the dynamically updated learning resource library, matching learning resources associated with the adjusted learning path is performed, and the learning resources are pushed to the employee.

[0005] As some embodiments of this application, the operational behavior information includes at least one of the following: code development activity records, code version management records, code review collaboration records, project task processing records, knowledge sharing and communication records, command line operation records, configuration file change records, and distributed system monitoring interaction records.

[0006] As some embodiments of this application, the step of analyzing the operational behavior information, identifying skill characteristics reflecting the employee's technical level, and updating the employee's competency profile based on the skill characteristics includes: The operational behavior information is analyzed in real time to extract behavioral characteristics related to the employee's actual operational ability in a preset technical field; Based on the behavioral characteristics, quantifiable skill characteristics reflecting the employee's proficiency in various skills in the technical field are generated; the skill characteristics include at least one of the following: frequency of use of preset technical components, complexity score of operation tasks, task solving efficiency index, and quality score of content contributed in collaborative interactions. The skill characteristics are updated in real time to the employee's competency profile to reflect the employee's skill level.

[0007] As some embodiments of this application, the step of performing real-time analysis of the operational behavior information and extracting behavioral characteristics related to the employee's actual operational capabilities in a preset technical field includes: From the operational behavior information, extract operation records that are continuous in time and involve execution on at least two different work tools to form a behavior sequence; Analyze the behavior sequence to identify preset behavior patterns contained in the behavior sequence; Based on the behavioral pattern, behavioral process indicators are generated to characterize the problem-solving process of employees in a preset technical field. The behavioral process indicators include at least one of the following: the number of tools used, the number of modifications, and the number of verification operations. The behavioral process indicators are used as behavioral characteristics related to the employee's actual operational capabilities in the technical field.

[0008] As some embodiments of this application, the step of extracting operation records that are sequential in time and involve execution on at least two different work tools from the operation behavior information to form a behavior sequence includes: Based on the operational behavior information, the employee's operation records in multiple time segments are identified; wherein the operation records in each time segment are temporally continuous, and the operation in at least one of the time segments involves at least two different work tools; If the same technical entity is detected in multiple operation records, or if the operations all point to subtasks derived from the same project task, then the relevant operation records in the multiple time segments are associated and integrated across time segments to form a behavior sequence spanning multiple time segments; wherein, the technical entity includes code files, configuration files, fault documents or server nodes with unique identifiers.

[0009] As some embodiments of this application, the step of adjusting the employee's learning path in response to the update of the competency profile and in conjunction with preset job skill requirement information includes: Based on the updated competency profile, identify the current skill gaps of the employees; Based on the preset job skill requirements information, determine the skill goals that the employee needs to master in the future; Based on the skill gaps and skill goals, the employee's learning path is adjusted; wherein, the adjustment of the learning path includes: adjusting the priority of learning content, replacing or supplementing learning content, and adjusting the difficulty and depth of learning content.

[0010] As some embodiments of this application, the step of matching learning resources associated with the adjusted learning path from a dynamically updated learning resource library and pushing the learning resources to the employee includes: Content analysis is performed on the learning materials in the dynamically updated learning resource library to identify and extract the key knowledge points in the learning materials; The extracted knowledge points are associated with a preset skill system to establish a dynamically adjustable mapping relationship between the learning materials and the skill system. Based on the skill system included in the adjusted learning path, learning resources containing corresponding knowledge points are matched from the learning resource library according to the mapping relationship; The matched learning resources will be pushed to the employees.

[0011] As some embodiments of this application, the dynamic adjustment step of the mapping relationship includes: The interaction data of the employees with the pushed learning resources is obtained. The interaction data includes completion rate, learning duration, number of times the learning is repeated, and the dwell time and interaction event records of asking questions, marking and attempting exercises in the learning resources. Track the changes in the numerical values ​​of skill characteristics in the employee's competency profile corresponding to the skills associated with the learning resources within a preset time period after the learning resources are pushed; Based on the interaction data and the numerical changes of the skill characteristics, the effectiveness of the learning resource delivery is evaluated, and an evaluation result is obtained. Based on the evaluation results, the mapping relationship between the learning materials and the skill system is adjusted.

[0012] As some embodiments of this application, after the step of matching learning resources associated with the adjusted learning path from the dynamically updated learning resource library and pushing the learning resources to the employee, the following steps are further included: Based on the skill system and the project tasks or technical problems currently being handled identified from the employee's operational behavior information, employees with proficiency in the corresponding skills of the skill system that is higher than a preset proficiency threshold and whose current workload is lower than a preset workload threshold are selected from the employee capability profiles within the enterprise and used as matched human collaborator resources. The matched human collaborator resources are pushed to the employees.

[0013] Secondly, this application also provides an AI-based personalized learning path planning and management system for improving the skills of employees in enterprises. The system includes: The information acquisition module is used to collect information on employees' operational behavior on the work platform; The skills analysis module is used to analyze the operational behavior information, identify skill characteristics that reflect the employee's technical level, and update the employee's ability profile based on the skill characteristics. The path adjustment module is used to adjust the employee's learning path in response to the update of the competency profile and in combination with preset job skill requirements information. The resource push module is used to match learning resources associated with the adjusted learning path from a dynamically updated learning resource library, and push the learning resources to the employee.

[0014] The technical solution according to the embodiments of this application has at least the following beneficial effects: The technical solution of this application enables real-time, detailed, and hands-on assessment of employee skills, overcoming the shortcomings of traditional assessments that are lagging and crude. It ensures a highly personalized and dynamically adaptable learning path, enabling immediate responses to changes in capabilities and evolving needs. Furthermore, it achieves precise matching and efficient utilization of learning resources through knowledge graphs, solving core problems in traditional corporate training such as inaccurate assessments, resource mismatches, slow feedback, and insufficient personalized support due to data silos and a lack of structured labeling of resources.

