A human resource management method and system

Abstract

The application discloses a kind of human resource management method and system, it is related to human resource management technical field, comprising: obtaining the initial professional anchor point set of target organization, each anchor point corresponds to a professional role, and the responsibility text section of professional role and its professional semantic feature are associated;Based on workflow system, capture the professional behavior trace generated in the process of executing work task by member;Based on the professional behavior trace captured, the dynamic mapping operation of professional anchor point is executed, the derived professional anchor point corresponding to the professional behavior trace is generated in the professional anchor point system, and the professional evolution path between derived professional anchor point and initial professional anchor point is established;Based on the professional deployment intention of organization manager, determine the target member and target professional anchor point matched with professional deployment intention semantics from professional anchor point system, and generate professional recommendation according to the professional evolution path associated with target member. By accessing workflow system, real behavior data support is provided for function matching.

Description

Technical Field

[0001] This invention relates to the field of human resource management technology, specifically to a human resource management method and system. Background Technology

[0002] Traditional human resource management methods primarily rely on static job descriptions, job level systems, and regular performance evaluations. Managers define the scope of responsibilities, qualifications, and key performance indicators for each position through job descriptions, and evaluate and allocate employees through interviews, annual reviews, and training. This static management model can maintain basic person-job fit during periods of relative organizational stability. However, with the rapid changes in modern organizational business, increasingly complex team collaborations, and the continuous evolution of employee capabilities, traditional job descriptions struggle to reflect the dynamic functions employees perform in actual work. Employees often assume functions beyond the scope of their job descriptions during cross-departmental collaborations, ad-hoc projects, or process optimization. Current technologies cannot capture these actual work contents, leading to discrepancies between person-job fit assessments and reality. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a human resource management method and system.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a human resource management method, specifically comprising the following steps: S1. Obtain the initial set of occupational anchor points for the target organization. The initial set of occupational anchor points is obtained by parsing the job structure and job description of the target organization. Each anchor point corresponds to an occupational role and is associated with the job responsibility text segment and its occupational semantic features. S2. Based on the workflow system, capture the traces of professional behavior generated by members during the execution of work tasks. The traces of professional behavior refer to the operation records, collaboration records or output records left by members when handling affairs. S3. Based on the captured occupational behavior traces, perform a dynamic mapping operation of occupational anchors. The dynamic mapping operation includes: generating a derivative occupational anchor corresponding to the occupational behavior trace in the occupational anchor system according to the semantic correlation between the occupational behavior trace and the initial occupational anchor, and establishing an occupational evolution path between the derivative occupational anchor and the initial occupational anchor. S4. Based on the career allocation intentions of organizational managers, determine the target members and target career anchors that semantically match the career allocation intentions from the career anchor system, and generate career recommendations based on the career evolution paths associated with the target members.

[0005] Preferably, in step S1, the job structure data and job description text of the target organization are obtained. The job structure data includes departmental hierarchy, job titles, and reporting relationships. The job description text includes job descriptions, qualification requirements, and key performance indicators for each job. After obtaining the job structure data, a career decision tree is constructed based on the job structure data. The root node of the career decision tree represents the organization's overall goal, the middle nodes represent departmental jobs, and the leaf nodes represent job roles. After the career decision tree is constructed, for each leaf node in the career decision tree, extract the corresponding job description text. The job description text includes the tasks to be performed by the position, the collaborating objects, and the types of deliverables. After extracting the job description text corresponding to each leaf node, the job description text is subjected to occupational semantic feature extraction to generate the initial occupational anchor point corresponding to the leaf node. Occupational semantic feature extraction includes: natural language preprocessing of the job description text, which includes word segmentation, entity recognition and syntactic analysis. Based on the preprocessing results, the core occupational elements of the job description text are extracted. The core occupational elements include task type, collaboration object type and output type. Based on the extracted core occupational elements, the occupational semantic feature vector of the job description text is generated. The position information of the leaf node in the occupational decision tree, the job description text corresponding to the leaf node and the generated occupational semantic feature vector are associated and stored together to form the initial occupational anchor point. After traversing all leaf nodes of the career decision tree and generating corresponding initial career anchors, the initial career anchors corresponding to all leaf nodes are collected to form an initial career anchor set. Each initial career anchor in the initial career anchor set is then bound to the corresponding leaf node in the career decision tree through its stored location information.

[0006] Preferably, in step S2, the workflow system is accessed, and operation records, collaboration records, and output records generated by members in the workflow system are obtained. The obtained operation records, collaboration records, and output records are deconstructed into behavioral semantics, and the task type, task object, task duration, collaborating members, and output characteristics corresponding to each record are extracted to generate the member's professional behavior trace. After generating the professional behavior traces of members, an initial set of professional anchor points is obtained. For each initial professional anchor point in the initial set of professional anchor points, the task type, task object, collaborating members and output features in the professional behavior trace are semantically encoded to generate a professional semantic feature vector of the professional behavior trace. The semantic distance between the professional semantic feature vector of the professional behavior trace and the professional semantic feature vector of the initial professional anchor point is calculated. The semantic distance is compared with a preset association threshold. When the semantic distance is less than or equal to the preset association threshold, it is determined that there is a professional association between the professional behavior trace and the initial professional anchor point. After determining that an occupational association exists, it is further determined whether the occupational behavior trace and the initial occupational anchor point meet any of the preset matching conditions, preset extension conditions, or preset deviation conditions. The preset matching condition means that the semantic distance between the occupational semantic feature vector of the occupational behavior trace and the occupational semantic feature vector of the initial occupational anchor point falls within a first preset threshold range. The preset extension condition means that the semantic distance falls within a second preset threshold range. The preset deviation condition means that the semantic distance falls within a third preset threshold range. When it is determined that any of the preset matching conditions, preset extension conditions, or preset deviation conditions are met, an occupational mapping event is determined to be captured, and the type of the occupational mapping event is recorded as a matching event, an extension event, or a deviation event. The intensity value of the occupational mapping event is recorded as the semantic distance value, and the unique identifier of the initial occupational anchor point associated with the occupational mapping event is recorded.

