An electric power engineering safety learning platform course resource intelligent recommendation method and system

Abstract

The application provides a power engineering safety learning platform course resource intelligent recommendation method and system, and relates to the technical field of learning platform.The method comprises the following steps: collecting student information;correlating the student information with the element nodes in the pre-constructed structured correlation system;according to the corresponding nodes of the student correlation, identifying the safety knowledge point set related to the student by traversing the correlation path in the structured correlation system;adjusting the recommendation priority of the course resources associated with the safety knowledge point set according to the dynamic information of the student to obtain the adjusted recommendation priority;according to the adjusted recommendation priority, screening and sorting the course resources, generating a recommendation list and pushing it to the student.The method of the application can more accurately and timely identify the personalized safety learning needs and weak links of students than the traditional method, and dynamically recommend the most relevant courses, greatly improving the pertinence and effectiveness of learning.

Description

Technical Field

[0001] This invention relates to the field of learning platform technology, and more specifically, to a method and system for intelligent recommendation of course resources for a power engineering safety learning platform. Background Technology

[0002] Safety is of paramount importance in the power industry, and power companies generally establish power engineering safety learning platforms to improve the safety skills and awareness of their employees. These platforms gather a wealth of safety training course resources, including procedure learning, operational demonstrations, accident case analysis, and emergency drills. After logging into the platform, trainees are required to learn safety knowledge relevant to their job positions.

[0003] The field of power engineering has a wide variety of job types, and the safety knowledge requirements for trainees vary depending on the position. Line inspectors need to focus on knowledge such as safety when working at heights, prevention of electric shock, and risk avoidance when working in the field, while substation operators need to be familiar with the operation of power equipment, safety of switching operations, fire prevention and explosion protection, etc.

[0004] With a vast amount of course resources available on learning platforms, finding the most relevant and urgently needed content for one's job is a challenge for learners. Traditionally, learners manually search or browse course catalogs by job title, which is inefficient and prone to missing key courses. Even when platforms offer course lists categorized by job title, such categorization is often too general and fails to meet learners' refined learning needs.

[0005] The actual safety knowledge required by trainees depends not only on the job title but also on a variety of deeper factors, such as job responsibilities, equipment type, working environment, and regional differences in safety regulations. A trainee's career development stage also influences their choice of learning content. Newly hired trainees need to systematically learn basic theories, general safety procedures, and basic operating standards; experienced trainees may need to learn advanced safety knowledge for specific complex equipment or high-risk work scenarios; and experienced expert trainees may need to stay informed about the latest technological developments, cutting-edge accident prevention practices, or participate in safety management courses.

[0006] Safety regulations, national standards, and internal company rules in the power industry are constantly evolving, and trainees must keep abreast of the latest regulations relevant to their specific roles. When relevant regulations or standards change, the platform needs to identify which roles and trainees are affected and proactively recommend updated courses or supplementary learning materials.

[0007] Trainees' individual learning history and performance are also important bases for assessing their safety knowledge level and identifying weaknesses. Trainees' learning records on the platform, online assessment scores, feedback on safety issues encountered in actual work, and knowledge points related to similar historical accident cases can all reflect trainees' current knowledge mastery and potential safety risks.

[0008] Existing power engineering safety learning platforms often struggle to effectively integrate the aforementioned multi-dimensional and dynamically changing information when recommending course resources, failing to provide personalized, dynamic, and precise recommendations for each learner. This insufficient information matching means learners may spend time learning courses less relevant to their current work, missing out on crucial courses most closely related to their job responsibilities, actual working environment, experience level, and the latest safety requirements—courses that can best enhance their safety skills. This mismatch of learning resources not only affects learning efficiency and motivation but, more importantly, may fail to address learners' blind spots in specific areas of safety knowledge, creating potential safety hazards and hindering safe production in the power industry. Summary of the Invention

[0009] The purpose of this invention is to provide an intelligent recommendation method and system for course resources in a power engineering safety learning platform. By constructing a structured knowledge system and combining it with multi-dimensional dynamic information of students for correlation and priority adjustment, it can more accurately and timely identify students' personalized safety learning needs and weaknesses than traditional methods, and dynamically recommend the most relevant courses, greatly improving the pertinence and effectiveness of learning.

[0010] In a first aspect, the present invention provides an intelligent recommendation method for course resources on a power engineering safety learning platform, comprising the following steps:

[0011] Collect student information; student information includes students' dynamic information;

[0012] The trainee information is associated with the element nodes in a pre-built structured association system; the structured association system includes various power safety-related element nodes and association paths formed by connecting different element nodes according to specific association relationships;

[0013] Based on the corresponding nodes associated with the students, the set of security knowledge points related to the students is identified by traversing the association paths in the structured association system.

[0014] Based on the dynamic information of trainees, the recommendation priority of course resources associated with the set of safety knowledge points is adjusted to obtain the adjusted recommendation priority;

[0015] Based on the adjusted recommendation priority, course resources are filtered and sorted, and a recommendation list is generated and pushed to students.

[0016] The power engineering safety learning platform course resource intelligent recommendation method provided by the application, in the power engineering safety learning platform, in the face of the multi-dimensional information such as the post, the work environment, the equipment type, the learning history, the assessment performance, and the dynamically changing regulations and accident cases of the student, the precise and dynamic association between the individual safety knowledge demand of the student and the massive course resources is established, and the most relevant course is intelligently recommended accordingly, so that the personalized and dynamic learning demand that cannot be met by the traditional simple matching mode is solved.

[0017] In a second aspect, the application provides a power engineering safety learning platform course resource intelligent recommendation system, comprising:

[0018] The acquisition module is used for acquiring student information; the student information includes dynamic information of the student;

[0019] The association module is used for associating the student information with the element nodes in the pre-constructed structured association system; the structured association system includes a plurality of power safety related element nodes and an association path formed by connecting different element nodes according to a specific association relationship;

[0020] The identification module is used for identifying a set of safety knowledge points related to the student by traversing the association path in the structured association system according to the corresponding node associated with the student;

[0021] The adjustment module is used for adjusting the recommendation priority of the course resource associated with the set of safety knowledge points according to the dynamic information of the student, to obtain an adjusted recommendation priority;

[0022] The generation module is used for screening and sorting the course resources according to the adjusted recommendation priority, generating a recommendation list and pushing the recommendation list to the student.

[0023] As can be seen from the above, the power engineering safety learning platform course resource intelligent recommendation method provided by the application establishes a structured association system reflecting the internal logical relationship in the power engineering safety field, the system includes a plurality of power safety related element nodes and specific association relationships therebetween. The dynamic information of the student (such as learning history, assessment score, associated regulation change, and related accident case) is mapped to the corresponding node in the system. By traversing the association path in the system, a set of safety knowledge points most relevant to the current state and demand of the student is determined, and courses are screened, sorted and recommended from the course resource library based on these knowledge points.

[0024] Other features and advantages of the application will be set forth in the specification, and in part will become apparent to those skilled in the art upon reading the specification, or will be learned from the practice of the application. The objectives and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 A flow chart of the power engineering safety learning platform course resource intelligent recommendation method provided by the embodiment of the present application.

[0026] Figure 2 A structural schematic diagram of the power engineering safety learning platform course resource intelligent recommendation system provided by the embodiment of the present application.