[0015] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0016] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0017] Figure 1 This is a flowchart illustrating an AI-based personalized learning path planning and management method provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the architecture of an AI-based personalized learning path planning and management system provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] Traditional corporate training methods often rely on fixed templates, making it difficult to adapt to differences in employee capabilities and changing job requirements. This leads to a mismatch between training resources and actual employee needs, and untimely feedback on learning progress, severely impacting training efficiency. Current employee capability assessment methods are updated infrequently, making it difficult to accurately measure employees' practical skills. Furthermore, with business development and the emergence of new technologies, job responsibilities and skill requirements are constantly updated, and the existing assessment system cannot keep pace. This results in learning resource allocation based on outdated information, leading to resource waste and decreased employee motivation. In addition, employees within a company have diverse backgrounds, learning preferences, and paces, making it difficult for a uniform training model to meet individual needs. Learning effectiveness assessments are delayed, and managers cannot adjust plans in a timely manner. A deeper problem lies in the lack of data sharing between different management systems within the company, preventing the real-time integration of key employee information and hindering accurate data support for personalized learning path planning. To cope with increasing training demands and constantly evolving skill requirements, companies have introduced a large amount of learning materials, but these materials lack a unified knowledge point labeling and capability association system, making it difficult for employees to effectively retrieve and match information based on their own skill gaps, thus reducing the efficiency of learning resource utilization.

[0022] In this regard, such as Figure 1 As shown, this application discloses an AI-based personalized learning path planning and management method, applied to improve the skills of enterprise employees. The method includes the following steps: S110 collects information on employees' operational behavior on the work platform; S120, Analyze the operational behavior information, identify skill characteristics that reflect the employee's technical level, and update the employee's competency profile based on the skill characteristics; S130, in response to the update of the competency profile and in combination with the preset job skill requirements information, adjust the employee's learning path; S140: From the dynamically updated learning resource library, match the learning resources associated with the adjusted learning path, and push the learning resources to the employee.

[0023] To better understand the technical solutions proposed in this application, some key terms involved are explained below.

[0024] "Operational behavior information" refers to the recordable data generated by employees when using various work platforms and tools in their daily work, such as code commit records, document editing history, system logs, and task updates in project management tools. This information serves as an objective basis for assessing employees' actual skill levels.

[0025] "Skill characteristics" refer to indicators that quantify or qualitatively describe an employee's abilities and proficiency in a specific technical field, obtained through analysis of operational behavior information. Examples include code quality scores, problem-solving efficiency, and tool usage proficiency.

[0026] A "competency profile" is a comprehensive and dynamic description of an employee's various skill characteristics. It can fully reflect an employee's current skill level, strengths, and weaknesses. Competency profiles are the foundation for personalized learning path planning.

[0027] "Job skill requirements information" refers to the list of skills and proficiency requirements that employees are expected to master for a given position. This information is usually developed jointly by the human resources and technical departments and is updated as the business develops.

[0028] A "learning path" refers to a series of learning activities, courses, and resources planned for employees to help them improve their skills or achieve job skill goals. Learning paths are dynamically adjusted to adapt to changes in employee capabilities and needs.

[0029] A "learning resource repository" refers to a collection of various learning materials, courses, documents, videos, and other resources. The learning resource repository in this application is dynamically updated and can continuously incorporate new knowledge and technical content.

[0030] The AI-based personalized learning path planning and management method proposed in this application is based on a series of closely related steps to achieve accurate assessment of employee skills, intelligent adjustment of learning paths, and personalized delivery of learning resources.

[0031] There are several ways to collect information on employee behavior on work platforms. For example, data collection agents can be deployed to monitor employee actions in real time on tools such as integrated development environments (IDEs), version control systems (like Git), project management tools (like Jira), knowledge-sharing platforms (like Confluence), and command-line interfaces. These agents can record employee code commits, branch merging, task status updates, document editing, search queries, and command execution. Another approach is to interface with existing enterprise IT systems via APIs to periodically or in real-time retrieve structured operation logs and behavioral data from these systems. For example, this could include code commit history from code repository services, build and deployment records from CI / CD pipelines, and problem resolution processes from fault management systems. Furthermore, plugins or extensions can be embedded in the work platform to capture employee interactions in a non-intrusive manner, such as comments, suggestions, and adoption status in code review tools.

[0032] To update employee competency profiles, the following methods can be employed. For example, pre-trained Natural Language Processing (NLP) models can be used to perform semantic analysis on textual data such as code commit messages, code review comments, and project documents to extract keywords and themes, thereby identifying employee activity and contributions in relevant technical fields. Simultaneously, machine learning algorithms can be combined to perform pattern recognition on employee operation sequences across different work tools. For instance, this can identify the steps employees follow, the tool combinations they use, and their operational efficiency when solving predefined types of problems. Through these analyses, a series of quantifiable skill characteristics can be generated, such as the number of code commits, code complexity scores, defect fix rates, and new feature development efficiency in predefined programming languages ​​or frameworks. These skill characteristics are then used to update employee competency profiles. A competency profile can be a multi-dimensional vector representation, where each dimension represents a skill, and its value represents the proficiency level of that skill. When new skill characteristics are identified, the corresponding skill dimensions in the competency profile are updated. For example, algorithms such as weighted averaging and exponential smoothing can be used to integrate new data into historical data to reflect the latest changes in employee skills.

[0033] The following methods can be used to adjust employee learning paths. For example, when an employee's competency profile is updated, the system triggers a path adjustment process. This process first compares the updated competency profile with preset job skill requirements. Job skill requirements can be a skill tree or skill map, containing all the skills required for the job and their corresponding proficiency requirements. Through comparison, the system can identify the gap between the employee's current competency and the job requirements, i.e., skill gaps. For example, if the job requires the employee to be proficient in Python programming, but the competency profile shows that the employee's proficiency in Python is low, then Python programming skills will be identified as a gap. Based on these gaps, the system will dynamically adjust the employee's learning path. Adjustments may include: adding new learning modules to compensate for gaps, such as recommending advanced Python courses; adjusting the priority of existing learning content, bringing forward courses related to gaps; or adjusting the difficulty and depth of learning content, for example, if the employee has weak foundational knowledge in a certain area, recommending more basic introductory courses. Adjustments to learning paths can also consider the employee's learning preferences and historical learning data to ensure that the recommended learning path is both effective and attractive.

[0034] To match learning resources associated with the adjusted learning path from a dynamically updated learning resource library and push these resources to employees, the following methods can be used. For example, the learning resource library is a continuously updated knowledge base containing various forms of learning materials, such as online courses, technical documents, video tutorials, and practical projects. To achieve accurate matching, content analysis can be performed on all learning materials in the library to extract their core knowledge points and associate them with a pre-defined skill system. For instance, through keyword extraction and topic modeling, a video tutorial covering knowledge points such as "microservice architecture design" and "Docker containerization" can be identified and mapped to corresponding skills in the skill system. When an employee's learning path is adjusted, the system will query these mapping relationships based on the skill system included in the adjusted learning path and match learning resources containing the corresponding knowledge points from the learning resource library. For example, if the learning path includes the skill objective of "cloud computing security," the system will match all courses, documents, and cases related to "cloud computing security." The matched learning resources will then be pushed to employees through various channels such as the company's internal learning management system (LMS), email notifications, and instant messaging tools, ensuring that employees can access personalized learning content in a timely manner.