[0007] Preferably, in step S3, a career mapping event is read from the career mapping event queue, the unique identifier of the initial career anchor associated with the career mapping event is extracted, the binding relationship between the initial career anchor set and the career decision tree is queried based on the unique identifier of the initial career anchor, and the position of the initial career anchor associated with the career mapping event in the career decision tree is determined. The position includes the node path information of the leaf node bound to the initial career anchor in the career decision tree. After determining the position of the initial career anchor point in the career decision tree, a derived career anchor point corresponding to the career mapping event is generated downstream of the leaf node to which the initial career anchor point is bound. Career behavior traces corresponding to the career mapping event are extracted from the career mapping event. Member identifiers associated with the career behavior traces corresponding to the career mapping event are extracted from the career mapping event. The type of the career mapping event is obtained from the career mapping event and converted into an association type identifier. The extracted career behavior traces, member identifiers, and association type identifiers are associated and stored to form the derived career anchor point. The derived career anchor point is then bound to the newly generated child node downstream of the leaf node to which the initial career anchor point is bound. After generating derived career anchors and binding them to newly generated child nodes, a career evolution path is established in the career decision tree from the leaf node bound to the initial career anchor to the child node bound to the derived career anchor. The leaf node bound to the initial career anchor is determined as the starting node of the path, and the child node bound to the derived career anchor is determined as the ending node of the path. The type of career mapping event is obtained and converted into an evolution type identifier according to the preset evolution type mapping rules. The evolution type identifier is stored as an attribute of the directed edge. A new edge record is added to the node connection relationship record table of the career decision tree. The edge record contains the starting node identifier, the ending node identifier, and the evolution type identifier, thus completing the establishment of the career evolution path.

[0008] Preferably, in step S4, the occupational allocation intention input by the organization manager is obtained through the natural language interaction interface in the graphical user interface, and the occupational allocation intention is preprocessed with natural language to obtain a structured allocation intention representation. The structured allocation intention representation includes an intention category field, a key entity field, and an original input text field. The structured allocation intent representation is semantically matched with all anchors in the career decision tree. One or more anchors with the highest matching degree are identified as target career anchors. All anchors in the career decision tree include initial career anchors and derived career anchors. Each anchor is associated with a career role node in the career decision tree. The key entity fields in the structured allocation intent representation are matched with the job title and job description text of each career role node. The intent category field in the structured allocation intent representation is matched with the hierarchical category of the career role node in the career decision tree. The entity matching degree and hierarchical matching degree are weighted and fused to generate a comprehensive matching degree score between the allocation intent and each anchor. One or more anchors with the highest comprehensive matching degree score are selected as target career anchors. After identifying one or more target career anchors, for each target career anchor, an occupational evolution subgraph centered on that target career anchor is extracted from the career decision tree. The node position of the target career anchor in the career decision tree is determined. All nodes in the career decision tree that have career evolution path connections with that node are traversed. Starting from that node, forward and backward traversals are performed along the direction of the career evolution path. The current node itself, the upstream nodes found by forward traversal, the downstream nodes found by backward traversal, and all career evolution paths connecting these nodes are gathered together to form an occupational evolution subgraph centered on the target career anchor. The upstream nodes include the upstream initial career anchors directly associated with the current node, and the downstream nodes include the downstream derived career anchors directly associated with the current node. After extracting the career evolution subgraph, the career evolution subgraph is matched with the historical career behavior traces of each member. The member with the highest degree of matching with the target career anchor is identified as the target member. The historical career behavior traces of all members in the organization are obtained. For each member, the task types and output characteristics involved in the member's historical career behavior traces are semantically compared with the responsibility description text of the career role nodes associated with the target career anchor. The fit score between the member and the target career anchor is calculated. The member with the highest fit score is selected as the target member, and the matching basis between the target member and the target career anchor is recorded. The career evolution subgraph is visualized and rendered to generate career recommendations. Each anchor point in the career evolution subgraph is mapped to a node graphic in the graph. The initial career anchor point is mapped to the first shape, and the derived career anchor points are mapped to the second shape. Each career evolution path in the career evolution subgraph is mapped to a directed edge graphic connecting two node graphics, with the arrow direction of the directed edge pointing from the upstream node to the downstream node. The directed edge graphic is rendered differently according to the evolution type identifier carried by each career evolution path, with different evolution type identifiers corresponding to different rendering colors. The matching relationship between target members and target career anchor points is displayed in the career recommendation graph, along with the text description of the matching basis. The rendered graph is then output to the career recommendation display area of ​​the graphical user interface.

[0009] The present invention also provides a human resource management system, comprising: Data acquisition module: Acquires the initial set of occupational anchor points of the target organization. The initial set of occupational anchor points is obtained by parsing the job structure and job description of the target organization. Each anchor point corresponds to a preset occupational role and is associated with the job responsibility text segment and its occupational semantic features. Behavior monitoring module: Based on the organizational workflow system, it captures the traces of professional behavior generated by members in the process of performing work tasks. The traces of professional behavior refer to the operation records, collaboration records or output records left by members when handling affairs. Occupational Anchor Reconstruction Module: Based on the captured occupational behavior traces, perform dynamic mapping operations on occupational anchors. The dynamic mapping operations include: generating a derivative occupational anchor corresponding to the occupational behavior trace in the occupational anchor system according to the semantic correlation between the occupational behavior trace and the initial occupational anchor, and establishing an occupational evolution path between the derivative occupational anchor and the initial occupational anchor. Career recommendation generation module: Based on the career allocation intentions of organizational managers, it determines the target members and target career anchors that semantically match the career allocation intentions from the career anchor system, and generates career recommendations based on the career evolution paths associated with the target members.

