Human resource file storage management system based on knowledge graph

Abstract

The invention relates to the technical field of human resource management, in particular to a human resource archive storage management system based on a knowledge graph, which comprises a post identification module, a path modeling module, a node linkage module, a fracture identification module and an access control module. According to the invention, through the normalized mapping of post description and skill combination, the standardized processing capability of unstructured archive content is improved, and in combination with the jump frequency and span characteristics of multi-stage path nodes in task records, a stage label system fitting a post responsibility process is constructed; identifying cross correlation of responsibility fields among nodes by utilizing a job cycle division result, dynamically adjusting a path structure mapping sequence, accurately identifying a change condition of a responsibility conduction relation in a structure path, and judging a responsibility chain interruption range through field group difference in continuous empty path identification; and based on sensitive field distribution characteristics in the access path and a role field coverage ratio, access permission judgment and permission check control of a path level are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of human resource management, and in particular to a human resource file storage management system based on a knowledge graph. BACKGROUND

[0002] The technical field of human resource management includes systematic planning, organization and coordination of internal personnel-related transactions, and automatic processing means, focusing on the whole process of information management of human resources recruitment, employment, training, assessment, deployment, incentive, file management and other aspects through information means, covering human resource information collection, human relationship management, post responsibility allocation, employee history record and permission control, and relying on standardized data structure, rule-driven modeling method, dynamic association of human and post matching relationship and other ways to build a system, with the continuous development of management needs, integrating semantic recognition, relationship construction and structured management mechanisms to improve information integration and management efficiency. Among them, the human resource file storage management system based on the knowledge graph refers to the conversion of unstructured information in the human resource file into structured graph data through semantic modeling and graph structure mapping, and the centralized and unified storage and management, including the identification of key entities in personnel files, attribute classification, semantic understanding of file content, graph construction of employee information and organizational relationship, establishment and query processing of entity relationship path, specifically including identifying entity and attribute content in file text using rule-driven natural language parsing method, converting personnel information into graph nodes and relationship edges using pre-set triple generation logic, and based on graph database, unified storage and retrieval management of information, completing the structured organization and graph expression of human resource file information.

[0003] The traditional human resource file storage management technology takes standardized modeling and graph mapping as the core construction method, and lacks the ability to finely describe the changes of node responsibilities in the case of multi-stage task recording and dynamic post rotation. In the process of post path evolution, it cannot effectively identify the structural migration trend caused by the cross coverage of responsibility fields, leading to the inability to timely alarm after the responsibility is split or the process is interrupted. In path access management, it only relies on static role field configuration, lacks a linkage judgment mechanism for path structure and sensitive field distribution, and in the scene involving high-frequency function change or permission penetration access, it is easy to form sensitive information leakage or permission configuration blind area, affecting the accuracy and security of file permission control. SUMMARY

[0004] The purpose of the present application is to solve the shortcomings in the prior art, and to provide a human resource file storage management system based on a knowledge graph.

[0005] In order to achieve the above purpose, the present application adopts the following technical scheme: the human resource file storage management system based on the knowledge graph comprises: The post identification module calls the employee archive information, analyzes the post description word group and the skill keyword combination content, compares the post word group and the field order and word difference in the standard template, carries out expression mapping between the post terminology and the standard template, and generates the post terminology normalization result; The path modeling module calculates the jump frequency and node span difference of the path node in the employee task record in multiple stages according to the post terminology normalization result, analyzes the path connection and the responsibility process fitting situation, screens the node reconstruction fragment, establishes the stage label, and generates the tenure cycle division result; The node linkage module uses the tenure cycle division result to judge the responsibility field intersection content between post nodes, screens the cross-path connection field group, adjusts the mapping order, calculates the upstream and downstream responsibility intersection situation, constructs the structure change description label, and generates the path evolution feature information; The fracture identification module combines the path evolution feature information, constructs the post connection path, analyzes the length of the continuous vacancy node group and the path proportion, judges the responsibility field difference between the upstream and downstream nodes, and delimits the responsibility interruption range to generate the structure risk identification result; The access control module analyzes the distribution of sensitive fields in the access path according to the structure risk identification result, calls the role field group, compares the field matching coverage ratio, judges the path access permission state, and generates the archive access management result.