[0027] Label explanation:

[0028] 100, acquisition module; 200, association module; 300, identification module; 400, adjustment module; 500, generation module. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0030] It should be noted that: similar labels and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0031] With reference to the accompanying drawings,the present application provides a power engineering safety learning platform course resource intelligent recommendation method, comprising the following steps: Figure 1

[0032] Acquiring student information; the student information includes static attribute information and dynamic information of the student;

[0033] Associating the student information with the element nodes in the pre-constructed structured association system; the structured association system includes a plurality of power safety related element nodes and an association path formed by connecting different element nodes according to a specific association relationship;

[0034] According to the corresponding nodes associated by the student, the set of safety knowledge points related to the student is identified by traversing the association path in the structured association system;

[0035] According to the dynamic information of the student, the recommendation priority of the course resource associated with the safety knowledge point set is adjusted to obtain an adjusted recommendation priority;

[0036] According to the adjusted recommendation priority, the course resource is screened and sorted, and a recommendation list is generated and pushed to the student.

[0037] Student information refers to data used to describe individual characteristics and behaviors of students, including static attribute information and dynamic information of students. Static attribute information can be represented by relatively stable data such as the post, department, length of service, and education level of the student, which is mainly to provide a basic portrait of the student. Dynamic information can be represented by data that changes over time, such as the student's learning record, test scores, browsing behavior, search keywords, and feedback on problems encountered in actual work, which is mainly to reflect the student's current learning status, knowledge mastery, and actual needs.

[0038] The structured association system refers to a pre-constructed knowledge network used to organize and associate power safety-related knowledge and elements, which includes various power safety-related element nodes and associated paths connected by different element nodes according to specific association relationships. Element nodes can be implemented by nodes representing concepts such as safety knowledge points, procedures, equipment, positions, accident cases, and work environments. Association paths can be implemented by edges representing semantic relationships such as "belongs to", "related to", "involves", and "needs to master", which is mainly to build a system that can reflect the complex knowledge structure of the power safety field, facilitating association discovery and reasoning.

[0039] Associating student information with element nodes in the pre-constructed structured association system means connecting the student individual with one or more element nodes in the structured association system, thereby placing the student in the knowledge system. This association can be achieved by directly matching student attributes with node labels, such as associating the student's position with the position node in the system; or by analyzing student behavior data and mapping it to relevant nodes, such as associating the student's browsed course content with the knowledge point node in the system, which is mainly to correspond the individual characteristics and behaviors of the student with specific content in the power safety knowledge system, laying a foundation for subsequent knowledge discovery.

[0040] The traversal of the association path in the structured association system refers to the process of exploring and accessing other element nodes along different types of association relationships in the system from the element node associated with the student. This process can be implemented using a graph traversal algorithm, such as depth-first search or breadth-first search, and can be guided by a pre-set strategy (such as the priority of different association types, node weight) to guide the traversal direction and depth, which is mainly to discover the safety knowledge points or other related elements hidden in the complex association relationship related to the student directly or indirectly.

[0041] According to the dynamic information of the student, adjusting the recommended priority of the course resources associated with the set of safety knowledge points refers to identifying the safety knowledge points related to the student, and using the latest behavior and state data of the student to correct the recommendation order or weight of the course resources corresponding to these knowledge points. This adjustment can use a dynamic information based on the student's recent learning behavior, test performance, actual problem feedback, etc. to calculate an influence factor, and apply it to the basic recommendation score of the course resources, which is mainly to make the recommendation results reflect the student's current most urgent learning needs, weaknesses or concerns in a timely manner, improve the timeliness and accuracy of the recommendation.

[0042] The working principle of the present application is to first construct a knowledge graph of the power engineering safety field, which contains various safety-related entities (such as posts, tasks, risks, procedures, knowledge points, equipment, environments, accidents, courses) and their pre-defined relationships. Then, the system collects the static information (such as post, equipment, environment) and dynamic information (such as learning progress, test score, procedure update, accident association) of the student, and associates these information with the corresponding entity nodes in the knowledge graph. When it is necessary to recommend courses for the student, the system starts from the node associated with the student, traverses along the association path in the graph, and preliminarily identifies the set of safety knowledge points related to the student. Then, the system dynamically adjusts the recommended priority of these knowledge points and their associated courses according to the student's learning history, test performance, and association with procedure changes and accident cases. Finally, the system filters and sorts the courses most suitable for the current needs of the student from the course resource library according to the adjusted priority, and presents them to the student. This process is dynamic and will be adjusted in real time as the student's state and external information (procedures, accidents) change.

[0043] The core innovation of the present application is to deeply integrate the multi-dimensional information of the student (including static attributes and dynamic behavior) with the pre-constructed structured power safety knowledge association system, and based on this system to carry out intelligent association discovery and dynamic recommendation priority adjustment, thereby solving the problem of effectively integrating multi-dimensional, dynamically changing information for personalized and precise course recommendation in the prior art, achieving the effect of improving the relevance, timeliness and effectiveness of the recommendation, and better meeting the individual learning needs of the student.

[0044] Specifically, the scheme of the present application constructs a comprehensive portrait of the trainee by collecting the static attribute information and dynamic information of the trainee. Then, the trainee information is associated with the element nodes in the pre-constructed structured association system, and the trainee individual is mapped into the power safety knowledge network. Based on the associated nodes of the trainee in the system, the relevant safety knowledge point set related to the trainee is systematically identified by traversing the associated path in the structured association system. This process utilizes the structure and association relationship of the knowledge system, and can discover the potential and indirectly related knowledge needs of the trainee. After identifying the relevant safety knowledge points, the course resources associated with these knowledge points are further obtained. The key is to dynamically adjust the recommendation priority of these course resources by using the dynamic information of the trainee, such as the recent learning progress, the knowledge blind spots exposed in the examination, the new problems encountered in the actual work, etc. This adjustment ensures that the recommended courses can respond to the current most urgent needs or the weakest links of the trainee in a timely manner. Finally, according to the dynamically adjusted recommendation priority, the course resources are screened and sorted to generate a personalized recommendation list and push it to the trainee. The whole process forms a closed loop, starting from the trainee information, deep mining through the knowledge system, real-time optimization combined with dynamic information, and finally realizing precise and timely course resource pushing.

[0045] The constructed structured association system of safety knowledge of electric power engineering can be stored and managed by a graph database. The node types in the system (such as "post", "operation task", "safety risk", "knowledge point", "course resource", etc.) represent entities in the graph, and the association relationship types (such as "undertake", "exist", "correspond", "contain", "explain", etc.) represent edges in the graph. After collecting the information of the students, the students are connected with the corresponding nodes in the graph database, for example, the student "Zhang San" is associated with the "high-voltage electrician" node, the "mountainous environment" node, and the "pole tower equipment" node. When it is necessary to identify the safety knowledge needs of Zhang San, the system starts from the node associated with Zhang San (such as "high-voltage electrician") and performs graph traversal query. For example, the query path can be set as: starting from the "post" node, following the "undertake" relationship to reach the "operation task" node, following the "exist" relationship to reach the "safety risk" node, then following the "correspond" relationship to reach the "regulation article" node, and finally following the "contain" relationship to reach the "knowledge point" node. At the same time, the "operation task" node can also be followed to reach the "equipment type" or "operation environment" node along the "involve" relationship, and then continue to traverse the related path from these nodes to reach the knowledge point. According to the set traversal rules and the starting node associated with the student, the graph database query engine performs traversal operation, collects all the "knowledge point" nodes reached, and forms a set of safety knowledge points related to Zhang San. For example, the traversal result may include "high-altitude operation safety", "correct use of safety belt", "electric shock prevention", "risk avoidance of mountainous field operation", "safety of pole tower structure", etc.