[0010] This invention provides a human resource management method and system, which has the following beneficial effects: This invention constructs an initial set of occupational anchor points, binding organizational job structures and occupational roles with job description texts and occupational semantic features to form a structured occupational reference benchmark. This establishes a calculable and comparable anchoring framework for dynamic monitoring and functional evolution tracking. By integrating with a workflow system, it captures operation records, collaboration records, and output records generated by members during the execution of work tasks. These records are then subjected to behavioral semantic deconstruction to generate structured occupational behavior traces. This allows for the traceable recording of the types of tasks actually undertaken by employees, collaborative relationships, and output characteristics, providing real behavioral data support for functional matching. Attached Figure Description

[0011] Figure 1 This is a flowchart of the present invention; Figure 2 This is a system block diagram of the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] Please see Figure 1 This invention provides a human resource management method, comprising the following steps: S1. Obtain the initial set of occupational anchor points for the target organization. The initial set of occupational anchor points is obtained by parsing the job structure and job description of the target organization. Each anchor point corresponds to an occupational role and is associated with the job responsibility text segment and its occupational semantic features. Furthermore, in S1, the job structure data and job description text of the target organization are obtained. The job structure data includes departmental hierarchy, job titles, and reporting relationships. The job description text includes job descriptions, qualification requirements, and key performance indicators for each job. After obtaining the job structure data, a career decision tree is constructed based on the job structure data. The root node of the career decision tree represents the organization's overall goal, the middle nodes represent departmental jobs, and the leaf nodes represent job roles. After the career decision tree is constructed, for each leaf node in the career decision tree, extract the corresponding job description text. The job description text includes the tasks to be performed by the position, the collaborating objects, and the types of deliverables. After extracting the job description text corresponding to each leaf node, the job description text is subjected to occupational semantic feature extraction to generate the initial occupational anchor point corresponding to the leaf node. Occupational semantic feature extraction includes: natural language preprocessing of the job description text, which includes word segmentation, entity recognition and syntactic analysis. Based on the preprocessing results, the core occupational elements of the job description text are extracted. The core occupational elements include task type, collaboration object type and output type. Based on the extracted core occupational elements, the occupational semantic feature vector of the job description text is generated. The position information of the leaf node in the occupational decision tree, the job description text corresponding to the leaf node and the generated occupational semantic feature vector are associated and stored together to form the initial occupational anchor point. After traversing all leaf nodes of the career decision tree and generating corresponding initial career anchors, the initial career anchors corresponding to all leaf nodes are collected to form an initial career anchor set. Each initial career anchor in the initial career anchor set is then bound to the corresponding leaf node in the career decision tree through its stored location information.

[0014] It should be noted that the job structure data is obtained by accessing the organization's human resource management system or reading the organizational structure configuration file. This data includes departmental hierarchy, job titles, and reporting relationships. The departmental hierarchy records the hierarchical structure between departments within the organization, the job titles record the standard titles of each job, and the reporting relationships record the superior job identifiers for each job. The job description text is obtained by reading the organization's human resource document library. This text includes the job description, qualification requirements, and key performance indicators for each job. The job description records the list of work content that the job needs to undertake, the qualification requirements record the knowledge, skills, and abilities required for the job, and the key performance indicators record the quantitative or qualitative indicators for measuring the job's performance. The obtained job structure data and job description text are stored in the organization's basic information database. After obtaining the job structure data, a career decision tree is constructed based on the job structure data. The construction of the career decision tree specifically includes the following operations: taking the overall organizational goal as the root node, the overall organizational goal is a general description of the organization's overall purpose and core tasks, traversing the departmental hierarchy in the job structure data, taking each department as an intermediate node, and determining the parent-child connection relationship between intermediate nodes according to the subordinate hierarchy relationship between departments, with the superior department node as the parent node and the subordinate department node as the child node. For each department node, iterate through all positions under that department, treat each position as a leaf node, and establish a parent-child connection between the leaf node and its department node. For each leaf node, extract the corresponding reporting relationship from the position structure data and store the reporting relationship as an attribute of the leaf node. Through the above operations, a career decision tree is formed. After the career decision tree is constructed, the initial career anchor point construction module extracts the job description text corresponding to each leaf node in the career decision tree. Specifically, it obtains the job name corresponding to the leaf node. The job name is information extracted from the job structure data and stored in the leaf node attributes when constructing the career decision tree. Based on the job name, it searches and locates the job description text entries that match the job name. It then extracts the content of the job description field from the job description text entries. The job description field includes the task list that the job needs to perform, the collaboration objects that the job needs to connect with in cross-job collaboration, and the type of output that the job needs to deliver after completing the work. After extraction, the job description text is associated with the corresponding leaf node and stored as the job description text attribute of the leaf node. After extracting the job description text corresponding to each leaf node, occupational semantic features are extracted for each job description text to generate an initial occupational anchor point corresponding to the leaf node. The occupational semantic feature extraction specifically includes the following steps: Natural language preprocessing is performed on the job description text. Key entities such as task verbs, collaborating object nouns, and output nouns are identified from the text through entity recognition operations. Modification and collocation relationships between language units are determined through syntactic analysis operations. Based on the preprocessing results, the core occupational elements of the position are extracted from the job description text. The core occupational elements include task type, collaborating object type, and output type. Based on the extracted core occupational elements, an occupational semantic feature vector of the job description text is generated. The occupational semantic feature vector is a vector representation formed by mapping the core occupational elements to a preset occupational semantic space. It is used to represent the position coordinates of the job role in the occupational semantic space. The position information of the leaf node in the occupational decision tree, the job description text corresponding to the leaf node, and the generated occupational semantic feature vector are associated and stored to form an initial occupational anchor point. The position information includes the complete path node identifier sequence from the root node to the leaf node. After traversing all leaf nodes of the career decision tree and completing the above processing, the initial career anchor points corresponding to all leaf nodes are collected to form an initial career anchor point set. The initial career anchor point set maintains a structural correspondence with the career decision tree. Each initial career anchor point is uniquely bound to the corresponding leaf node in the career decision tree through its stored location information. Through the binding relationship, each job role node in the career decision tree has a traceable and comparable career semantic representation. This representation includes the job description text and its vector coordinates in the career semantic space. After the initial career anchor point set is constructed, it serves as a benchmark reference for subsequent monitoring of members' career behavior traces. It is used to semantically compare with the actual career behavior traces generated by members to achieve dynamic tracking of the matching, expansion, or deviation relationship between members' actual career roles and preset career roles.