[0006] As a further scheme of the application, the post terminology normalization result includes a standard post field, a skill attribution label and an expression mapping structure, the tenure cycle division result includes a stage behavior identifier, a task span range and a path fragment number, the path evolution feature information includes a responsibility migration label, a node connection order and an evolution path structure, the structure risk identification result includes a vacancy connection segment, a responsibility breakpoint position and a link interruption range, and the archive access management result includes a permission path label, a sensitive field density and an access judgment level.

[0007] As a further scheme of the application, the post identification module includes: The post word group extraction submodule obtains the employee archive information, extracts the post description word group and the skill keyword, analyzes the word group and the keyword combination, establishes the post skill dataset, and generates the post word group data volume; The structure format comparison submodule compares the post word group structure and the field order and word difference in the standard word group template based on the post word group data volume, judges the appearance order of the keyword in the template, screens the fields matched with the template structure, and generates the post structure difference volume; The terminology mapping adjustment submodule calls the post structure difference volume, adjusts the original post word group description according to the combination matching result between the post and the skill, maps it to the standard template expression, and obtains the post terminology normalization result.

[0008] As a further scheme of the present application, the path modeling module comprises: The node frequency analysis submodule obtains the post term normalization result, counts the appearance frequency of each stage path node and the number of stage nodes in the employee task record, analyzes the order distribution of the nodes in each stage, establishes the node appearance change interval, and generates the stage jump change amount; The task path evaluation submodule calls the stage jump change amount, compares the connection order between the task path nodes and the key nodes in the post standard task process, filters the node distribution continuity and missing paragraphs, judges the co-occurrence proportion of actual task nodes and standard path nodes, and obtains the key path fitting degree coefficient; The behavior segmentation construction submodule monitors the distribution position of the node reconstruction behavior in the time sequence according to the key path fitting degree coefficient, collects the cumulative frequency of each behavior type in the period of time, the segmented node number, the node time interval and the span, calculates the node behavior change index, groups the behavior types and divides the period, forms the structured tenure period segmentation identifier, and establishes the tenure period division result.

[0009] As a further scheme of the present application, the node linkage module comprises: The responsibility intersection extraction submodule calls the responsibility field set in the original post node and the transferred post node according to the tenure period division result, calculates the overlap proportion between the matching content item number in the text expression and the field semantic range expression vector, constructs the intersection expression vector according to the coverage rate and content approximation score of the responsibility item, obtains the overlapped expression set and judges the information structure consistency, and obtains the post responsibility intersection ratio; The cross-post mapping submodule filters the field group with path continuity mark based on the post responsibility intersection ratio, detects the original position label of the target field group in the transferred post path structure, adjusts the mapping order and judges the coincidence distribution degree of the remapped field position, extracts the overlapping structure segment and counts the connected path number range, and obtains the cross-node field connection density; The path change recognition submodule calculates the difference item number between the upstream and downstream node binding responsibility fields in the structure path, the connection order change degree, the expression span difference, calculates the responsibility expression order offset density in the post path, constructs the structure change description label between the reconstructed path structure and the original path structure, and generates the path evolution feature information.

[0010] As a further scheme of the present application, the breakage recognition module comprises: The empty post section identification submodule obtains the path evolution feature information, detects a node group with a continuous unconfigured personnel binding field in the post connection path, calculates the length of the node group, and compares the length with the total length of the path to obtain an empty post continuous ratio; The responsibility field comparison submodule acquires a responsibility field set of upstream and downstream connection nodes on the empty post section based on the empty post continuous ratio, analyzes the number of items in the field set, judges the difference between the number and a responsibility configuration benchmark value, and obtains a responsibility configuration offset degree; The fracture interval determination submodule filters a position where a responsibility field coverage is broken, marks a fracture section range in a node sequence, and establishes a structure risk identification result according to the responsibility configuration offset degree and in combination with node distribution in the post connection path and the responsibility field coverage.