[0046] As a preferred embodiment, the scheme of the present application is implemented as follows: a student information database can be established to store the static attributes of students, such as their positions, departments, and length of service, as well as their dynamic behavior data on the learning platform, such as their browsing records, learning duration, test scores, and course completion status. At the same time, a power safety knowledge graph is constructed as a structured association system, in which the nodes can include specific safety regulation provisions, device models, operation steps, accident types, risk points, and position names, and the edges can represent various association relationships between these elements, such as "Position A needs to master Regulation B", "Device C's operation involves Step D", and "Accident E is related to Risk Point F". When the student logs in to the platform, the system first reads their static attributes and recent dynamic information from the database. Then, the student's position information is associated with the position nodes in the knowledge graph, and the course content the student has recently browsed or learned is associated with the knowledge point nodes in the graph. Next, starting from these associated nodes, the system traverses the knowledge graph according to pre-set traversal rules (e.g., preferentially traversing the "need to master" relationship, and then traversing the "related to" relationship), to identify a set of safety knowledge points related to the student's job responsibilities and recent learning content. For example, if the student is a substation attendant, the system will traverse to knowledge points such as substation equipment operation and switching operation safety. If the student has recently learned a course on a certain device, the system will further traverse to knowledge points such as detailed operation procedures for the device and related accident cases. After identifying the set of knowledge points, the system searches for course resources associated with these knowledge points. At the same time, the system analyzes the student's dynamic information, such as the student's low score in the recent switching operation simulation test or the student's feedback on a problem related to the operation of a certain device. Based on this dynamic information, the system increases the recommendation priority of course resources related to switching operation and the operation of the device. For example, the priority of the switching operation safety regulation course and the demonstration video of the operation of the device is increased. Finally, the course resource list is sorted according to the adjusted priority, and the top-ranked courses are generated into a recommendation list, which is displayed on the student's learning platform homepage or pushed to the student through a message.

[0047] Through the above scheme, the present application can effectively integrate the multi-dimensional information of students and the structured knowledge in the field of power safety, overcoming the limitations of traditional recommendation methods that rely on single-dimensional information or simple classification, and achieving personalized, dynamic, and precise course resource recommendation for each student, improving the relevance and efficiency of student learning, and helping to timely fill the knowledge gaps of students and improve the safety skills and awareness of power engineering practitioners, thereby better ensuring the safety production of the power industry.

[0048] In some embodiments, the step of associating the student information with the element nodes in the pre-constructed structured association system includes:

[0049] By analyzing static attribute information, identifying the static attribute elements of the trainees, and according to the identified static attribute elements, finding the element nodes corresponding to the static attribute elements in the structured association system as a first set, establishing the static association relationship between the trainee individual and the first set;

[0050] By analyzing dynamic information, identifying the dynamic elements of the trainees, and according to the identified dynamic elements, finding the element nodes related to the dynamic elements in the structured association system as a second set, establishing the dynamic association relationship between the trainee individual and the related second set; the dynamic elements include information type, information content and information time; the dynamic association relationship includes association strength information and association timeliness information.

[0051] "Analyzing static attribute information" refers to the structured processing of the relatively stable information of the trainees, such as extracting information such as post, department, length of service, etc. from the trainee's file. "Identifying the static attribute elements of the trainees" refers to determining the specific attribute values from the analyzed static attribute information, such as the post name "substation attendant", the department name "operation and maintenance department", etc. These elements are the basic description of the identity of the trainees. "Different granularity element nodes" refers to the nodes related to the same static attribute element in the structured association system, which can exist in different levels or degrees of refinement, such as "substation attendant job responsibilities" (coarse granularity), "substation equipment operation procedures" (medium granularity), and "110kV transformer inspection points" (fine granularity). "Static association relationship" refers to the relatively long-term stable connection between the trainee individual and these element nodes determined based on static attributes. "Analyzing dynamic information" refers to the processing of real-time or recent information that changes, such as learning records, test scores, safety hazard feedback, accident case learning, etc. "Identifying the dynamic elements of the trainees" refers to extracting specific events or content from dynamic information, such as "complete course《High Voltage Switch Operation Safety》", "test score 90 points", "submit hazard feedback: the grounding wire of a certain device is loose", "learn accident case: a substation fire accident". "Dynamic association relationship" refers to the connection between the trainee individual and these element nodes determined based on dynamic elements, which has timeliness and strength. "Association strength information" is a numerical value that measures the closeness of the association between dynamic elements and related element nodes, for example, an accident feedback may have a higher association strength than an ordinary learning record. "Association timeliness information" is a numerical value that measures the validity period of the dynamic association relationship, reflecting the influence of the time of the occurrence of the dynamic element on the current association relevance, for example, a recent event has a higher timeliness than a long time ago.

[0052] The scheme defines the specific method of associating the student information with the element nodes in the structured association system in detail, the core of which is to distinguish and process the static attribute information and dynamic information of the students, and to establish static association relationship and dynamic association relationship respectively, so that the learning needs of the students are more comprehensively and finely described. Specifically, first, by analyzing the static attribute information of the students, the relatively stable and basic static attribute elements of the students are identified. Then, according to these identified static attribute elements, the corresponding element nodes of different granularity in the pre-constructed structured association system are searched as a first set, and the static association relationship between the student individual and the first set is established. This static association relationship reflects the basic safety knowledge needs of the students based on their basic identity and background, and by searching nodes of different granularity, the basic association is more flexible and accurate, providing a stable basis for subsequent knowledge point identification. Secondly, by analyzing the dynamic information of the students, the dynamic elements of the students which change in real time and are personalized are identified. According to these identified dynamic elements, the element nodes related to the dynamic elements in the structured association system are searched as a second set, and the dynamic association relationship between the student individual and the related second set is established. This dynamic association relationship can capture the real-time learning state, weak link or focus of the students, and is an effective supplement to the static association. The dynamic elements include information type, information content and information time, which means that when establishing the dynamic association, not only the knowledge points themselves (information content) are considered, but also the type of information source (information type) and the time of information occurrence (information time). The dynamic association relationship contains association strength information and association timeliness information, making the dynamic association more fine and accurate, and being able to more truly reflect the most urgent and most relevant learning needs of the students. By distinguishing and processing static and dynamic information, and establishing dynamic association with strength and timeliness, the scheme can solve the problems of granularity mismatch between student information and element nodes in the structured association system, and the precision of association between student dynamic information and system nodes in the power engineering safety learning scene, so as to map the student individual to the structured knowledge system. This association method combining static attributes and dynamic information makes the mapping between the student individual and the element nodes in the structured association system more rich and accurate, provides a more solid and fine basis for subsequent knowledge point identification and course recommendation based on the mapping, and overcomes the limitations brought by relying only on coarse-grained static information for association.