[0015] S2. Based on the workflow system, capture the traces of professional behavior generated by members during the execution of work tasks. The traces of professional behavior refer to the operation records, collaboration records or output records left by members when handling affairs. Furthermore, in S2, the workflow system is accessed to obtain the operation records, collaboration records, and output records generated by the members in the workflow system. The obtained operation records, collaboration records, and output records are deconstructed into behavioral semantics to extract the task type, task object, task duration, collaborating members, and output characteristics corresponding to each record, thereby generating the member's professional behavior traces. After generating the professional behavior traces of members, an initial set of professional anchor points is obtained. For each initial professional anchor point in the initial set of professional anchor points, the task type, task object, collaborating members and output features in the professional behavior trace are semantically encoded to generate a professional semantic feature vector of the professional behavior trace. The semantic distance between the professional semantic feature vector of the professional behavior trace and the professional semantic feature vector of the initial professional anchor point is calculated. The semantic distance is compared with a preset association threshold. When the semantic distance is less than or equal to the preset association threshold, it is determined that there is a professional association between the professional behavior trace and the initial professional anchor point. After determining that an occupational association exists, it is further determined whether the occupational behavior trace and the initial occupational anchor point meet any of the preset matching conditions, preset extension conditions, or preset deviation conditions. The preset matching condition means that the semantic distance between the occupational semantic feature vector of the occupational behavior trace and the occupational semantic feature vector of the initial occupational anchor point falls within a first preset threshold range. The preset extension condition means that the semantic distance falls within a second preset threshold range. The preset deviation condition means that the semantic distance falls within a third preset threshold range. When it is determined that any of the preset matching conditions, preset extension conditions, or preset deviation conditions are met, an occupational mapping event is determined to be captured, and the type of the occupational mapping event is recorded as a matching event, an extension event, or a deviation event. The intensity value of the occupational mapping event is recorded as the semantic distance value, and the unique identifier of the initial occupational anchor point associated with the occupational mapping event is recorded.

[0016] It should be noted that a data interface is established with the workflow system, which includes, but is not limited to, project management systems, task collaboration platforms, document management systems, or enterprise resource planning systems. Through the data interface, the professional behavior monitoring module accesses the data flow of the workflow system through periodic polling or event listening, and obtains various records generated by members during the execution of work tasks in real time or near real time. Operation records include members' operations such as creation, editing, submission, approval, and closing in the system, as well as their corresponding timestamps and operation object identifiers. Collaboration records include members' interactive behaviors in the workflow system, such as comments, @mentions, task assignments, and group discussions, as well as their corresponding collaboration object identifiers and collaboration content summaries. Output records include documents, code, design schemes, reports, and other deliverables submitted by members in the workflow system, as well as their corresponding file attributes, storage paths, and submission times. After obtaining the operation records, collaboration records, and output records, they are stored in the behavior cache area. Each record is accompanied by the member identifier that generated the record. After acquiring operation records, collaboration records, and output records, behavioral semantic deconstruction is performed on these records. This deconstruction includes the following steps: For each record, type identification is performed. Based on the source system fields or record format characteristics, the record is categorized into operation record type, collaboration record type, or output record type. For operation record types, the task type and task object involved in the operation are extracted based on the operation object identifier and operation type field. The task type is determined by the category attribute of the operation object or the project tag associated with the operation. The task object is determined by the name or number field of the operation object. The task duration spent by the member on the task is calculated based on the consecutive operation timestamps under the same task object. For collaboration... For record types, the collaboration topic is extracted based on the collaboration content summary, and the identifiers of other collaborating members who have a collaboration relationship with the member are extracted based on the @mention field or comment reply relationship. For output record types, output features are extracted based on the metadata of the output file. Output features include output type, output size, and task identifier associated with the output. The extracted task type, task object, task duration, collaborating members, and output features are aggregated according to member identifiers to form the member's professional behavior trace within a preset time window. The professional behavior trace is stored in the form of structured data, including member identifier, task type field, task object field, task duration field, collaborating member list field, and output feature field. After generating the professional behavior traces of members, an initial set of professional anchor points is obtained. This set contains multiple initial professional anchor points, each associated with a job description text and its corresponding professional semantic feature vector. For each initial professional anchor point in the set, the task type, task object, collaborating members, and output features in the professional behavior trace are semantically encoded to generate a professional semantic feature vector for that professional behavior trace. This professional semantic feature vector and the professional semantic feature vector of the initial professional anchor point reside in the same preset professional semantic space. The semantic distance between the professional semantic feature vector of the professional behavior trace and the professional semantic feature vector of the initial professional anchor point is calculated. This semantic distance is obtained through a distance metric in the vector space and is used to characterize the semantic closeness between the actual task content performed by the member and the preset job description. The calculated semantic distance is compared with a preset association threshold. When the semantic distance is less than or equal to the preset association threshold, it is determined that there is a professional association between the professional behavior trace and the initial professional anchor point, and the corresponding semantic distance value is recorded. When the semantic distance is greater than the preset association threshold, it is determined that there is no professional association. After comparing all initial career anchors, for each comparison result with a career association, it is further determined whether the career behavior trace and the initial career anchor meet any of the preset matching conditions, extension conditions, or deviation conditions. The preset matching condition means that the semantic distance between the career semantic feature vector of the career behavior trace and the career semantic feature vector of the initial career anchor falls within a first preset threshold range, and the task type field is consistent with the core task type in the job description text, indicating that the task actually performed by the member is highly consistent with the preset responsibilities of the position. The preset extension condition means that the semantic distance between the career semantic feature vector of the career behavior trace and the career semantic feature vector of the initial career anchor falls within a second preset threshold range, and the task type or collaborating member field in the career behavior trace contains semantically relevant content not explicitly stated in the job description text, indicating that the task actually performed by the member is highly consistent with the preset responsibilities of the position. The task exceeds the preset responsibilities of the position but is related. The preset deviation condition means that the semantic distance between the professional semantic feature vector of the professional behavior trace and the professional semantic feature vector of the initial professional anchor falls into the third preset threshold range, and the task type or output feature in the professional behavior trace is semantically inconsistent with the core content in the responsibility description text, indicating that the task actually performed by the member has a directional difference from the preset responsibilities of the position. When any of the above conditions are met, it is determined that a professional mapping event is captured, and the type of the professional mapping event is recorded as a matching event, an extended event, or a deviation event. The intensity value of the professional mapping event is recorded as the calculated semantic distance value, and the unique identifier of the initial professional anchor associated with the professional mapping event is recorded. The capture and recording operation of the professional mapping event is completed, and the event is stored in the professional mapping event queue for subsequent dynamic reconstruction module to call.