[0011] As a further scheme of the application, the access control module comprises: The field cumulative statistics submodule obtains the structure risk identification result, analyzes a salary field, an evaluation field and an organizational function field in the access path, counts the cumulative occurrence number of each type of field in the path, and obtains a total amount of path sensitive fields; The role permission comparison submodule calls a field group corresponding to an access level of a user role based on the total amount of path sensitive fields, compares a coverage ratio of the number of sensitive fields and the number of accessible fields in the role field set, and obtains a permission coverage ratio; The path permission determination submodule judges a matching relationship between the sensitive fields and the role field set in the path nodes according to the permission coverage ratio, marks an access permission state of the path, and establishes an archive access management result.

[0012] Compared with the prior art, the application has the advantages and positive effects that: In the application, the standardization processing capability of unstructured archive content is improved through the normalized mapping of post descriptions and skill combinations, a stage label system that fits post responsibility processes is constructed in combination with the jump frequency and span features of multi-stage path nodes in task records, the cross correlation of responsibility fields between nodes is identified by using the tenure cycle division result, the mapping order of path structures is dynamically adjusted, the change of responsibility transmission relationships in the structure path is accurately identified, the responsibility chain interruption range is judged in the continuous empty post path identification through the difference between field groups, and the access permission judgment and permission checking control at the path level are realized based on the sensitive field distribution features in the access path and the role field coverage ratio. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 The system flowchart of the application; Figure 2 The post identification module flowchart of the application; Figure 3A path modeling module flowchart of the present application; Figure 4 A node linkage module flowchart of the present application; Figure 5 A fracture identification module flowchart of the present application; Figure 6 An access control module flowchart of the present application. DETAILED DESCRIPTION

[0014] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0015] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and simplify the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0016] Please refer to Figure 1 The human resource file storage management system based on a knowledge graph comprises: The post identification module calls the employee file information, analyzes the post description word group and the skill keyword combination content, compares the post word group and the field order and word difference in the standard template, performs expression mapping between the post terminology and the standard template, and generates a post terminology normalization result; The path modeling module calculates the jump frequency and node span difference of the path nodes in the employee task record in multiple stages according to the post terminology normalization result, analyzes the path connection and the responsibility process fitting situation, screens the node reconstruction fragments, establishes the stage label, and generates a tenure division result; The node linkage module uses the tenure division result to judge the responsibility field intersection content between the post nodes, screens the cross-path connection field group, adjusts the mapping order, calculates the upstream and downstream responsibility intersection, constructs the structure change description label, and generates path evolution feature information; The fracture identification module combines the path evolution feature information, constructs the post connection path, analyzes the length of the continuous vacancy node group and the path proportion, judges the responsibility field difference between the upstream and downstream nodes, delimits the responsibility interruption range, and generates a structure risk identification result; The access control module analyzes the distribution of sensitive fields in the access path according to the structural risk identification result, calls the role field group comparison field matching coverage ratio, judges the path access permission state, and generates the archive access management result.

[0017] The post term normalization result includes a standard post field, a skill attribution label, and an expression mapping structure. The tenure division result includes a stage behavior identifier, a task span range, and a path segment number. The path evolution feature information includes a responsibility migration label, a node connection order, and an evolution path structure. The structural risk identification result includes an empty post connection segment, a responsibility breakpoint position, and a link interruption range. The archive access management result includes a permission path label, a sensitive field density, and an access judgment level.