[0035] This application proposes an AI-based personalized learning path planning and management method, which works by constructing an intelligent closed loop of "perception-analysis-planning-allocation". The system first continuously collects operational behavior information from various work platforms used by employees daily, serving as a real-time, objective data source for skills assessment. Next, AI technology is used to analyze this behavioral information, extracting quantifiable skill characteristics and dynamically updating the employee's digital competency profile, thereby achieving detailed and continuous assessment of practical skills. Based on the real-time updated competency profile, the system compares it with preset job skill requirements, automatically identifying the employee's skill gaps and dynamically adjusting their learning path accordingly, including priority, content, and difficulty, ensuring that the learning plan matches the individual's weaknesses and job requirements in real time. Finally, the system conducts in-depth content analysis of materials in the learning resource library, establishing a mapping relationship between knowledge points and skill systems, forming a knowledge graph. This allows the system to accurately match and push the most relevant learning resources based on the adjusted learning path.

[0036] The technical solution of this application enables real-time, detailed, and hands-on assessment of employee skills, overcoming the shortcomings of traditional assessments that are lagging and crude. It ensures a highly personalized and dynamically adaptable learning path, enabling immediate responses to changes in capabilities and evolving needs. Furthermore, it achieves precise matching and efficient utilization of learning resources through knowledge graphs, solving core problems in traditional corporate training such as inaccurate assessments, resource mismatches, slow feedback, and insufficient personalized support due to data silos and a lack of structured labeling of resources.

[0037] It should be noted that the operational behavior information preferably includes at least one of the following: code development activity records, code version management records, code review collaboration records, project task processing records, knowledge sharing and communication records, command line operation records, configuration file change records, and distributed system monitoring interaction records.

[0038] Code development activity logs refer to various operation logs generated by employees during the software development process, such as detailed records of code writing, debugging, and testing. Their purpose is to reflect employees' programming skills and development efficiency.

[0039] Code version management records refer to the operation records such as commits, merges, and branch management generated by employees when using version control systems (such as Git and SVN). The purpose is to evaluate the employee's code management standards, collaboration ability, and proficiency in version control tools.

[0040] Code review collaboration logs refer to records of employees' comments, suggestions, modifications, and other interactive behaviors during the code review process. Their purpose is to reflect employees' code quality awareness, problem-finding ability, and teamwork level.

[0041] Project task processing records refer to the records of employees' operations in project management tools (such as Jira and Trello) such as task creation, assignment, status updates, and completion status. The purpose is to reflect employees' task management ability, project progress efficiency, and sense of responsibility.

[0042] Knowledge sharing and exchange records refer to the interactive records of employees creating, editing, commenting, asking questions, and answering questions in internal knowledge bases, forums, and instant messaging tools. The purpose is to evaluate employees' knowledge contributions, learning abilities, and communication and collaboration skills.

[0043] Command line operation logs refer to the records of various instructions and scripts executed by employees through the command line interface in the operating system or development environment. Their purpose is to reflect the employees' proficiency in low-level system operations, automated script writing, and troubleshooting.

[0044] Configuration file change logs refer to records of employee modifications and updates to system, application, or service configuration files. Their purpose is to demonstrate the employee's depth of understanding of system architecture, service configuration, and troubleshooting. Distributed system monitoring interaction logs refer to records of employee actions while monitoring distributed systems, including querying monitoring tools, handling alarms, and analyzing logs. Their purpose is to assess the employee's ability to maintain complex systems, locate faults, and optimize performance.

[0045] This application's solution, by collecting the aforementioned diverse types of operational behavior information, can capture employees' behavioral trajectories and technical performance in actual work from multiple dimensions and in greater detail. This rich and multi-source operational behavior information provides a solid data foundation for subsequent identification of employees' skill characteristics. Consequently, a more comprehensive and accurate employee competency profile can be constructed, providing a more reliable basis for subsequent adjustments to learning paths.

[0046] Compared to traditional methods that rely on a single or limited data source, this application integrates diverse information such as code development, version management, review collaboration, project tasks, knowledge sharing, command-line operations, configuration file changes, and distributed system monitoring. This allows for a more accurate portrayal of employees' proficiency in various skills, effectively avoiding assessment biases caused by data bias. Consequently, it provides more robust data support for employees' personalized learning path planning, ensuring the relevance and effectiveness of the learning path.

[0047] In some embodiments of this application, the step of analyzing operational behavior information, identifying skill characteristics reflecting employee technical levels, and updating employee competency profiles based on these skill characteristics preferably includes: The operational behavior information is analyzed in real time to extract behavioral characteristics related to the employee's actual operational ability in a preset technical field; Based on the behavioral characteristics, quantifiable skill characteristics reflecting the employee's proficiency in various skills in the technical field are generated; the skill characteristics include at least one of the following: frequency of use of preset technical components, complexity score of operation tasks, task solving efficiency index, and quality score of content contributed in collaborative interactions. The skill characteristics are updated in real time to the employee's competency profile to reflect the employee's skill level.

[0048] "Real-time analysis" refers to the continuous and immediate data processing and analysis of employees' operational behavior information on the work platform to ensure that the extracted behavioral characteristics can promptly reflect the employees' latest operational status and technical performance. In this way, specific behavioral patterns and ability tendencies exhibited by employees in predefined technical fields, such as software development and system maintenance, can be captured during actual operations. These behavioral characteristics are a direct reflection of employees' actual operational capabilities; for example, during code development, the frequency of employees calling relevant APIs, the average time spent troubleshooting, and their activity level in code reviews.

[0049] Based on the extracted behavioral characteristics, a series of quantifiable skill characteristics can be generated. These skill characteristics aim to objectively and accurately measure an employee's proficiency in various skills within a predefined technical field. Specifically, these skill characteristics may include, but are not limited to: Based on the usage frequency of preset technical components: generated through statistical analysis and contextual verification models. The system not only counts the absolute number of times employees call specific technical components (such as Spring Boot annotations, ReactHooks, and Kubernetes resource declarations) during code submissions, configuration modifications, and other operations, but also combines code review results, static code analysis reports (such as SonarQube), and runtime logs to verify the correctness of their use and their compliance with best practices.