[0053] As a specific implementation, the association of learner information and element nodes in the structured association system can be implemented according to the following steps. For example, for a learner, the static attribute information can include the post "high-voltage tester" and the department "test center". The system parses this information and identifies the static attribute elements "high-voltage tester" and "test center". Then, the nodes corresponding to "high-voltage tester" are found in the structured association system, and nodes such as "high-voltage tester job responsibilities" (coarse granularity), "high-voltage test technology" (medium granularity), and "insulation withstand test" (fine granularity) are found, which are taken as a first set, and the static association relationship between the learner and these nodes is established. At the same time, the nodes related to "test center" are found, such as "test equipment management" and "test safety regulations", which are also included in the first set and static association is established. Then, the system parses the dynamic information of the learner, such as that the learner has recently completed a course "characteristics test of mutual inductor", submitted a hidden danger feedback on test equipment failure, and an examination result. The system identifies dynamic elements such as "completed course "characteristics test of mutual inductor", "hidden danger feedback: test equipment failure", and "examination result". For "completed course "characteristics test of mutual inductor", the system finds nodes related to "characteristics test of mutual inductor" in the structured association system, such as "mutual inductor test technology" and "test data analysis", as a second set. According to the information type (learning record), the association strength calculation method is determined, and the association strength between the learner and these nodes is calculated. According to the information time (the time of completing the course), the association timeliness is calculated. For "hidden danger feedback: test equipment failure", the system finds nodes related to "test equipment failure handling" and "equipment maintenance safety", as a second set. According to the information type (hidden danger feedback), the association strength calculation method (which can be higher than the learning record) is determined, and the association strength is calculated. According to the information time (feedback time), the association timeliness is calculated. For "examination result", the system finds knowledge point nodes related to the examination content as a second set, and calculates the association strength and timeliness according to the information type (examination result) and time. Finally, the learner individual establishes static association with the nodes in the first set and dynamic association with the nodes in the second set, and each dynamic association includes the calculated association strength and association timeliness information.

[0054] By the technical means, the granularity mismatching problem of the student information and the element nodes in the structured association system can be effectively solved, and the nodes of different granularities are associated through the static attributes to provide the student with comprehensive basic knowledge association. Meanwhile, the dynamic association containing the strength and timeliness is established by analyzing the dynamic information, the accuracy of the student individual and the system node association is improved, and the current learning state and actual demand of the student can be more accurately reflected. Therefore, the student individual is more finely and accurately mapped to the structured knowledge system, and a solid foundation is laid for subsequent personalized knowledge point identification and course recommendation.

[0055] In some embodiments, according to the identified dynamic element, the element nodes related to the dynamic element in the structured association system are searched as the second set, and the step of establishing the dynamic association relationship between the student individual and the related second set includes:

[0056] According to the information content, the element nodes related to the information content in the structured association system are searched as the second set;

[0057] According to the information type, the association strength calculation method between the student individual and each element node in the second set is determined, and the association strength between the student individual and each element node in the second set is calculated based on the determined association strength calculation method;

[0058] According to the information time, the association timeliness calculation method between the student individual and each element node in the second set is determined, and the association timeliness between the student individual and each element node in the second set is calculated based on the determined association timeliness calculation method;

[0059] According to the association strength between the student individual and each element node in the second set, and the association timeliness between the student individual and each element node in the second set, the dynamic association relationship between the student individual and the related second set is established.

[0060] The association strength calculation manner refers to rules or models for determining how to quantify the close degree of the association between the behavior of the trainee and the relevant element node according to the type of the behavior (for example, browsing, searching, learning, examination, simulation operation, etc.), which can be implemented by using a preset weight table based on the behavior type, a function model based on the behavior frequency, or a mapping relationship based on the behavior result (such as examination score, operation score). The association strength refers to a numerical value reflecting the close degree of the association between the trainee and the specific element node, which is calculated according to the determined association strength calculation manner. The numerical value can be a real number within a specific range, and the larger the numerical value, the closer the association. The association timeliness calculation manner refers to rules or models for determining how to quantify the influence of the behavior of the trainee on the current association relationship in terms of the duration of the behavior, which can be implemented by using a decay function based on the time interval (for example, linear decay, exponential decay), a reinforcement factor based on a specific event (such as procedure update, accident occurrence), or different timeliness periods based on the behavior type. The association timeliness refers to a numerical value reflecting the effective or urgent degree of the association between the trainee and the specific element node, which is calculated according to the determined association timeliness calculation manner. The numerical value can be a real number within a specific range, and the larger the numerical value, the higher the timeliness or the more urgent the association.

[0061] The above scheme of the present application can realize the following working principle through the synergistic effect of each step: first, by analyzing the specific information content of the dynamic behavior of the trainee, the element node set related to the content is accurately found and determined as the second set in the pre-constructed structured association system, which ensures that the dynamic association relationship established subsequently is based on the behavior content actually occurred by the trainee. Then, according to the information type of the dynamic element, the system can intelligently select or determine the association strength calculation method between the trainee individual and each element node in the second set, and calculate the specific association strength value based on this. Different behavior types naturally reflect the differences in the trainee's attention to the relevant content, learning depth or mastery level, for example, one active search behavior may have a higher association strength than one passive browsing behavior, and one examination passing may have a higher association strength than one learning completion. By distinguishing the information type to calculate the association strength, the dynamic association relationship established can more accurately reflect the trainee's status in knowledge or skills. At the same time, according to the information time of the dynamic element, the system can determine the association timeliness calculation method between the trainee individual and each element node in the second set, and calculate the specific association timeliness value based on this. The occurrence time of the trainee's behavior is a direct indicator of his current interest and demand, and recent behavior is usually more reflective of the trainee's current dynamic changes and urgent needs than long-term behavior, especially for knowledge with strong timeliness in the power safety field (such as the latest regulations, recent accidents). By considering the information time to calculate the association timeliness, the dynamic association relationship established can reflect the influence of time decay or reinforcement, and preferentially reflect the trainee's recent dynamic changes and potential risk points. Finally, the dynamic association relationship between the trainee individual and the related second set is established by comprehensively utilizing the calculated association strength and association timeliness. This dynamic association relationship combining strength and timeliness can more comprehensively and accurately depict the trainee's current dynamic learning portrait, which not only reflects which knowledge points the trainee is interested in or the mastery level (through strength), but also reflects whether these interests or knowledge states occurred recently or earlier (through timeliness), thereby providing high-quality input with time dimension for more accurate identification of related safety knowledge points and recommendation of more suitable course resources. This dynamic association relationship with strength and timeliness is a further refinement and enhancement of the basic scheme of only establishing an association relationship, making the depiction of the trainee's portrait based on dynamic information more accurate and timely, especially suitable for the high requirements of the power safety field on information timeliness.

[0062] To make the embodiments of the present application clearer, the following will be described in combination with a specific example: assume that a student has completed an online examination on “substation switching operation risk analysis” on the platform, with a score of 60 (out of 100) and the examination completion time being 10 o’clock this morning. The system identifies the dynamic element, with the information type being “examination”, the information content being “substation switching operation risk analysis, score 60”, and the information time being “10 o’clock this morning”. First, according to the information content “substation switching operation risk analysis”, the system searches for element nodes related to the content in the structured association system, such as “substation”, “switching operation”, “risk analysis”, “operation regulation”, “accident case”, etc., taking these nodes as the second set. Then, according to the information type “examination” and “score 60”, the system determines the association strength calculation method. For example, the preset rule can be: the association strength of the examination type behavior is negatively correlated with the examination score, or positively correlated with the number of knowledge points not mastered. Assume that according to the score calculation, the student’s association strength with the nodes of “substation”, “switching operation”, “risk analysis”, etc. is calculated as 0.4 (range 0-1, 1 indicating the strongest). At the same time, according to the information time “10 o’clock this morning”, the system determines the association timeliness calculation method. For example, the preset rule can be: the timeliness factor of the behavior occurring on the same day is 1.0, and it decays by 0.1 every day. Since it is a behavior occurring on the same day, the timeliness factor is calculated as 1.0. Finally, the system combines the calculated association strength 0.4 and association timeliness 1.0 to establish the dynamic association relationship between the student individual and each element node in the second set of “substation”, “switching operation”, “risk analysis”, etc. This association relationship can be stored as a record, such as containing the student ID, element node ID, association strength value, association timeliness value, and update timestamp.