[0018] Please refer to Figure 2 The post recognition module includes: The post phrase extraction submodule obtains employee archive information, extracts post description phrases and skill keywords, analyzes the combination of phrases and keywords, establishes a post skill dataset, and generates a post phrase data volume. Obtaining employee archive information refers to reading structured and unstructured fields containing the current post name, post responsibility description, work records over the years, key performance indicators, and other content from the personnel information system. The operation of extracting post description phrases is performed through two steps of regular expression and keyword matching. First, take the "post name" and "post responsibility description" fields as target fields, and extract verb phrases and noun phrases with an appearance frequency greater than 5 from them to combine as post description phrase candidates. For example, extract "sales plan" and "customer resource maintenance" from "responsible for developing sales plans and maintaining customer resources" as two description phrases, and store them as description phrases. In the step of extracting skill keywords, take the combination of nouns and adjectives appearing in the "required skills", "post requirements", and "assessment standards" fields as the target selection object. On this basis, an initial set of skill keywords is established. Assuming that "data analysis ability" and "communication and coordination ability" are mentioned in the post description, they are divided into "data analysis" and "communication and coordination", and their word frequencies are recorded as 9 and 5, respectively. Then, take the post ID as the index field to construct a post-skill combination structure, and generate a post skill dataset containing the correspondence between post description phrases and skill keywords. In this dataset, each post record is mapped to a specific skill set and description phrase combination to form a key-value structure mapping. The total number of extracted phrases in all post descriptions is 1865, as shown in the following table: Table 1: Post phrase and keyword extraction example table ; As shown in Table 1, the extracted description groups and key word frequencies in different positions are different, which are used as the basis for frequency threshold judgment and reordering operation in subsequent structure comparison and term mapping, and finally the number of position group data is 1865.

[0019] The structure format comparison submodule compares the field order and word difference between the position group structure and the standard group template based on the position group data, judges the appearance order of the key words in the template, filters the fields matching the template structure, and generates the position structure difference amount; Based on the position group data, the field order and word content difference between each position group structure and the standard group template are compared. In the field order comparison process, the three fields "verb + noun + qualifier" in the standard template are selected as the control order, the actual position group is field split and reordered, and the deviation degree from the standard order is judged. If a position description is "resource maintenance customer", there is an order difference with the standard "maintenance + customer + resource", which is recorded as 1 deviation. The frequency of position difference deviation is counted. If the deviation position is the first position, it is counted as a class deviation. If the deviation is at the end position, it is classified as another class of deviation. In the field content difference comparison, the Jaccard similarity between the key words in the group and the standard template word library is calculated one by one. If the template group is "analyze sales data" and the position group is "statistical sales information", the intersection of the key words is "sales", and the union is "analysis, sales, data, statistics, information", and the similarity is This similarity is lower than the set threshold 0.6, which is counted as a mismatch item. The appearance order of the key words in the template is mapped and the sequence difference is generated. Taking "formulate plan to maintain customers" as the original group and "plan to formulate customer maintenance" as the standard group, the original sequence index is [1, 2, 3], the standard index is [2, 1, 3], and the average displacement is , which is judged whether it is greater than the structure deviation reference 0.5. If it is greater, it is recorded as a high displacement difference item. In the matching field selection process, the same word items as the standard template fields are extracted from each group of position groups, recorded as matching field items and the total number of matches is counted. The total position is traversed, and finally the position structure difference amount is generated. The total difference group is 783, accounting for 42% of the total position group number.

[0020] The term mapping adjustment submodule calls the position structure difference amount, adjusts the original position group description according to the combination matching result between the position and the skill, maps it to the standard template expression, and obtains the position term normalization result; The 783 difference items marked in the post structure difference amount are called back to the corresponding post skill data set record according to the post ID, and the word group reordering and keyword adjustment operation is performed on each post data based on the comparison result of the post description word group and the skill keyword combination. In the adjustment order operation, read the index order of each word in the original post word group, refer to the highest matching template order template in the standard template, reorder the word group field according to the maximum overlap principle, for example, the original word group is "planning activity organization", the corresponding template is "organization activity planning", and the original word group is adjusted to "organization activity planning". In the keyword adjustment operation, the skill keyword frequency is used as the main sequence sorting weight, and the threshold frequency is set to 5. The words below 5 will not be included in the keyword adjustment target group. A total of 412 keywords with a frequency greater than or equal to 5 are selected, and these keywords are inserted, replaced or deleted according to the corresponding word group position in the post description. The adjusted word group constructs a standard expression format description content in the post description, and in the normalization mapping process, the adjusted word group is searched item by item using the standard post word library, and the closest standard post expression item is matched according to the minimum editing distance principle. The maximum character distance threshold is set to 3. If the original description "maintain guest" and the standard expression "maintain customer resources" have an editing distance of 3 (character replacement + addition), the mapping condition is met, and the mapping result is "maintain customer resources". Finally, 1865 post term normalization results are obtained, of which 783 are adjusted items and 1082 are items that do not need to be adjusted.