[0017] S3. Based on the captured occupational behavior traces, perform a dynamic mapping operation of occupational anchors. The dynamic mapping operation includes: generating a derivative occupational anchor corresponding to the occupational behavior trace in the occupational anchor system according to the semantic correlation between the occupational behavior trace and the initial occupational anchor, and establishing an occupational evolution path between the derivative occupational anchor and the initial occupational anchor. Furthermore, in S3, career mapping events are read from the career mapping event queue, the unique identifier of the initial career anchor associated with the career mapping event is extracted, the binding relationship between the initial career anchor set and the career decision tree is queried based on the unique identifier of the initial career anchor, and the position of the initial career anchor associated with the career mapping event in the career decision tree is determined. The position includes the node path information of the leaf node bound to the initial career anchor in the career decision tree. After determining the position of the initial career anchor point in the career decision tree, a derived career anchor point corresponding to the career mapping event is generated downstream of the leaf node to which the initial career anchor point is bound. Career behavior traces corresponding to the career mapping event are extracted from the career mapping event. Member identifiers associated with the career behavior traces corresponding to the career mapping event are extracted from the career mapping event. The type of the career mapping event is obtained from the career mapping event and converted into an association type identifier. The extracted career behavior traces, member identifiers, and association type identifiers are associated and stored to form the derived career anchor point. The derived career anchor point is then bound to the newly generated child node downstream of the leaf node to which the initial career anchor point is bound. After generating derived career anchors and binding them to newly generated child nodes, a career evolution path is established in the career decision tree from the leaf node bound to the initial career anchor to the child node bound to the derived career anchor. The leaf node bound to the initial career anchor is determined as the starting node of the path, and the child node bound to the derived career anchor is determined as the ending node of the path. The type of career mapping event is obtained and converted into an evolution type identifier according to the preset evolution type mapping rules. The evolution type identifier is stored as an attribute of the directed edge. A new edge record is added to the node connection relationship record table of the career decision tree. The edge record contains the starting node identifier, the ending node identifier, and the evolution type identifier, thus completing the establishment of the career evolution path.

[0018] It should be noted that after capturing a career mapping event and storing it in the career mapping event queue, the career mapping event is read from the career mapping event queue, and the unique identifier of the initial career anchor associated with the event is extracted. The unique identifier is the encoded information assigned to each initial career anchor during the initial career anchor construction stage to uniquely identify the anchor. Based on the unique identifier, the binding relationship between the initial career anchor set and the career decision tree is queried. The binding relationship records the node path information of the leaf node bound to each initial career anchor in the career decision tree. The node path information includes the hierarchical sequence of all intermediate nodes traversed from the root node to the leaf node and the node identifier of each node. By parsing the node path information, the precise position of the initial career anchor associated with the career mapping event in the career decision tree is determined. The precise position includes the hierarchical number of the leaf node bound to the initial career anchor in the tree structure, the identifier of the parent node of the leaf node, the node identifier of the leaf node itself, and the sequential index of the leaf node among its sibling nodes. Once the location is determined, this location information is used as the insertion point for generating subsequent derivative occupation anchor points, and this location information is associated with and stored in relation to occupation mapping events.

[0019] After determining the position of the initial career anchor in the career decision tree, a derivative career anchor corresponding to the career mapping event is generated downstream of the leaf node to which the initial career anchor is bound. Downstream means that a child node is added below the leaf node as the parent node. The child node serves as the carrying node of the derivative career anchor and is located at the level directly below the leaf node. Extract the corresponding professional behavior traces from the captured professional mapping events. The professional behavior traces are structured data generated after deconstructing the operation records, collaboration records, and output records of members in the professional behavior monitoring phase. They include task type, task object, task duration, collaborating members, and output characteristics. Extract the member identifier associated with the professional behavior trace corresponding to the professional mapping event. The member identifier is an identity code that uniquely identifies the member in the workflow system. Obtain the type of the event from the professional mapping event. The type is one of matching event, extended event, or deviation event. Convert the type into an association type identifier. The association type identifier is used to characterize the professional relationship between the derived professional anchor point and its upstream initial professional anchor point. The association type identifier corresponding to the matching event indicates functional matching, the association type identifier corresponding to the extended event indicates functional expansion, and the association type identifier corresponding to the deviation event indicates functional deviation. The extracted occupational behavior traces, member identifiers, and association type identifiers are associated and stored together to form a derived occupational anchor point. The derived occupational anchor point is then bound to the newly generated child node. At the same time, the generation timestamp of the derived occupational anchor point and the associated occupational mapping event identifier are recorded in the child node attributes, thereby completing the insertion operation of the derived occupational anchor point in the occupational decision tree. After generating a derived career anchor point and binding it to a new child node in the career decision tree, a career evolution path is established in the career decision tree from the leaf node bound to the initial career anchor point to the child node bound to the derived career anchor point. The career evolution path is a directed edge, with its direction pointing from the initial career anchor point to the derived career anchor point, which is used to record the evolution process of the career relationship between the two. The leaf node bound to the initial career anchor point is determined as the starting node of the path, and the child node bound to the derived career anchor point is determined as the ending node of the path. The node identifiers of the starting node and the ending node are obtained respectively. The type of career mapping event is obtained, and the type is converted into an evolution type identifier according to a preset evolution type mapping rule. The preset evolution type mapping rule includes: when the career mapping event is a matching event, it is mapped to the career enhancement evolution type; when the career mapping event is an expansion event, it is mapped to the career expansion evolution type; when the career mapping event is a deviation event, it is mapped to the career offset evolution type. The evolution type identifier is stored as an attribute of the directed edge. The attribute also includes the establishment timestamp of the career evolution path and the career mapping event identifier on which it is based. In the node connection relationship record table of the career decision tree, a new edge record is added. The edge record includes the starting node identifier, the ending node identifier, the evolution type identifier, the establishment timestamp, and the career mapping event identifier.

[0020] S4. Based on the career allocation intentions of organizational managers, determine the target members and target career anchors that semantically match the career allocation intentions from the career anchor system, and generate career recommendations based on the career evolution paths associated with the target members.