[0063] By the technical solution, the following technical effects can be achieved: the dynamic correlation relationship is established based on the content actually concerned or contacted by the trainee, the dynamic correlation relationship can reflect the difference in the attention degree or mastery level of the trainee to the related content by determining the correlation strength calculation method according to the information type and calculating the strength, the dynamic correlation relationship can reflect the time value of the trainee's behavior and preferentially reflect the recent dynamic change and urgent demand of the trainee by determining the correlation timeliness calculation method according to the information time and calculating the timeliness, the dynamic correlation relationship can more comprehensively and accurately depict the current dynamic learning portrait of the trainee by combining the correlation strength and the correlation timeliness, especially can reflect the time value of the information, so that the knowledge points related to the recent behavior can be given a higher timeliness weight, thereby obtaining a higher priority in subsequent recommendation, improving the accuracy and timeliness of the recommendation, and helping the trainee to timely pay attention to and learn the latest safety knowledge most related to the current state of the trainee, thereby reducing the potential safety risk.

[0064] In some embodiments, according to the corresponding nodes associated with the trainee, the step of identifying a set of safety knowledge points related to the trainee by traversing the correlation paths in the structured correlation system includes:

[0065] According to the static attribute information of the trainee, a traversal strategy parameter for the structured correlation system is determined; the traversal strategy parameter includes a traversal priority of different correlation relationship types and a weight of different element node types;

[0066] Based on the static correlation relationship between the trainee and the element nodes in the structured correlation system, starting from the element nodes associated with the trainee, the correlation paths in the structured correlation system are traversed according to the determined traversal strategy parameter;

[0067] In the traversal process, the relevance degree of the element nodes to the trainee is evaluated according to the accessed element node type, correlation path type and traversal strategy parameter;

[0068] According to the evaluated relevance degree, a set of safety knowledge points related to the trainee is identified.

[0069] The student static attribute information refers to the personal characteristic data of the student that is relatively stable and unchanging, which is registered in the learning platform or input by the administrator, and can be represented by the post, department, job type, area, responsible equipment type, and work experience. The structured association system refers to abstracting various types of information in the field of power safety, such as safety regulations, equipment models, work environments, risk types, accident cases, and safety knowledge points, as element nodes, and establishing the association relationship between these nodes to form a graph structure, which can be constructed and stored in the form of a knowledge graph, semantic network, or association database. The traversal strategy parameter refers to the rule set used to guide the traversal direction and evaluate the importance of the node when traversing the structured association system, which can be represented by a mapping table containing the association relationship type to the priority value and a mapping table containing the element node type to the weight value. The association relationship type refers to the category of the edge connecting different element nodes in the structured association system, which can be distinguished by predefined types such as "belongs to", "involves", "may cause", "prevention measures are", and "related equipment is". The element node type refers to the classification of different nodes in the structured association system, which can be distinguished by predefined types such as "safety regulations", "equipment", "work environment", "risk", "knowledge point", and "accident case". The traversal priority refers to the degree of priority selection or higher exploration weight given to different types of association relationships during traversal, which can be represented by a numerical value, with a larger value indicating a higher priority. The weight refers to the importance or influence degree of different types of element nodes themselves when evaluating the relevance of the element nodes to the student, which can be represented by a numerical value, with a larger value indicating a higher weight. The static association relationship refers to the relatively fixed connection between the student individual and certain element nodes in the structured association system based on the student's static attributes, which can be implemented by storing the correspondence between the student ID and the associated node ID in the database. The association path refers to a sequence of element nodes connected in sequence by association relationships in the structured association system. The relevance degree refers to the degree of close association between the element nodes in the structured association system and the student individual, which can be quantitatively represented by a calculated numerical value, with a larger value indicating a higher relevance.

[0070] The scheme solves the challenge of personalized knowledge point identification in a complex knowledge system by introducing a traversal strategy based on the static attributes of the trainee. First, according to the static attribute information of the trainee, the system can generate a set of traversal strategy parameters, which reflects the emphasis of the trainee's post, environment, and other factors on knowledge point demand. For example, for high-altitude operation personnel, the associated relationships and node types related to "high-altitude operation risk", "safety belt use", etc. will be given higher priority or weight. Then, based on the established static association starting point of the trainee and the knowledge system, the system conducts directional traversal according to the customized strategy parameters. This traversal method is no longer a blind full-map exploration, but rather a priority along the path highly related to the trainee's static attributes. At each step of the traversal, the system will dynamically calculate the relevance of the current element node to the individual trainee by combining the node type accessed, the association path type, and the preset strategy parameters. This evaluation process ensures that each relevant element encountered during traversal is quantitatively measured. Finally, the system filters out safety knowledge points that meet a certain threshold of relevance based on these evaluated relevance levels, forming a knowledge point set that highly matches the individual needs of the trainee. In this way, the scheme effectively integrates the trainee's static attribute information into the traversal and identification process of the knowledge system, making the identified knowledge point set more accurate and personalized, and avoiding interference from a large amount of irrelevant information. The combination of the traversal strategy guided by static attributes and the trainee's static association starting point makes the knowledge point identification process more targeted and efficient.

[0071] For example, assume that a trainee's static attribute information indicates that their post is "power line inspection personnel" and their work environment involves "outdoors" and "high altitude". The system determines the traversal strategy parameters based on these static attributes. For example, set the traversal priority of association relationship types such as "involving risk" and "required skills" to high, and set the weight of element node types such as "work environment", "equipment type", and "risk type" to high. Based on the static association established between this trainee and the "power line inspection personnel" element node, the system starts traversal from this node. During the traversal process, if a "high-altitude fall risk" node is encountered through a "risk-related" association relationship, since the association relationship type has high priority and the node type has high weight, the system will evaluate the relevance of the node to the trainee as high. If further connected to the "safety belt correct use" knowledge point node through the "prevention measure is" association relationship, the system will evaluate the relevance of the knowledge point to the trainee by combining the priority and weight of the path. In contrast, if it is connected to a node related to "substation operation procedures" through a low-priority association relationship, the relevance evaluation value will be lower. Finally, the system identifies a set of safety knowledge points such as "safety belt use specifications", "snake and insect bite prevention", and "high-altitude operation emergency handling" that are highly relevant to the trainee.

[0072] By determining the traversal strategy parameters for the structured association system according to the static attribute information of the trainee, and performing targeted traversal and correlation evaluation according to the strategy parameters, the scheme can identify a set of safety knowledge points highly relevant to the individual trainee from a complex power safety knowledge system. This solves the problem that traditional methods are difficult to accurately identify knowledge points according to the specific attributes of trainees, improves the accuracy and relevance of knowledge point identification, and thus can provide trainees with personalized learning content that is more in line with their job responsibilities and risk scenarios, improving learning efficiency and relevance.