[0021] Please refer to Figure 3 The path modeling module comprises: The node frequency analysis submodule obtains the post term normalization result, counts the appearance frequency and stage node number of each stage path node of the employee task record, analyzes the order distribution of the node in each stage, establishes the node appearance change interval, and generates the stage jump change amount. The standardized post identification code corresponding to each employee at different time periods and the related skill field are extracted from the processed normalized post terminology library, and a mapping table with the task record is established. The path node frequency of each stage in the employee's task record is counted. The tenure period is divided according to a fixed window, and the time period length is set to 30 days. The task record of each employee is divided into several stage paragraphs according to the timestamp. For example, employee A records "data collection, data cleaning, modeling training, result analysis, report writing" in period 1, and the frequency of each node in this stage is counted. Assuming that "data collection" appears 4 times and "modeling training" appears 1 time, the frequency array is [4, 1, 1, 1, 1], and the number of nodes in this stage is 5. The node distribution order of each stage is counted. The task nodes are sorted in ascending order of timestamp, and the difference between the node order and the standard post task process is compared. If a node is ranked 2 in the standard process and ranked 4 in the actual record, the order offset value is 2. The order offset sum of all nodes in this stage is obtained by traversing all nodes, and the position change interval of each node in different stages is established. For example, "report writing" appears in the last position in stage 1 and in the second last position in stage 2, so the change interval is [5, 4]. After recording the position change interval set of all nodes in the whole period, the appearance order difference and frequency difference of all nodes in adjacent two stages are combined and summed, and then normalized to obtain the stage jump change value between each stage. If the jump change value between stage 1 and stage 2 is 11, the change intensity value of stage pair (1→2) is recorded.

[0022] The task path evaluation submodule calls the stage jump change value, compares the connection order between the task path nodes and the key nodes of the standard post task process, filters the node distribution continuity and missing paragraphs, judges the co-occurrence proportion of the actual task nodes and the standard path nodes, and obtains the key path fitting degree coefficient; The stage jump change value is called, and the task path nodes of each employee and the key nodes of the corresponding standard post task process are compared in sequence. When comparing the order, the standard process nodes are indexed and listed, for example, "data collection-cleaning-modeling-analysis-reporting" corresponds to the number [1, 2, 3, 4, 5], and the actual path nodes are numbered as [1, 3, 2, 4, 5] according to the task order. Each node is compared in turn to determine whether it is arranged in the standard order, and the continuity of the connection order matching paragraph is counted. If the order is inverted or jumps more than 2 positions, it is determined that the order is not continuous, and the node number of the order jump segment is selected and added to the discontinuous node set, for example, the node [2, 3] position jump is in the non-continuous interval of the standard, and it is recorded as 1 jump segment. Then the co-occurrence proportion of the actual task nodes and the standard path nodes is judged. Assuming that the standard process contains 8 key task nodes, and an employee completes 5 of them in a stage, the co-occurrence proportion is A multi-stage averaging method is used, averaging the co-occurrence ratios of each stage throughout the entire lifecycle as the critical path fit coefficient. If an employee's co-occurrence ratios across the three stages are 0.625, 0.75, and 0.5, then their fit coefficient is... Finally, the fit coefficient between the employee's task path and the standard job task process is recorded.