[0021] Furthermore, in S4, the occupational allocation intention input by the organization manager is obtained through the natural language interaction interface in the graphical user interface. The occupational allocation intention is preprocessed with natural language to obtain a structured allocation intention representation. The structured allocation intention representation includes an intention category field, a key entity field, and an original input text field. The structured allocation intent representation is semantically matched with all anchors in the career decision tree. One or more anchors with the highest matching degree are identified as target career anchors. All anchors in the career decision tree include initial career anchors and derived career anchors. Each anchor is associated with a career role node in the career decision tree. The key entity fields in the structured allocation intent representation are matched with the job title and job description text of each career role node. The intent category field in the structured allocation intent representation is matched with the hierarchical category of the career role node in the career decision tree. The entity matching degree and hierarchical matching degree are weighted and fused to generate a comprehensive matching degree score between the allocation intent and each anchor. One or more anchors with the highest comprehensive matching degree score are selected as target career anchors. After identifying one or more target career anchors, for each target career anchor, an occupational evolution subgraph centered on that target career anchor is extracted from the career decision tree. The node position of the target career anchor in the career decision tree is determined. All nodes in the career decision tree that have career evolution path connections with that node are traversed. Starting from that node, forward and backward traversals are performed along the direction of the career evolution path. The current node itself, the upstream nodes found by forward traversal, the downstream nodes found by backward traversal, and all career evolution paths connecting these nodes are gathered together to form an occupational evolution subgraph centered on the target career anchor. The upstream nodes include the upstream initial career anchors directly associated with the current node, and the downstream nodes include the downstream derived career anchors directly associated with the current node. After extracting the career evolution subgraph, the career evolution subgraph is matched with the historical career behavior traces of each member. The member with the highest degree of matching with the target career anchor is identified as the target member. The historical career behavior traces of all members in the organization are obtained. For each member, the task types and output characteristics involved in the member's historical career behavior traces are semantically compared with the responsibility description text of the career role nodes associated with the target career anchor. The fit score between the member and the target career anchor is calculated. The member with the highest fit score is selected as the target member, and the matching basis between the target member and the target career anchor is recorded. The career evolution subgraph is visualized and rendered to generate career recommendations. Each anchor point in the career evolution subgraph is mapped to a node graphic in the graph. The initial career anchor point is mapped to the first shape, and the derived career anchor points are mapped to the second shape. Each career evolution path in the career evolution subgraph is mapped to a directed edge graphic connecting two node graphics, with the arrow direction of the directed edge pointing from the upstream node to the downstream node. The directed edge graphic is rendered differently according to the evolution type identifier carried by each career evolution path, with different evolution type identifiers corresponding to different rendering colors. The matching relationship between target members and target career anchor points is displayed in the career recommendation graph, along with the text description of the matching basis. The rendered graph is then output to the career recommendation display area of ​​the graphical user interface.

[0022] It should be noted that after obtaining the structured allocation intent representation, semantic matching is performed on all anchor points in the career decision tree. All anchor points include initial career anchor points and derived career anchor points. Each anchor point is associated with a career role node in the career decision tree. This career role node contains the corresponding job title, job description text, and the node's hierarchical category in the tree. The key entity fields in the structured allocation intent representation are then compared with the job title and job description text of each career role node to calculate entity matching degree. This entity matching degree is determined by comparing whether the job title and skill keywords identified in the allocation intent are consistent with or contain similar entities in the job title and job description text of the career role node. The meaning association is used to determine the hierarchical matching degree between the intent category field in the structured allocation intent representation and the hierarchical category to which the occupational role node belongs in the occupational decision tree. The hierarchical matching degree is determined based on the preset mapping relationship between the intent category and the node hierarchical category. For example, the job filling intent has the highest matching degree with the leaf node level, and the structural reorganization intent has the highest matching degree with the intermediate node level. The entity matching degree and the hierarchical matching degree are weighted and fused to generate a comprehensive matching degree score between the allocation intent and each anchor point. The comprehensive matching degree scores of all anchor points are sorted in descending order, and one or more anchor points with the highest comprehensive matching degree scores are selected as target occupational anchor points. The target occupational anchor points are used for the subsequent extraction of occupational evolution subgraphs. The extraction of the career evolution subgraph specifically includes the following steps: First, determine the node position of the target career anchor point in the career decision tree. The node position includes the node identifier of the node bound to the anchor point in the tree structure and the node's hierarchy information. Second, traverse all nodes in the career decision tree that have career evolution path connections with the target node. The career evolution path is a directed edge previously established by the career anchor point dynamic reconstruction module. Each directed edge records the starting node identifier, ending node identifier, and evolution type identifier. Third, starting from the target node, perform forward and backward traversals along the direction of the career evolution path. Forward traversal refers to traversing along the path from the upstream node... The current node searches for upstream nodes directly associated with it in the direction of forward traversal. Backward traversal means searching for downstream nodes directly associated with it in the direction from the current node to the downstream node. The current node itself, the upstream nodes found by forward traversal, the downstream nodes found by backward traversal, and all career evolution paths connecting these nodes are gathered together to form a career evolution subgraph centered on the target career anchor point. The career evolution subgraph includes the target career anchor point itself, the upstream initial career anchor point or downstream derived career anchor point directly associated with the target career anchor point, and the career evolution paths connecting these anchor points. After identifying one or more target career anchors, for each target career anchor, an occupational evolution subgraph centered on that target career anchor is extracted from the career decision tree. The node position of the target career anchor in the career decision tree is determined. All nodes in the career decision tree that have career evolution path connections with that node are traversed. Starting from that node, forward and backward traversals are performed along the direction of the career evolution path. The current node itself, the upstream nodes found through forward traversal, the downstream nodes found through backward traversal, and all career evolution paths connecting these nodes are gathered to form an occupational evolution subgraph centered on the target career anchor. The upstream nodes include the upstream initial career anchors directly associated with the current node, and the downstream nodes include the downstream derived career anchors directly associated with the current node. After extracting the career evolution subgraph, the subgraph is matched with the historical career behavior traces of each member. The member with the highest degree of matching with the target career anchor is identified as the target member. The historical career behavior traces of all members in the organization are obtained. For each member, the task types and output characteristics involved in the member's historical career behavior traces are semantically compared with the responsibility description text of the career role nodes associated with the target career anchor. The fit score between the member and the target career anchor is calculated. The member with the highest fit score is selected as the target member, and the matching basis between the target member and the target career anchor is recorded. The career evolution subgraph is visualized and rendered to generate a career recommendation graph. Each anchor point in the career evolution subgraph is mapped to a node graphic in the graph, with the initial career anchor point mapped to the first shape and the derived career anchor points mapped to the second shape. Each career evolution path in the career evolution subgraph is mapped to a directed edge graphic connecting two node graphics, with the arrow direction of the directed edge pointing from the upstream node to the downstream node. The directed edge graphic is rendered differently according to the evolution type identifier carried by each career evolution path, with different evolution type identifiers corresponding to different rendering colors. The matching relationship between the target member and the target career anchor point is displayed in the career recommendation graph, along with a text description of the matching basis. The rendered graph is then output to the career recommendation graph display area of ​​the graphical user interface.