[0050] Task complexity score: Generated through a multi-dimensional analysis model of task metadata and changesets. The system extracts multiple dimensions of data from tasks handled by employees (such as Jira Issues): technical tags (e.g., "distributed transactions," "performance optimization"), the scale of associated code changes (number of lines added / deleted, number of files involved), architectural impact (whether core interfaces or database schemas were modified), and the diversity of required skills (the number of different technology stack keywords identified in the task description). These dimensions are used to calculate a complexity score using a pre-trained regression model or a rule-based weighted scoring card. The average complexity of tasks successfully completed by employees constitutes a persistent indicator of their "ability to handle complex tasks."

[0051] Task resolution efficiency metrics: generated through time series analysis and workflow mining techniques. The system tracks the complete cycle from task start (status changes to "in progress") to completion (status changes to "resolved" or "closed") and analyzes the behavioral sequences within it.

[0052] Task resolution efficiency metrics may include: Average Repair / Resolve Time: Tracks the average time it takes for a task status to change from "In Progress" to "Resolved" or "Closed".

[0053] Delivery throughput and stability: Analyze the number of lines of code, feature points, or tasks submitted by employees that pass automated tests and code reviews on the first attempt within a specific period.

[0054] First-time fix success rate: This refers to the percentage of times an employee's initial solution to a defect or issue is verified as effective without requiring multiple iterations. This metric comprehensively reflects the accuracy of problem identification and the comprehensiveness of the solution. High-efficiency, high-quality, and stable output will correspond to a higher "Execution and Delivery Efficiency" skill trait value.

[0055] Quality score of content contributed in collaborative interactions: This indicator is generated through natural language processing and collaborative network analysis. It uses natural language processing technology and collaborative behavior analysis to quantify the value contribution of employees in team knowledge sharing and technical collaboration.

[0056] After the aforementioned skill characteristics are generated, the system typically normalizes them and maps them to a unified measurement system (e.g., a scale score of 0-100, or level labels such as "beginner," "intermediate," "advanced," and "expert"). For a macro-level skill dimension defined in the competency profile (e.g., "cloud-native application development"), its overall score can be calculated by fusing multiple related micro-level skill characteristics (e.g., "frequency of Kubernetes resource configuration and usage," "complexity of microservice troubleshooting tasks," and "quality of distributed system design review") according to a preset weight model. Ultimately, these continuously generated and updated quantifiable skill characteristics are integrated into the employee's dynamic competency profile in real time, providing a direct and reliable decision-making basis for subsequent personalized learning path planning and precise matching of learning resources.

[0057] Therefore, the generated skill characteristics are updated in real time to the employee's competency profile. The competency profile is a dynamic and comprehensive employee skills file that integrates information such as the employee's proficiency in various skills, learning preferences, and career development goals. By updating skill characteristics in real time, the competency profile can accurately and promptly reflect the employee's current skill level and development trend, providing a reliable data foundation for subsequent adjustments to learning paths.

[0058] This solution dynamically identifies employees' skill characteristics based on their actual operational data and presents them in a quantifiable manner, greatly improving the accuracy and reliability of skills assessment. Furthermore, the real-time updated competency profiles ensure timely identification of employees' skill weaknesses and strengths, enabling more targeted and timely adjustments to subsequent learning paths, thereby effectively improving employee skill levels and the company's overall competitiveness.

[0059] In a further embodiment of this application, the step of performing real-time analysis of operational behavior information and extracting behavioral characteristics related to the employee's actual operational capabilities in a preset technical field may include: Extract operation records that are sequential in time and involve execution on at least two different work tools from the operational behavior information to form a behavior sequence; Analyze the behavior sequence to identify preset behavior patterns contained in the behavior sequence; Based on the behavioral pattern, behavioral process indicators are generated to characterize the problem-solving process of employees in a preset technical field. The behavioral process indicators include at least one of the following: the number of tools used, the number of modifications, and the number of verification operations. The behavioral process indicators are used as behavioral characteristics related to the employee's actual operational capabilities in the technical field.

[0060] When generating a sequence of actions, the system filters through a vast amount of employee activity data on the work platform, identifying a series of closely linked action records. These records are not only sequential in time but must also involve the employee using at least two different work tools. For example, an employee might first write code in a code editor, then switch to a version control tool to commit the code, and finally use a command-line tool to compile and test it. These cross-tool, sequential actions are integrated into a sequence of actions to reflect the complete process by which the employee solves a specific problem or completes a task.

[0061] Next, the predefined behavioral patterns within the behavioral sequences are identified. These behavioral patterns are predefined sequence patterns that reflect predefined technical activities or problem-solving strategies. For example, a "debugging pattern" might manifest as frequent compilation, running, and logging operations after code modifications; a "refactoring pattern" might manifest as extensive adjustments to the code structure, accompanied by multiple tests. Pattern recognition algorithms can extract these meaningful patterns from complex behavioral sequences.

[0062] The pattern recognition algorithms mainly include the following two technical approaches: Rule-based template-based sequence matching: This method relies on a pre-defined behavioral pattern library. Each pattern in the library is defined as a rule template, explicitly specifying the key event type, the order of events, the maximum allowed time interval, and optional event attribute constraints. The system compares the input behavioral sequence with these templates, employing sequence pattern mining techniques such as the improved Apriori algorithm and PrefixSpan algorithm to find subsequences that completely or highly conform to the template definition. This method is logically clear and highly interpretable.

[0063] Sequence classification based on machine learning models: This method employs supervised learning. First, a large number of historical behavior sequences labeled with behavioral pattern categories (such as "debug," "refactor," and "exploration") by experts are used as the training set. Each atomic behavior event is converted into a numerical vector using word embedding technology. Then, a recurrent neural network (such as LSTM) or a Transformer encoder is used to model the event vector sequence to capture its long-term dependencies and contextual features. Finally, a classifier is trained to predict the probability that the input sequence belongs to each preset behavior pattern. This method can automatically learn complex and flexible pattern features, has better robustness to noise and individual differences in behavior sequences, and has stronger recognition capabilities.

[0064] Subsequently, behavioral process indicators are generated to characterize employees' problem-solving processes within a predefined technical field. These indicators aim to quantify the specific behavioral characteristics of employees when solving problems. For example, the number of tools used can reflect the breadth of tools used by employees when solving problems; the number of modifications can reflect the frequency of attempts and adjustments made by employees during the problem-solving process; and the number of verification operations can reflect the rigor with which employees verify solutions. These behavioral process indicators can characterize employees' actual operational capabilities from multiple dimensions.