[0073] In some embodiments, the step of identifying a set of safety knowledge points relevant to the trainee according to the corresponding node associated with the trainee by traversing the association path in the structured association system comprises:

[0074] Obtaining the experience level information of the trainee;

[0075] According to the static attribute information of the trainee and the experience level information of the trainee, determine the traversal strategy parameters for the structured association system; the traversal strategy parameters include the traversal priority of different association relationship types and the weight of different element node types;

[0076] Based on the static association relationship between the individual trainee and the element nodes in the structured association system, starting from the element nodes associated with the individual trainee, according to the determined traversal strategy parameters, traverse the association path in the structured association system;

[0077] During the traversal process, according to the accessed element node type, association path type, traversal strategy parameter and experience level information of the trainee, evaluate the relevance of the element node to the individual trainee;

[0078] According to the evaluated relevance and the experience level information of the trainee, identify a set of safety knowledge points relevant to the individual trainee, so that the granularity of the set of safety knowledge points matches the experience level of the trainee.

[0079] The experience level information refers to data reflecting the knowledge mastery, skill proficiency, or career development stage of the trainee in the field of power engineering safety. It can be achieved by using the trainee's length of service, title, post level, obtained professional certificates, historical training records, assessment results, or experience level evaluation results through the platform. The traversal strategy parameter refers to a set of rules or values used to guide the traversal process of the structured association system. It can be achieved by using a pre-set parameter table, dynamically calculated parameters based on machine learning models, or parameters set in combination with expert experience. The traversal priority of different association relationship types refers to setting different priority access orders or weights for various association relationships (such as "belongs to", "related to", "is a component", "is prerequisite knowledge", etc.) connecting element nodes in the traversal process. It can be achieved by using numerical weights, ordering lists, or rule-based dynamic adjustment mechanisms. The weight of different element node types refers to setting different importance or access tendency values for different types of element nodes (such as "post", "equipment", "operation procedures", "accident cases", "knowledge points", etc.) in the structured association system. It can be achieved by using a fixed weight table, dynamically adjusted weights according to context, or weights calculated based on node attributes. The relevance of the evaluation element node to the individual trainee refers to calculating the correlation degree between a certain element node in the structured association system and the learning needs or interests of a specific trainee. It can be achieved by using a calculation method based on path length and weight, a calculation method based on semantic similarity, or a calculation model combining trainee historical behavior data. The granularity of the safety knowledge point set matches the experience level of the trainee refers to the detailed, deep, and broad knowledge contained in the identified safety knowledge point set being adapted to the current experience level of the trainee. For example, basic and general knowledge points are recommended for trainees with less experience, and in-depth and specific knowledge points are recommended for experienced trainees. It can be achieved by filtering knowledge points according to experience level thresholds, adjusting relevance evaluation models according to experience levels, or expanding or aggregating knowledge points according to experience levels.

[0080] In the process of identifying the set of safety knowledge points related to the learner, the experience level information of the learner is first obtained, which reflects the current knowledge mastery and learning stage of the learner. Then, when determining the traversal strategy parameters for the structured association system, not only the static attribute information of the learner is considered, but also the experience level information of the learner is combined. This means that for learners with different experience levels, even if their static attributes are the same, the strategy for traversing the association system will be adjusted, thereby affecting the scope and focus of the traversal. Next, based on the static association relationship established between the learner individual and the system, starting from the association node, the structured association system is traversed according to the traversal strategy parameters combined with the experience level. In the traversal process, when evaluating the relevance of the accessed element nodes and the learner individual, in addition to considering the node and path types, traversal strategy parameters, the experience level information of the learner is also explicitly included. This makes the relevance evaluation more refined and can distinguish the demand degree of different experience level learners for the same knowledge point. Finally, when identifying the set of safety knowledge points according to the evaluated relevance, the experience level information of the learner is again combined, the key is to ensure that the granularity of the identified set of safety knowledge points can match the experience level of the learner. For example, for learners with less experience, more basic and general knowledge points may be identified; for experienced learners, more in-depth and professional knowledge points may be identified. By fully considering the experience level of the learner in the traversal strategy determination, relevance evaluation and final knowledge point identification process, this scheme can more accurately identify the safety knowledge points suitable for the current learning needs and cognitive level of the learner, thereby improving the effectiveness of the recommendation and the learning efficiency of the learner. This method introduces the key dimension of learner experience level on the basis of basic knowledge point identification based on static attributes, making the knowledge point identification process more intelligent and personalized, and effectively solving the problem of mismatching the granularity of knowledge demand of learners with different experience levels, thereby providing a more accurate knowledge basis for subsequent course resource recommendation.

[0081] For example, assume that a student's static attribute information shows that his / her post is "substation operator." If the student's experience level information shows that he / she is a "junior operator," the system obtains the experience level information. In determining the traversal strategy parameters, the system sets the traversal strategy parameters in combination with the static attribute "substation operator" and the experience level "junior," for example, sets the associated path priority of the nodes pointing to basic operation procedures, general safety common sense class nodes high, sets the associated path priority of the nodes pointing to complex equipment fault handling, advanced emergency plan class nodes low, and simultaneously reduces the weight of complex equipment nodes. The system starts from the "substation operator" node associated with the student and traverses the structured association system according to these parameters. In the traversal process, when the "transformer operation safety procedure" node is accessed, the system evaluates its relevance in combination with its node type, associated path type, traversal strategy parameters, and the student's "junior" experience level. The evaluation result can indicate that the node has high relevance, but for a junior student, more emphasis is placed on basic operation parts. When the "transformer fault diagnosis and handling" node is accessed, the evaluation result can indicate that it has low relevance or requires more basic prerequisite knowledge. Finally, according to the evaluated relevance and in combination with the "junior" experience level, the set of safety knowledge points identified by the system will mainly include "substation basic safety procedures," "basic points of switching operation," "common device safety operation specifications," and other knowledge points with a more basic granularity. Conversely, if the student's experience level information shows that he / she is a "senior operator," the system obtains the experience level information. In determining the traversal strategy parameters, the system sets different traversal strategy parameters in combination with "substation operator" and "senior" experience level, for example, sets the associated path priority of the nodes pointing to complex equipment fault handling, advanced emergency plan class nodes high, and increases the weight of complex equipment nodes. The system starts from the "substation operator" node and traverses according to these new parameters. In the traversal process, when the "transformer operation safety procedure" node is accessed, the relevance is evaluated by considering that a senior student can need to understand deeper principles or special situation handling. When the "transformer fault diagnosis and handling" node is accessed, the evaluation result can indicate that it has high relevance and matches the needs of a senior student. Finally, according to the evaluated relevance and in combination with the "senior" experience level, the set of safety knowledge points identified by the system will mainly include "advanced operation and maintenance of specific transformer models," "complex fault mode analysis and emergency handling," "substation risk assessment and management," and other knowledge points with a more in-depth granularity. In this way, the system can dynamically adjust the focus and depth of knowledge point identification according to the student's experience level, ensuring that the identified set of knowledge points better meets the student's current learning needs.