[0023] The behavior segmentation construction submodule monitors the distribution of node reconstruction behaviors in the time series based on the critical path fit coefficient, collecting the cumulative frequency, number of segmented nodes, node time interval, and span for each behavior type within a periodic time period, using the formula: ; Calculate the node behavior change index, group the behavior types and divide them into periodic segments to form a structured employment period segmentation identifier and establish the employment period division results; in, For the first The frequency normalization value of each path node is obtained by counting the number of times the node appears in the task records within the term of the node and dividing by the maximum node frequency in the same term. For the first The normalized value of the number of path nodes within a period segment is obtained by counting the number of nodes within that period segment and dividing by the maximum number of nodes in the entire period. For the first The normalized time interval between each node and its predecessor is obtained by calculating the difference in timestamps of the nodes and dividing it by the maximum time difference over the entire period. For the first The normalized span value of the path to which each node belongs within this period is obtained by calculating the span of the node index in the path and dividing it by the maximum span of the entire period. The total number of behavior nodes within the current behavior cycle segment is obtained by directly counting the number of nodes within this segment. This is a node behavior change index, representing the degree of change in the behavior sequence within a period, used for subsequent division of behavior stages. This is the index number of the behavior node within the periodic segment; According to the key path adherence coefficient, the employee task path stage with an adherence value less than 0.6 is taken as the key monitoring object. The node reconstruction behavior in these stages is identified, that is, the actual path node sequence appears sequence inversion, merging, disassembly, replacement and other behaviors with the standard process. By comparing the occurrence sequence of each node and the time stamp interval, it is judged whether the reconstruction behavior occurs. In the time sequence, the time stamp corresponding to all nodes is collected and the occurrence frequency of each behavior type is counted per stage, for example, the "node merging" of a certain employee occurs 6 times and the "node replacement" occurs 2 times in a 3-month cycle, so the behavior type frequency array is [6, 2]. The segmented node number, average time interval and span between nodes are recorded for each behavior period. The node number directly counts the total number of task nodes in the behavior segment, the time interval is calculated by the average of the node time stamp difference, and the span is the maximum value minus the minimum value of the node index. Then the above participation items are normalized and brought into the following formula: ; In the formula, represents the frequency normalized value of the th node in the period, which is obtained by dividing the number of occurrences of the node in the period task by the maximum frequency in the same period; is the normalized value of the number of path nodes in the th period segment, which is obtained by dividing the number of nodes in the period segment by the maximum number of nodes in the whole period; is the time interval normalized value between the th node and its previous node, which is obtained by dividing the time stamp difference by the maximum time difference; is the path span normalized value of the node in the period; is the total number of nodes in the current behavior segment.

[0024] Suppose there are 3 nodes in a behavior segment, record their frequency, time interval and path span as shown in the following table.

[0025] Table 2 Behavior segmentation normalization parameter table ; As shown in Table 2, the original frequency, time difference and span value of each node are used for subsequent calculation of normalized parameters. The above parameters are normalized to obtain: , , , ; , , ; , , ; and the formula is: ; The results are calculated as follows: the first term is: ; the second term is: ; the third term is: ; ; wherein the node behavior change index is an index for measuring the degree of change of the node task in the post behavior path of the employee within a certain time period. Its specific meaning is: in the structured knowledge graph path, the higher the density of task behavior nodes within a certain time period, the more compact the time distribution, and the more concentrated the span, the greater the value of the node behavior change index of the cycle, indicating that the employee has undergone centralized task reconstruction or duty switching behavior within the cycle. On the contrary, if the node frequency is dispersed, the path span is wide, and the behavior span is weak, the index value is small, representing strong behavior stability in the cycle. This parameter is directly used to construct the time section of the tenure period, and is the key data basis for subsequent node linkage recognition and path evolution judgment, and can be used to dynamically reveal the employee behavior rhythm and post transition potential. The results show that the node behavior change index , since the behavior stage variation division threshold is set to 0.3, the stages below this value are classified as "stable stage", and if higher than 0.6, it is "severe change stage", thus determining that the cycle segment is "moderate change stage". According to this, the tenure period segmentation identifier is established, and the cycle label is recorded in the personnel archives as "stage B (moderate reconstruction)".