[0065] Ultimately, these generated behavioral process indicators are used as behavioral characteristics related to employees' actual operational capabilities in the aforementioned technical fields. These behavioral characteristics are a quantitative reflection of employees' actual operational capabilities, providing a more refined and accurate data foundation for subsequent updates to capability profiles.

[0066] By performing refined sequence extraction and pattern recognition on operational behavior information, this application can more accurately and comprehensively capture employees' actual operational behaviors in complex technical tasks and quantify them into specific behavioral process indicators. Therefore, the extracted behavioral features can more realistically reflect employees' actual operational and problem-solving abilities within a pre-defined technical field, providing a more reliable and refined data foundation for subsequently generating quantifiable skill characteristics and updating capability profiles, thereby significantly improving the accuracy and effectiveness of employee skill assessment.

[0067] In a further embodiment of this application, the step of extracting operation records that are sequential in time and involve execution on at least two different work tools from the operation behavior information to form a behavior sequence may include: Based on the operational behavior information, the employee's operation records in multiple time segments are identified; wherein the operation records in each time segment are temporally continuous, and the operation in at least one of the time segments involves at least two different work tools; If the same technical entity is detected in multiple operation records, or if the operations all point to subtasks derived from the same project task, then the relevant operation records in the multiple time segments are associated and integrated across time segments to form a behavior sequence spanning multiple time segments; wherein, the technical entity includes code files, configuration files, fault documents or server nodes with unique identifiers.

[0068] Identifying employee operation records across multiple time segments refers to the system analyzing all operational information of employees on the work platform and dividing it into a series of segments with time boundaries. Operation records within each time segment are required to be temporally continuous, meaning that the employee's operations do not have significant interruptions within a single segment. Furthermore, to ensure that the identified operations possess a certain degree of complexity and cross-tool collaboration, operations within at least one of the time segments need to involve at least two different work tools. For example, an employee might, within a given time period, first write code in a code editor, then switch to a version control tool to commit the code, and finally run tests in a testing tool.

[0069] When the system detects that multiple different operation records all relate to the same technical entity, or that these operations all point to sub-tasks derived from the same project task, it triggers a cross-time-segment association and integration mechanism. For example, an employee might process a fault ticket in the morning, involving modifying a code file, and then continue processing the same fault ticket in the afternoon, modifying a related configuration file in another tool. Although these two operations occur in different time segments, because they are both associated with the same fault ticket or the same code file, the system will treat them as a continuous and logically complete sequence of behaviors. The technical entity can be understood as a resource or object with a unique identifier in the software development or operation and maintenance process, specifically including but not limited to code files, configuration files, fault tickets, or server nodes. These technical entities serve as anchor points for association, enabling the system to identify sets of operations that, even if not completely continuous in time, logically belong to the same task or problem-solving process. In this way, even if an employee's operations are interrupted or scattered across different time points and tools while handling a complex problem, the solution of this application can effectively integrate them into a complete sequence of behaviors, thereby more accurately reflecting the employee's actual workflow and problem-solving capabilities.

[0070] This application's solution, by introducing technical entities and project sub-tasks as the basis for association, can more comprehensively and accurately capture employees' actual operational processes, forming a more logically complete and semantically richer behavioral sequence. This not only improves the quality of behavioral sequence construction but also provides more reliable input for subsequent behavioral pattern recognition, thereby making the identification of employee skill characteristics more accurate. Ultimately, this enhances the effectiveness and relevance of AI personalized learning path planning, thus better supporting the skill improvement of enterprise employees.

[0071] In some embodiments of this application, the step of adjusting the employee's learning path in response to the update of the competency profile and in conjunction with preset job skill requirement information preferably includes: Based on the updated competency profile, identify the current skill gaps of the employees; Based on the preset job skill requirements information, determine the skill goals that the employee needs to master in the future; Based on the skill gaps and skill goals, the employee's learning path is adjusted; wherein, the adjustment of the learning path includes: adjusting the priority of learning content, replacing or supplementing learning content, and adjusting the difficulty and depth of learning content.

[0072] Based on the updated competency profile, identifying current skill gaps in employees involves the system analyzing the latest competency profile data, such as proficiency scores and mastery levels for various skills, and comparing it with industry standards, profiles of outstanding employees in the same position, or preset minimum skill requirements. This allows the system to accurately pinpoint the specific skills in which employees are lacking or have not reached the expected level. The purpose is to provide a clear starting point and direction for subsequent adjustments to the learning path.

[0073] By combining pre-defined job skill requirements with the determination of future skill goals for employees, the system not only considers current employee shortcomings but also references future skill requirements set by the company or industry for the corresponding positions. For example, a particular position might require mastering a new programming language or tool within the next year. By integrating this requirement information with employees' career development plans, forward-looking and practical skill enhancement goals can be set for them. The aim is to ensure that adjustments to the learning path not only address current deficiencies but also adapt to future development needs.

[0074] Adjusting employee learning paths based on skill gaps and skill goals involves redesigning those paths based on identified weaknesses and defined objectives. For example, if an employee has a weakness in Python programming, but the job objective requires mastering data analysis, then Python data analysis-related courses will be added to the learning path. Adjustments to the learning path can include various methods: prioritizing learning content, such as advancing courses directly related to skill gaps; replacing or supplementing learning content, such as adding new course modules or removing already mastered content; and adjusting the difficulty and depth of the learning content, such as recommending introductory courses for beginners and advanced project practice for those seeking more advanced skills. The aim is to make the learning path more personalized, efficient, and targeted.

[0075] This application, by clearly identifying skill gaps and setting skill goals, provides a solid basis and clear direction for adjusting learning paths. This not only ensures that employees' learning content closely matches their actual ability gap and future career development, avoiding unnecessary learning burdens, but also greatly improves the relevance and efficiency of learning by flexibly adjusting the priority of learning content, adding or replacing content, and adjusting the difficulty and depth. This accelerates the employee skills enhancement process and better meets the company's needs for refined talent management.

[0076] In a specific embodiment of this application, the step of matching learning resources associated with the adjusted learning path from a dynamically updated learning resource library and pushing the learning resources to the employee preferably includes: Content analysis is performed on the learning materials in the dynamically updated learning resource library to identify and extract the key knowledge points in the learning materials; The extracted knowledge points are associated with a preset skill system to establish a dynamically adjustable mapping relationship between the learning materials and the skill system. Based on the skill system included in the adjusted learning path, learning resources containing corresponding knowledge points are matched from the learning resource library according to the mapping relationship; The matched learning resources will be pushed to the employees.