[0023] It should be noted that the preset threshold in this invention is obtained by those skilled in the art by collecting multiple sets of sample data and setting a corresponding preset ratio coefficient for each set of sample data; substituting the set preset ratio coefficient and the collected sample data into the formula, any two formulas constitute a system of two first equations, and the calculated coefficients are filtered and averaged to obtain the value. The size of the coefficient is a specific value obtained by quantifying each parameter to facilitate subsequent comparison. The size of the coefficient depends on the amount of sample data and the preset proportional coefficient initially set by those skilled in the art for each set of sample data; as long as it does not affect the proportional relationship between the parameter and the quantified value.

[0024] Please see Figure 2 This invention provides a human resource management system, comprising: Data acquisition module: Acquires the initial set of occupational anchor points of the target organization. The initial set of occupational anchor points is obtained by parsing the job structure and job description of the target organization. Each anchor point corresponds to a preset occupational role and is associated with the job responsibility text segment and its occupational semantic features. Behavior monitoring module: Based on the organizational workflow system, it captures the traces of professional behavior generated by members in the process of performing work tasks. The traces of professional behavior refer to the operation records, collaboration records or output records left by members when handling affairs. Occupational Anchor Reconstruction Module: Based on the captured occupational behavior traces, perform dynamic mapping operations on occupational anchors. The dynamic mapping operations include: generating a derivative occupational anchor corresponding to the occupational behavior trace in the occupational anchor system according to the semantic correlation between the occupational behavior trace and the initial occupational anchor, and establishing an occupational evolution path between the derivative occupational anchor and the initial occupational anchor. Career recommendation generation module: Based on the career allocation intentions of organizational managers, it determines the target members and target career anchors that semantically match the career allocation intentions from the career anchor system, and generates career recommendations based on the career evolution paths associated with the target members.

[0025] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0026] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0027] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0028] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0029] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A human resource management method, characterized in that, include: S1. Obtain the initial set of occupational anchor points for the target organization. The initial set of occupational anchor points is obtained by parsing the job structure and job description of the target organization. Each anchor point corresponds to an occupational role and is associated with the job responsibility text segment and its occupational semantic features. S2. Based on the workflow system, capture the traces of professional behavior generated by members during the execution of work tasks. The traces of professional behavior refer to the operation records, collaboration records or output records left by members when handling affairs. S3. Based on the captured occupational behavior traces, perform a dynamic mapping operation of occupational anchors. The dynamic mapping operation includes: generating a derivative occupational anchor corresponding to the occupational behavior trace in the occupational anchor system according to the semantic correlation between the occupational behavior trace and the initial occupational anchor, and establishing an occupational evolution path between the derivative occupational anchor and the initial occupational anchor. S4. Based on the career allocation intentions of organizational managers, determine the target members and target career anchors that semantically match the career allocation intentions from the career anchor system, and generate career recommendations based on the career evolution paths associated with the target members.

2. The human resource management method according to claim 1, characterized in that, In S1, the job structure data and job description text of the target organization are obtained. The job structure data includes departmental hierarchy, job titles, and reporting relationships. The job description text includes job descriptions, qualification requirements, and key performance indicators for each job. After obtaining the job structure data, a career decision tree is constructed based on the job structure data. The root node of the career decision tree represents the organization's overall goal, the middle nodes represent departmental jobs, and the leaf nodes represent job roles. After the career decision tree is constructed, for each leaf node in the career decision tree, extract the corresponding job description text. The job description text includes the tasks to be performed by the position, the collaborating objects, and the types of deliverables. After extracting the job description text corresponding to each leaf node, occupational semantic features are extracted for each job description text to generate the initial occupational anchor point corresponding to the leaf node. Occupational semantic feature extraction includes natural language preprocessing of the job description text, which includes word segmentation, entity recognition and syntactic analysis.

3. The human resource management method according to claim 2, characterized in that, Based on the preprocessing results, the core occupational elements of the job description text are extracted. The core occupational elements include task type, collaboration object type and output type. Based on the extracted core occupational elements, the occupational semantic feature vector of the job description text is generated. The position information of the leaf node in the occupational decision tree, the job description text corresponding to the leaf node and the generated occupational semantic feature vector are associated and stored together to form the initial occupational anchor point. After traversing all leaf nodes of the career decision tree and generating corresponding initial career anchors, the initial career anchors corresponding to all leaf nodes are collected to form an initial career anchor set. Each initial career anchor in the initial career anchor set is then bound to the corresponding leaf node in the career decision tree through its stored location information.

4. The human resource management method according to claim 1, characterized in that, In S2, the workflow system is accessed to obtain the operation records, collaboration records and output records generated by members in the workflow system. The obtained operation records, collaboration records and output records are deconstructed into behavioral semantics to extract the task type, task object, task duration, collaborating members and output characteristics corresponding to each record, and generate the professional behavior trace of the member. After generating the professional behavior traces of members, an initial set of professional anchor points is obtained. For each initial professional anchor point in the initial set of professional anchor points, the task type, task object, collaborating members, and output features in the professional behavior trace are semantically encoded to generate a professional semantic feature vector of the professional behavior trace. The semantic distance between the professional semantic feature vector of the professional behavior trace and the professional semantic feature vector of the initial professional anchor point is calculated. The semantic distance is compared with a preset association threshold. When the semantic distance is less than or equal to the preset association threshold, it is determined that there is a professional association between the professional behavior trace and the initial professional anchor point.

5. A human resource management method according to claim 4, characterized in that, After determining that an occupational association exists, it is further determined whether the occupational behavior trace and the initial occupational anchor point meet any of the preset matching conditions, preset extension conditions, or preset deviation conditions. The preset matching condition means that the semantic distance between the occupational semantic feature vector of the occupational behavior trace and the occupational semantic feature vector of the initial occupational anchor point falls within a first preset threshold range. The preset extension condition means that the semantic distance falls within a second preset threshold range. The preset deviation condition means that the semantic distance falls within a third preset threshold range. When it is determined that any of the preset matching conditions, preset extension conditions, or preset deviation conditions are met, an occupational mapping event is determined to be captured, and the type of the occupational mapping event is recorded as a matching event, an extension event, or a deviation event. The intensity value of the occupational mapping event is recorded as the semantic distance value, and the unique identifier of the initial occupational anchor point associated with the occupational mapping event is recorded.