[0082] By obtaining the experience level information of the trainee and incorporating it into the determination of the traversal strategy parameters, the evaluation of the element node relevance, and the identification process of the final safety knowledge point set, the scheme can dynamically adjust the granularity of the identified safety knowledge point set according to the actual experience level of the trainee. This makes it possible to recommend basic and general knowledge points to trainees with insufficient experience and more in-depth and specific knowledge points to trainees with rich experience, thereby effectively solving the problem of mismatch between the granularity of safety knowledge required by trainees with different experience levels and improving the accuracy and applicability of the identified safety knowledge point set.

[0083] In some embodiments, the step of adjusting the recommendation priority of the course resources associated with the safety knowledge point set according to the dynamic information of the trainee includes:

[0084] obtaining course resource information associated with the safety knowledge point set;

[0085] determining the basic impact weight of the dynamic element on the relevant course resources according to the information type of the dynamic element;

[0086] evaluating the degree of association between the information content and the relevant course resources according to the information content of the dynamic element, and calculating a preliminary impact value based on the basic impact weight;

[0087] calculating a time decay factor of the dynamic element on the preliminary impact value according to the time information of the dynamic element;

[0088] combining the preliminary impact value and the time decay factor to obtain the final impact value of the dynamic element on the relevant course resources;

[0089] For multiple dynamic elements associated with the same course resource, the final impact values of the multiple dynamic elements are considered comprehensively, and a comprehensive priority adjustment value of the course resource is calculated according to a preset comprehensive rule;

[0090] adjusting the recommendation priority of the course resources associated with the safety knowledge point set using the comprehensive priority adjustment value to obtain the adjusted recommendation priority.

[0091] The information type of the dynamic element refers to the category to which the learner dynamic information belongs, such as learning behavior records, assessment results, procedure standard update notifications, accident case feedback, etc. It can be achieved by using preset classification tags or extracting category information from text through natural language processing technology. The basic influence weight refers to the preset influence size of different types of dynamic information on the priority of course resource recommendation. It can be achieved by using fixed values set by expert experience, weight values obtained based on historical data statistical analysis, or values obtained through machine learning model training. The information content refers to the specific carrying data of the dynamic information, such as specific assessment scores, procedure numbers and versions, accident description texts, etc. It can exist in the form of structured data fields or unstructured texts. The correlation degree refers to the relatedness and closeness of the information content and the specific course resource in terms of theme, scope or detail. It can be achieved by using keyword matching, semantic similarity calculation, ontology-based reasoning or artificial rule mapping. The preliminary influence value refers to the initial influence evaluation value of a single dynamic element on the relevant course resource without considering time factors. It can be calculated by multiplying the basic influence weight and the correlation degree or using a function relationship. The time information refers to the timestamp of the occurrence or recording of the dynamic event. It can be recorded in the standard date and time format. The time decay factor refers to the coefficient of the influence of dynamic elements weakening over time. It can be calculated by using a linear decay function, an exponential decay function or a piecewise function based on time difference. The final influence value refers to the actual influence evaluation value of a single dynamic element on the relevant course resource after considering time decay. It can be calculated by multiplying the preliminary influence value and the time decay factor. The preset comprehensive rule refers to the logic or algorithm used to integrate the influence of multiple dynamic elements on the same course resource. It can be achieved by using weighted summation, rule-based decision tree, priority sorting or more complex machine learning models. The comprehensive priority adjustment value refers to the overall adjustment amount of the priority of a specific course resource under the joint action of multiple dynamic elements. It can be calculated by applying the comprehensive rule to multiple final influence values. The adjusted recommendation priority refers to the final priority obtained by superimposing the comprehensive priority adjustment value on the basis of the original recommendation priority. It can be calculated by using simple numerical addition or more complex priority mapping functions.

[0092] The scheme obtains the course resource information associated with the set of safety knowledge points related to the trainee, and determines the object that needs to be adjusted in priority. Then, for each dynamic information of the trainee, first, a basic influence weight is determined according to the information type, which reflects the different indicative significance of different types of events (such as examination failure, procedure update) on learning needs. Then, the content of the dynamic information is analyzed in depth to evaluate the close degree of its association with specific course resources, and the content association is combined with the basic weight of the information type to calculate a preliminary influence value, which reflects the effective use of the content of the dynamic information. Considering that the influence of dynamic information will decay over time, the scheme further calculates a time decay factor according to the time of the dynamic information, and applies it to the preliminary influence value to obtain the final influence value considering the timeliness. Since the trainee may have multiple dynamic information related to the same course, the scheme provides a mechanism to consider the final influence value of these single dynamic elements, and calculates the priority adjustment amount that the course resource should obtain as a whole according to the preset rules. Finally, the comprehensive priority adjustment value calculated is applied to the original priority of the course resource, thereby obtaining the recommended priority reflecting the most real and urgent needs of the trainee. By refining the influence of dynamic information into type, content, time and comprehensive consideration of multiple factors, and applying it to the course resources related to the trainee identified in the previous step, the scheme can overcome the shortcomings of simply processing dynamic information, achieve fine and dynamic adjustment of the priority of course recommendation, and make the final recommended result more in line with the actual situation and needs of the trainee.

[0093] For example, assume that the learner Xiao Wang recently has two dynamic information related to the course resource of "High-altitude Operation Safety Regulations": one is that he failed an online examination about the regulations three months ago with a score of 40; the other is that the latest revised version of the regulations was released last week. The system first acquires the information of the course resource related to "High-altitude Operation Safety Regulations", such as the course "Interpretation of New Version of High-altitude Operation Safety Regulations". For the dynamic element of failed examination, its information type is "examination result (negative)", information content is "score 40", and time information is "three months ago". The system determines a higher negative basic influence weight according to the "examination result (negative)" type, for example -0.8. According to the "score 40", the system evaluates the relevance degree with the course "Interpretation of New Version of High-altitude Operation Safety Regulations", for example, through the gap between the score and the passing line to determine that the relevance degree is high, and calculates the preliminary influence value, for example -0.8*0.9=-0.72. According to the time information "three months ago", the system calculates a timeliness decay factor, for example 0.5 (indicating that the influence is halved after three months). Combining the preliminary influence value with the timeliness decay factor, the final influence value of the examination dynamic element is obtained: -0.72*0.5=-0.36. For the dynamic element of regulations revision, its information type is "regulations update (important)", information content is "New Version of High-altitude Operation Safety Regulations", and time information is "last week". The system determines a higher positive basic influence weight according to the "regulations update (important)" type, for example +1.0. According to "New Version of High-altitude Operation Safety Regulations", the system evaluates the relevance degree with the course "Interpretation of New Version of High-altitude Operation Safety Regulations", and the relevance degree is extremely high, and calculates the preliminary influence value, for example +1.0*1.0=+1.0. According to the time information "last week", the system calculates a higher timeliness decay factor, for example 0.95 (indicating that the influence is very small after last week). Combining the preliminary influence value with the timeliness decay factor, the final influence value of the regulations update dynamic element is obtained: +1.0*0.95=+0.95. For the course "Interpretation of New Version of High-altitude Operation Safety Regulations", there are two associated dynamic elements. The system considers the two final influence values of -0.36 and +0.95 according to the preset comprehensive rule (for example, recent important regulations update is prior to earlier examination result, or adopts weighted summation, and recent event weight is higher). If the weighted summation rule is adopted, and the weight of recent event is higher, the comprehensive priority adjustment value is calculated, for example 0.95*0.7+(-0.36)*0.3=0.665-0.108=+0.557. Finally, the original recommended priority of the course "Interpretation of New Version of High-altitude Operation Safety Regulations" is adjusted by using the comprehensive priority adjustment value +0.557 to obtain the adjusted recommended priority.