[0077] Content analysis of learning materials in a dynamically updated learning resource repository refers to the in-depth analysis of various learning materials, such as documents, video courses, code examples, and online exercises, using technologies such as Natural Language Processing (NLP), text mining, image recognition, or video analysis. The goal is to accurately identify and extract the core concepts, key skills, knowledge structures, and logical relationships within these materials, forming structured knowledge points. For example, from a technical document on cloud computing, knowledge points such as "containerization technology," "serverless architecture," and "elastic scaling" can be extracted.

[0078] Associating extracted knowledge points with a pre-defined skill system involves mapping specific knowledge points identified from learning materials to a hierarchical skill system common within the enterprise or industry. This skill system can be a tree structure or graph, containing detailed divisions from macro-technical fields to micro-operational skills. Establishing the mapping relationship between the learning materials and the skill system aims to clarify which specific skills each learning material can improve in employees, and to what extent. This mapping relationship is dynamically adjusted, meaning that as new learning materials are added, existing materials are updated, or the skill system itself evolves, the mapping relationship will be updated and optimized in real time to maintain its accuracy and timeliness.

[0079] Based on the skill system included in the adjusted learning path, and according to the mapping relationship, learning resources containing corresponding knowledge points are matched from the learning resource library. This means that after an employee's competency profile is updated, resulting in an adjustment of their learning path, the path will clearly indicate the skill goals that the employee currently needs to master or improve. The system will then use the established dynamic mapping relationship to accurately filter learning materials from the vast learning resource library that effectively cover these skill points based on these skill goals. For example, if the learning path indicates that an employee needs to improve their "Python asynchronous programming" skill, the system will match all learning resources associated with the "Python asynchronous programming" knowledge point, such as relevant tutorials, code libraries, or practical projects.

[0080] Pushing matched learning resources to employees refers to proactively sending selected learning resources to employees through various channels, such as internal corporate learning platforms, email notifications, instant messaging tools, or mobile applications. The push can take the form of a personalized learning list, a recommended course list, or a direct learning link, ensuring that employees can access and begin learning promptly.

[0081] This application establishes a dynamically adjusted mapping relationship that can respond in real time to changes in learning resources and skill systems, ensuring that the pushed resources remain timely and relevant. This precise and dynamic matching mechanism enables employees to obtain learning content that truly matches their current skill level and development needs, thereby effectively addressing skill gaps, accelerating skill improvement, and greatly enhancing learning efficiency and experience.

[0082] In a more specific embodiment of this application, the dynamic adjustment step of the mapping relationship preferably includes: The interaction data of the employees with the pushed learning resources is obtained. The interaction data includes completion rate, learning duration, number of times the learning is repeated, and the dwell time and interaction event records of asking questions, marking and attempting exercises in the learning resources. Track the changes in the numerical values ​​of skill characteristics in the employee's competency profile corresponding to the skills associated with the learning resources within a preset time period after the learning resources are pushed; Based on the interaction data and the numerical changes of the skill characteristics, the effectiveness of the learning resource delivery is evaluated, and an evaluation result is obtained. Based on the evaluation results, the mapping relationship between the learning materials and the skill system is adjusted.

[0083] Acquiring employee interaction data on pushed learning resources refers to the system collecting all quantifiable and recordable behavioral information related to employee interactions with these resources during the learning process. Completion rate can be understood as the percentage of time an employee completes the entire learning resource, such as video viewing progress, document reading progress, or exercise completion rate. Learning duration refers to the total time an employee spends on the learning resources, reflecting the effort invested. Repeated learning frequency indicates how often an employee studies the same learning resource multiple times, potentially indicating the difficulty or importance of that knowledge point. Furthermore, the duration of time spent asking questions, marking items, and attempting exercises within the learning resources, along with recorded interaction events, can provide a more detailed reflection of specific points of confusion, focus, and practical attempts by employees during the learning process. For example, spending too much time on a code example, marking a concept multiple times, or repeatedly attempting a programming exercise. This interaction data collectively forms the basis for evaluating learning effectiveness.

[0084] Tracking changes in the numerical values ​​of skill characteristics corresponding to the skills associated with the learning resources within a preset time period after the resources are pushed means that the system continuously monitors the updates to the employee's competency profile. For example, if the pushed learning resources aim to improve an employee's "Python programming" skills, the system will observe whether the value of the "Python programming" skill characteristic in the employee's competency profile has increased, and by what extent, for a period of time after the resources are pushed. This tracking mechanism can directly reflect the contribution of the learning resources to the actual improvement of the employee's skill level.

[0085] Based on the interaction data and the numerical changes in the skill characteristics, the effectiveness of the learning resource delivery is evaluated to obtain an evaluation result. Specifically, machine learning models or preset evaluation rules can be used to comprehensively analyze the interaction data (such as high completion rate, moderate learning time, repeated learning of key points, effective questioning, etc.) and the numerical changes in skill characteristics (such as significant improvement in skill proficiency), thereby quantitatively evaluating the effectiveness of the learning resource in improving employee skills. For example, learning resources with high completion rate, active interaction, and significant improvement in skill characteristics will be evaluated as "excellent"; conversely, resources with low completion rate, little interaction, and no significant change in skill characteristics may be evaluated as "poor".

[0086] Based on the evaluation results, the mapping relationship between the learning materials and the skill system is adjusted. For example, if a learning resource is evaluated as having a significant effect on improving skills, the mapping weight between the learning resource and the skill can be increased, making it easier to match in future recommendations; if a learning resource is ineffective, its mapping weight can be reduced, or even its content and relevance to the skill system can be re-evaluated, and it can be corrected or removed.

[0087] Through the aforementioned technical solution, this application enables precise evaluation of the effectiveness of learning resource recommendations, thereby allowing the mapping relationship between learning materials and skill systems to more dynamically and accurately reflect the actual value of learning resources. This not only significantly improves the personalization and effectiveness of learning path planning, avoiding the repeated recommendation of ineffective or inefficient learning resources, but also continuously optimizes the quality of the learning resource library and recommendation strategies based on employees' actual learning feedback and skill development, ultimately promoting the rapid and efficient improvement of employees' skills.

[0088] The following is a specific example to illustrate this.