6. A human resource management method according to claim 1, characterized in that, In step S3, career mapping events are read from the career mapping event queue, the unique identifier of the initial career anchor associated with the career mapping event is extracted, the binding relationship between the initial career anchor set and the career decision tree is queried based on the unique identifier of the initial career anchor, and the position of the initial career anchor associated with the career mapping event in the career decision tree is determined. The position includes the node path information of the leaf node bound to the initial career anchor in the career decision tree. After determining the position of the initial career anchor point in the career decision tree, a derived career anchor point corresponding to the career mapping event is generated downstream of the leaf node to which the initial career anchor point is bound. Career behavior traces corresponding to the career mapping event are extracted from the career mapping event. Member identifiers associated with the career behavior traces corresponding to the career mapping event are extracted from the career mapping event. The type of the career mapping event is obtained from the career mapping event and converted into an association type identifier. The extracted career behavior traces, member identifiers, and association type identifiers are associated and stored to form the derived career anchor point. The derived career anchor point is then bound to the newly generated child node downstream of the leaf node to which the initial career anchor point is bound.

7. A human resource management method according to claim 6, characterized in that, After generating derived career anchors and binding them to newly generated child nodes, a career evolution path is established in the career decision tree from the leaf node bound to the initial career anchor to the child node bound to the derived career anchor. The leaf node bound to the initial career anchor is determined as the starting node of the path, and the child node bound to the derived career anchor is determined as the ending node of the path. The type of career mapping event is obtained and converted into an evolution type identifier according to the preset evolution type mapping rules. The evolution type identifier is stored as an attribute of the directed edge. A new edge record is added to the node connection relationship record table of the career decision tree. The edge record contains the starting node identifier, the ending node identifier, and the evolution type identifier, thus completing the establishment of the career evolution path.

8. A human resource management method according to claim 1, characterized in that, In step S4, the occupational allocation intention input by the organization manager is obtained through the natural language interaction interface in the graphical user interface. The occupational allocation intention is preprocessed with natural language to obtain a structured allocation intention representation. The structured allocation intention representation includes an intention category field, a key entity field, and an original input text field. The structured allocation intent representation is semantically matched with all anchors in the career decision tree. One or more anchors with the highest matching degree are identified as target career anchors. All anchors in the career decision tree include initial career anchors and derived career anchors. Each anchor is associated with a career role node in the career decision tree. The key entity fields in the structured allocation intent representation are matched with the job title and job description text of each career role node. The intent category field in the structured allocation intent representation is matched with the hierarchical category of the career role node in the career decision tree. The entity matching degree and hierarchical matching degree are weighted and fused to generate a comprehensive matching degree score between the allocation intent and each anchor. One or more anchors with the highest comprehensive matching degree score are selected as target career anchors.

9. A human resource management method according to claim 8, characterized in that, After identifying one or more target career anchors, for each target career anchor, an occupational evolution subgraph centered on that target career anchor is extracted from the career decision tree. The node position of the target career anchor in the career decision tree is determined. All nodes in the career decision tree that have career evolution path connections with that node are traversed. Starting from that node, forward and backward traversals are performed along the direction of the career evolution path. The current node itself, the upstream nodes found by forward traversal, the downstream nodes found by backward traversal, and all career evolution paths connecting these nodes are gathered together to form an occupational evolution subgraph centered on the target career anchor. The upstream nodes include the upstream initial career anchors directly associated with the current node, and the downstream nodes include the downstream derived career anchors directly associated with the current node. After extracting the career evolution subgraph, the career evolution subgraph is matched with the historical career behavior traces of each member. The member with the highest degree of matching with the target career anchor is identified as the target member. The historical career behavior traces of all members in the organization are obtained. For each member, the task types and output characteristics involved in the member's historical career behavior traces are semantically compared with the responsibility description text of the career role nodes associated with the target career anchor. The fit score between the member and the target career anchor is calculated. The member with the highest fit score is selected as the target member, and the matching basis between the target member and the target career anchor is recorded. The career evolution subgraph is visualized and rendered to generate career recommendations. Each anchor point in the career evolution subgraph is mapped to a node graphic in the graph. The initial career anchor point is mapped to the first shape, and the derived career anchor points are mapped to the second shape. Each career evolution path in the career evolution subgraph is mapped to a directed edge graphic connecting two node graphics, with the arrow direction of the directed edge pointing from the upstream node to the downstream node. The directed edge graphic is rendered differently according to the evolution type identifier carried by each career evolution path, with different evolution type identifiers corresponding to different rendering colors. The matching relationship between target members and target career anchor points is displayed in the career recommendation graph, along with the text description of the matching basis. The rendered graph is then output to the career recommendation display area of ​​the graphical user interface.

10. A human resource management system for executing a human resource management method as described in any one of claims 1-9, characterized in that, include: Data acquisition module: Acquires the initial set of occupational anchor points of the target organization. The initial set of occupational anchor points is obtained by parsing the job structure and job description of the target organization. Each anchor point corresponds to a preset occupational role and is associated with the job responsibility text segment and its occupational semantic features. Behavior monitoring module: Based on the organizational workflow system, it captures the traces of professional behavior generated by members in the process of performing work tasks. The traces of professional behavior refer to the operation records, collaboration records or output records left by members when handling affairs. Occupational Anchor Reconstruction Module: Based on the captured occupational behavior traces, perform dynamic mapping operations on occupational anchors. The dynamic mapping operations include: generating a derivative occupational anchor corresponding to the occupational behavior trace in the occupational anchor system according to the semantic correlation between the occupational behavior trace and the initial occupational anchor, and establishing an occupational evolution path between the derivative occupational anchor and the initial occupational anchor. Career recommendation generation module: Based on the career allocation intentions of organizational managers, it determines the target members and target career anchors that semantically match the career allocation intentions from the career anchor system, and generates career recommendations based on the career evolution paths associated with the target members.

Images