[0094] By the above technical means, the scheme can analyze various dynamic information of the students in detail, including the nature, content relevance and occurrence time, and provide a mechanism to comprehensively process the influence of multiple dynamic factors on the same course, even if these factors may seem contradictory. This enables the system to more accurately assess the current real learning needs and knowledge weaknesses of the students, especially to respond to important and time-sensitive events such as procedural updates in a timely manner, so as to generate a more personalized, dynamic and accurate course recommendation list, effectively solving the problem of inaccurate recommendation caused by simply processing dynamic information.

[0095] Reference is made to the accompanying drawings Figure 2 The present application provides an electric power engineering safety learning platform course resource intelligent recommendation system, comprising:

[0096] The acquisition module 100 is configured to acquire student information, wherein the student information comprises dynamic information of the students.

[0097] The association module 200 is configured to associate the student information with element nodes in a pre-constructed structured association system; the structured association system comprises a plurality of power safety related element nodes and an association path formed by connecting different element nodes according to a specific association relationship.

[0098] The identification module 300 is configured to identify a set of safety knowledge points related to the students by traversing the association path in the structured association system according to the corresponding nodes associated with the students.

[0099] The adjustment module 400 is configured to adjust the recommendation priority of the course resources associated with the set of safety knowledge points according to the dynamic information of the students, to obtain an adjusted recommendation priority.

[0100] The generation module 500 is configured to filter and sort the course resources according to the adjusted recommendation priority, and generate and push a recommendation list to the students.

[0101] In this document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any such actual relationship or order between such entities or actions.

[0102] The above description is only an embodiment of the present application and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A power engineering safety learning platform course resource intelligent recommendation method, characterized in that, The method comprises the following steps: Collecting student information; the student information comprises dynamic information and static attribute information of the student; Associating the student information with element nodes in a pre-constructed structured association system; the structured association system comprises a plurality of power safety-related element nodes and an association path formed by connecting different element nodes according to a specific association relationship; According to the corresponding nodes associated with the student, the set of safety knowledge points related to the student is identified by traversing the association path in the structured association system; According to the dynamic information of the student, the recommendation priority of the course resources associated with the set of safety knowledge points is adjusted to obtain an adjusted recommendation priority; According to the adjusted recommendation priority, the course resources are screened and sorted, and a recommendation list is generated and pushed to the student; The step of associating the student information with the element nodes in the pre-constructed structured association system comprises: By analyzing the static attribute information, the static attribute elements of the student are identified, and according to the identified static attribute elements, the element nodes corresponding to the static attribute elements of different granularities in the structured association system are found as a first set, and a static association relationship between the student individual and the first set is established; By analyzing the dynamic information, the dynamic elements of the student are identified, and according to the identified dynamic elements, the element nodes related to the dynamic elements in the structured association system are found as a second set, and a dynamic association relationship between the student individual and the related second set is established; the dynamic association relationship comprises association strength information and association timeliness information; the dynamic elements comprise information type, information content and information time.

2. The method of claim 1, wherein, The step of finding the element nodes related to the dynamic elements in the structured association system as a second set according to the identified dynamic elements, and establishing a dynamic association relationship between the student individual and the related second set comprises: According to the information content, the element nodes related to the information content in the structured association system are found as the second set; According to the information type, the association strength calculation method between the student individual and each element node in the second set is determined, and the association strength between the student individual and each element node in the second set is calculated based on the determined association strength calculation method; According to the information time, the association timeliness calculation method between the student individual and each element node in the second set is determined, and the association timeliness between the student individual and each element node in the second set is calculated based on the determined association timeliness calculation method; According to the association strength between the student individual and each element node in the second set, and the association timeliness between the student individual and each element node in the second set, the dynamic association relationship between the student individual and the related second set is established. 3.The method of claim 1, wherein, The step of identifying the set of safety knowledge points related to the student according to the corresponding nodes associated with the student, by traversing the association path in the structured association system, comprises: According to the static attribute information of the student, the traversal strategy parameters of the structured association system are determined; Based on the static association relationship between the student individual and the element nodes in the structured association system, the element nodes associated with the student individual are started, and the association path in the structured association system is traversed according to the determined traversal strategy parameters; In the traversal process, the relevance degree of the element node to the individual learner is evaluated according to the accessed element node type, the associated path type and the traversal strategy parameter; According to the evaluated relevance degree, a set of safety knowledge points related to the individual learner is identified.

4. The method of claim 1, wherein, The step of identifying the set of safety knowledge points related to the individual learner according to the corresponding node associated with the learner by traversing the associated path in the structured association system includes: Obtaining the experience level information of the learner; According to the static attribute information of the learner and the experience level information of the learner, determine the traversal strategy parameter for the structured association system; Based on the static association relationship between the individual learner and the element node in the structured association system, starting from the element node associated with the individual learner, traversing the associated path in the structured association system according to the determined traversal strategy parameter; In the traversal process, the relevance degree of the element node to the individual learner is evaluated according to the accessed element node type, the associated path type and the traversal strategy parameter; According to the evaluated relevance degree and the experience level information of the learner, a set of safety knowledge points related to the individual learner is identified.

5. The method of claim 3 or 4, wherein, The traversal strategy parameter includes the traversal priority of different association relationship types and the weight of different element node types.

6. The method of claim 1, wherein, According to the dynamic information of the learner, the recommendation priority of the course resource associated with the set of safety knowledge points is adjusted to obtain the adjusted recommendation priority, which includes: Obtaining the course resource information associated with the set of safety knowledge points; According to the information type of the dynamic element, determine the basic influence weight of the dynamic element on the related course resource; According to the information content of the dynamic element, evaluate the relevance degree of the information content and the related course resource, and calculate the preliminary influence value based on the basic influence weight; According to the time information of the dynamic element, calculate the timeliness decay factor of the dynamic element on the preliminary influence value; Combine the preliminary influence value and the timeliness decay factor to obtain the final influence value of the dynamic element on the related course resource; For multiple dynamic elements associated with the same course resource, the final influence values of the multiple dynamic elements are considered comprehensively, and the comprehensive priority adjustment value of the course resource is calculated according to the preset comprehensive rule; The recommendation priority of the course resource associated with the set of safety knowledge points is adjusted by using the comprehensive priority adjustment value to obtain the adjusted recommendation priority.

7. An intelligent recommendation system for power engineering safety learning platform course resources, which adopts the method according to any one of claims 1-6. It includes: The acquisition module is used to acquire learner information; The learner information includes the dynamic information of the learner; The association module is used to associate the learner information with the element nodes in the pre-constructed structured association system; the structured association system includes multiple power safety related element nodes and associated paths connected by different element nodes according to specific association relationships; The identification module is used to identify the set of safety knowledge points related to the learner by traversing the associated path in the structured association system according to the corresponding node associated with the learner; The adjustment module is used to adjust the recommendation priority of the course resource associated with the set of safety knowledge points according to the dynamic information of the learner to obtain the adjusted recommendation priority; The generation module is used to filter and sort the course resources according to the adjusted recommendation priority, generate a recommendation list and push it to the learner.

Images