[0089] Suppose an employee is recommended an online course on "Cloud Computing Architecture Design" by the system. The system continuously collects interaction data from the employee during the course. For example, the employee completes 90% of the course, with a total study time of 15 hours, including three chapters studied twice, asking two questions in the discussion forum on "Microservice Architecture," and attempting the sample code for "Containerized Deployment" multiple times. Simultaneously, the system tracks the changes in the employee's skill profile for "Cloud Computing Architecture Design" and "Microservice Architecture" within one month of the course's recommendation. For instance, the employee's proficiency in "Cloud Computing Architecture Design" might increase from 60 to 75 points, and their proficiency in "Microservice Architecture" from 50 to 65 points. Based on this interaction data and changes in skill characteristics, the system evaluates the effectiveness of the course recommendation. For example, the system might determine that the course significantly improves the employee's cloud computing architecture design capabilities, but there is still room for improvement in microservice architecture. Based on this assessment, the system will adjust the mapping weight between the online course and the "cloud computing architecture design" skill system, giving it a higher priority when recommending it to employees with similar skill gaps in the future. At the same time, the system may also more closely link the microservice architecture part of the course with more in-depth microservice topic resources to make up for its deficiencies in the corresponding sub-skills.

[0090] In a preferred embodiment of this application, after the step of matching learning resources associated with the adjusted learning path from a dynamically updated learning resource library and pushing the learning resources to the employee, the following step is further included: Based on the skill system and the project tasks or technical problems currently being handled identified from the employee's operational behavior information, employees with proficiency in the corresponding skills of the skill system that is higher than a preset proficiency threshold and whose current workload is lower than a preset workload threshold are selected from the employee capability profiles within the enterprise and used as matched human collaborator resources. The matched human collaborator resources are pushed to the employees.

[0091] The skill system refers to a structured system associated with the learning materials, used to describe corresponding skill domains and knowledge points. It can be understood as a refined classification and labeling of the skills covered by the learning resources. The operational behavior information may include at least one of the following: employee's code development activity records on the work platform, code version management records, code review collaboration records, project task processing records, knowledge sharing and communication records, command-line operation records, configuration file change records, and distributed system monitoring interaction records. This information comprehensively reflects the employee's daily work status and technical practice. The currently processed project task or technical problem refers to the specific technical difficulties that the employee actually encounters and needs to solve during the work process, or the ongoing project work. This can be identified through analysis of operational behavior information, such as through keyword recognition and task management system integration. The employee competency profile within the enterprise is a comprehensive data model reflecting the employee's proficiency in various skills, constructed based on the employee's skill characteristics (such as the frequency of use of technical components, operational task complexity rating, task resolution efficiency indicators, and collaboration contribution quality rating). The proficiency threshold and workload threshold are pre-set values ​​used to quantify screening conditions, ensuring that the matched human collaborators not only possess the required skills but also have the ability to provide assistance. The aforementioned human collaborator resources refer to employees within the company who possess the corresponding skills and currently have the capacity to provide assistance. They can act as mentors, consultants, or collaborators to provide direct support to employees who need help.

[0092] Through the aforementioned technical solution, this application can provide employees with more comprehensive and personalized learning and growth support. This solution not only provides static learning resources but also introduces dynamic human collaborator resources, enabling employees to obtain immediate and precise expert guidance and collaboration when encountering practical operational difficulties. This significantly improves employees' efficiency in solving practical problems and accelerates the mastery and application of new skills, especially in complex or urgent project scenarios, effectively avoiding inefficiencies or project delays caused by a lack of immediate support. Furthermore, this solution helps promote knowledge sharing and a collaborative culture within the enterprise, building a mutually beneficial learning ecosystem, thereby comprehensively enhancing the enterprise's overall technological capabilities and innovation vitality.

[0093] The following is a specific example to illustrate this.

[0094] Suppose a software development engineer encounters a complex technical challenge related to distributed transaction processing while working on a new microservice architecture transformation project. This problem cannot be quickly solved by consulting existing learning resources such as documentation or videos. At this point, the engineer's actions on the work platform (e.g., frequently searching for related keywords, attempting various distributed transaction solutions in code without success, posting help requests on the internal forum, etc.) will be captured by the system in real time. The system will identify that the engineer's current project task is "microservice architecture transformation," and the technical problem encountered is "distributed transaction processing." Based on this, the system will identify the skill set required for this problem, such as "distributed system design" and "transaction consistency assurance." Subsequently, the system will filter from the competency profiles of all engineers within the enterprise those whose proficiency scores in skills such as "distributed system design" and "transaction consistency assurance" are above a preset threshold (e.g., a proficiency score of 90 or higher), and whose current workload (e.g., judged based on their project task list and schedule) is below a preset threshold (e.g., current project task saturation is below 70%). Suppose the system identifies engineer Zhang San, who has extensive experience in distributed systems and currently has available time. The system will then recommend engineer Zhang San as a human collaborator resource, sending a notification via internal communication tools (such as instant messaging or email) to the engineer facing the problem, suggesting they communicate and collaborate with Zhang San to jointly solve the distributed transaction processing challenge. In this way, the engineer can obtain immediate, targeted expert support, thereby quickly overcoming technical bottlenecks and efficiently completing the project task.

[0095] like Figure 2 As shown, this application also discloses an AI-based personalized learning path planning and management system 200, which is applied to improve the skills of employees in an enterprise. The system includes: Information acquisition module 210 is used to collect information on employees' operational behavior on the work platform; The skills analysis module 220 is used to analyze the operational behavior information, identify skill characteristics that reflect the employee's technical level, and update the employee's ability profile based on the skill characteristics. The path adjustment module 230 is used to adjust the learning path of the employee in response to the update of the competency profile and in combination with the preset job skill requirements information. The resource push module 240 is used to match learning resources associated with the adjusted learning path from a dynamically updated learning resource library, and push the learning resources to the employee.

[0096] The information acquisition module 210 can be configured to periodically retrieve raw operation logs from various work platforms used by employees (e.g., code repositories, project management systems, internal enterprise communication tools, etc.) using simple log capture tools or preset API interfaces. For example, the information acquisition module 210 can be set to pull data once at fixed time intervals (e.g., every hour or every day), or to synchronize data during off-peak hours through batch processing.

[0097] The skills analysis module 220 can be configured to use a simple rule-based matching algorithm to identify predefined operational behavior patterns and generate preliminary skill features based on the matching results. For example, the number of code submissions an employee makes in a programming language file can be used as a skill feature indicating their proficiency in that language. Subsequently, these skill features can be directly updated into the employee's competency profile.

[0098] The path adjustment module 230 can be configured to determine whether an employee has skill gaps after receiving a capability profile update notification, through a simple threshold comparison. For example, if the proficiency of a certain skill is below a preset passing grade, the learning content corresponding to that skill will be added to the learning path. The adjustment of the learning path can be limited to adding new learning content, without involving fine-tuning of priority or difficulty.

[0099] The resource push module 240 can be configured to search for learning materials in the learning resource library that match the skill names in the learning path using keyword matching, and then push these materials to employees. For example, if the learning path includes the skill "Java programming," then all learning resources with "Java programming" in their titles or tags will be pushed to employees.

[0100] It should be noted that the information interaction and execution process between the above modules are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0101] The preferred embodiments of this application have been described in detail above, but this application is not limited to the above embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application.

Claims

1. An AI-based personalized learning path planning and management method, applied to improve the skills of enterprise employees, characterized in that... The method includes the following steps: Collect information on employees' operational behavior on the work platform; The operational behavior information is analyzed to identify skill characteristics that reflect the employee's technical level, and the employee's competency profile is updated based on these skill characteristics. In response to the update of the competency profile and in combination with the preset job skill requirements information, the learning path of the employee is adjusted. From the dynamically updated learning resource library, matching learning resources associated with the adjusted learning path is performed, and the learning resources are pushed to the employee.

2. The AI-based personalized learning path planning and management method according to claim 1, characterized in that, The operational behavior information includes at least one of the following: code development activity records, code version management records, code review collaboration records, project task processing records, knowledge sharing and communication records, command line operation records, configuration file change records, and distributed system monitoring interaction records.

3. The AI-based personalized learning path planning and management method according to claim 1, characterized in that, The steps of analyzing the operational behavior information, identifying skill characteristics reflecting the employee's technical level, and updating the employee's competency profile based on the skill characteristics include: The operational behavior information is analyzed in real time to extract behavioral characteristics related to the employee's actual operational ability in a preset technical field; Based on the behavioral characteristics, quantifiable skill characteristics reflecting the employee's proficiency in various skills in the technical field are generated; the skill characteristics include at least one of the following: frequency of use of preset technical components, complexity score of operation tasks, task solving efficiency index, and quality score of content contributed in collaborative interactions. The skill characteristics are updated in real time to the employee's competency profile to reflect the employee's skill level.

4. The AI-based personalized learning path planning and management method according to claim 3, characterized in that, The step of performing real-time analysis of the operational behavior information and extracting behavioral characteristics related to the employee's actual operational capabilities in a preset technical field includes: From the operational behavior information, extract operation records that are continuous in time and involve execution on at least two different work tools to form a behavior sequence; Analyze the behavior sequence to identify preset behavior patterns contained in the behavior sequence; Based on the behavioral pattern, behavioral process indicators are generated to characterize the problem-solving process of employees in a preset technical field. The behavioral process indicators include at least one of the following: the number of tools used, the number of modifications, and the number of verification operations. The behavioral process indicators are used as behavioral characteristics related to the employee's actual operational capabilities in the technical field.

5. The AI-based personalized learning path planning and management method according to claim 4, characterized in that, The step of extracting operation records that are sequential in time and involve execution on at least two different work tools from the operation behavior information to form a behavior sequence includes: Based on the operational behavior information, the employee's operation records in multiple time segments are identified; wherein the operation records in each time segment are temporally continuous, and the operation in at least one of the time segments involves at least two different work tools; If the same technical entity is detected in multiple operation records, or if the operations all point to subtasks derived from the same project task, then the relevant operation records in the multiple time segments are associated and integrated across time segments to form a behavior sequence spanning multiple time segments; wherein, the technical entity includes code files, configuration files, fault documents or server nodes with unique identifiers.

6. The AI-based personalized learning path planning and management method according to claim 1, characterized in that, The step of adjusting the employee's learning path in response to the update of the competency profile and in conjunction with preset job skill requirement information includes: Based on the updated competency profile, identify the current skill gaps of the employees; Based on the preset job skill requirements information, determine the skill goals that the employee needs to master in the future; Based on the skill gaps and skill goals, the employee's learning path is adjusted; wherein, the adjustment of the learning path includes: adjusting the priority of learning content, replacing or supplementing learning content, and adjusting the difficulty and depth of learning content.

7. The AI-based personalized learning path planning and management method according to claim 1, characterized in that, The step of matching learning resources associated with the adjusted learning path from a dynamically updated learning resource library and pushing the learning resources to the employee includes: Content analysis is performed on the learning materials in the dynamically updated learning resource library to identify and extract the key knowledge points in the learning materials; The extracted knowledge points are associated with a preset skill system to establish a dynamically adjustable mapping relationship between the learning materials and the skill system. Based on the skill system included in the adjusted learning path, learning resources containing corresponding knowledge points are matched from the learning resource library according to the mapping relationship; The matched learning resources will be pushed to the employees.

8. The AI-based personalized learning path planning and management method according to claim 7, characterized in that, The dynamic adjustment steps of the mapping relationship include: The interaction data of the employees with the pushed learning resources is obtained. The interaction data includes completion rate, learning duration, number of times the learning is repeated, and the dwell time and interaction event records of asking questions, marking and attempting exercises in the learning resources. Track the changes in the numerical values ​​of skill characteristics in the employee's competency profile corresponding to the skills associated with the learning resources within a preset time period after the learning resources are pushed; Based on the interaction data and the numerical changes of the skill characteristics, the effectiveness of the learning resource delivery is evaluated, and an evaluation result is obtained. Based on the evaluation results, the mapping relationship between the learning materials and the skill system is adjusted.

9. The AI-based personalized learning path planning and management method according to claim 7, characterized in that, After the step of matching learning resources associated with the adjusted learning path from the dynamically updated learning resource library and pushing the learning resources to the employee, the method further includes the following steps: Based on the skill system and the project tasks or technical problems currently being handled identified from the employee's operational behavior information, employees with proficiency in the corresponding skills of the skill system that is higher than a preset proficiency threshold and whose current workload is lower than a preset workload threshold are selected from the employee capability profiles within the enterprise and used as matched human collaborator resources. The matched human collaborator resources are pushed to the employees.

10. An AI-based personalized learning path planning and management system, applied to improve the skills of enterprise employees, characterized in that... The system includes: The information acquisition module is used to collect information on employees' operational behavior on the work platform; The skills analysis module is used to analyze the operational behavior information, identify skill characteristics that reflect the employee's technical level, and update the employee's ability profile based on the skill characteristics. The path adjustment module is used to adjust the employee's learning path in response to the update of the competency profile and in combination with preset job skill requirements information. The resource push module is used to match learning resources associated with the adjusted learning path from a dynamically updated learning resource library, and push the learning resources to the employee.

Images