[0026] Please refer to Figure 4 , the node linkage module includes: The responsibility intersection extraction submodule calls the responsibility field set in the original post node and the transferred post node according to the tenure period division result, calculates the overlap ratio between the number of matching content items in the text expression of the responsibility field and the expression vector of the semantic range of the field, constructs the intersection expression vector according to the coverage rate and content approximation score of the responsibility item, obtains the expression set after the overlap and judges the consistency of the information structure, and obtains the responsibility intersection ratio; According to the division result of the tenure cycle, the system calls the original post node and the transferred post node of the employee in a specific cycle, and extracts the corresponding responsibility field set respectively. When extracting the field, the "core responsibility section" field in the post description and the responsibility record table is taken as the target field, from which independent responsibility item phrases such as "formulate annual budget", "audit supplier contract", "monitor project progress" and the like are split and standardized into a field sequence set. Then, the matching operation between the responsibility fields is performed. First, the number of overlapping items in the text expression of the responsibility fields in the original post and the transferred post is calculated. The method is to calculate the longest common substring at the character level for each group of phrases. For example, the original post contains the field "manage budget expenditure", and the transferred post contains the field "budget expenditure approval", and the common substring is "budget expenditure", which is counted as 1 matching item. After repeated operation, the total number of matching fields is 12, the total number of fields is 20, and the field matching rate is . Further, the semantic vector expression model is called to construct semantic vectors for all responsibility items. The semantic similarity is calculated by the cosine angle between the vectors. For example, the cosine similarity between "cost analysis" and "budget control" is 0.82. The approximate degree judgment threshold is set to 0.75. The number of field pairs greater than the threshold is 8, and the semantic approximate coverage rate is . The weighted responsibility intersection expression score is calculated as by matching rate and approximate coverage rate with a set weight ratio of 1:2. The intersection expression vector is constructed accordingly. Then, the field order of the original post expression structure and the overlapping expression structure is aligned. If the field order structure is completely consistent, it is determined that the information structure is consistent, otherwise it is marked as structural difference. Finally, the post responsibility intersection ratio is calculated as 46.67%.

[0027] The cross-post mapping submodule filters the field group with path continuity label based on the post responsibility intersection ratio, detects the original position label of the target field group in the transferred post path structure, adjusts the mapping order and judges the overlap distribution degree of the remapped field position, extracts the overlapping structure segment and counts the connected path number range, and obtains the cross-node field connection density. Based on the above job responsibility intersection ratio results, records with a ratio higher than 40% are screened as job pairs with a mappable basis, all responsibility fields are extracted in these job pairs, and field groups with path continuity are marked, that is, these fields appear twice or more times in the task flow path. For example, the "project review" field appears 3 times in the path of multiple task stages, and the field group is marked as a continuous field group. The target job path structure is located for these field groups, and the position label where the field first appears in the imported job structure is identified, for example, "project review" first appears in node 6 in the imported path node, and the original position index is recorded as 6. Next, the field remapping sorting operation is performed, the original path field order is aligned to the imported path structure, the field index order is adjusted according to the shortest jump path principle, and whether the adjusted field index is continuous is judged. If the three new sorted fields are [6, 7, 8], it is recorded as complete coincidence, and if the sorting is [6, 9, 11], it is recorded as partial jump coincidence. Extract the coincidence structure segment, and count the number of paths connected by all field nodes in these structure segments. If there are 4 paths connected to the "approval process segment", the connection path number is 4. If the number of fields in the structure segment is 3, the connection density is The final cross-node field connection density is 1.33. This value will be used in the subsequent steps to calculate the degree of change of responsibility path.

[0028] The path change identification submodule calculates the number of difference items between the upstream and downstream node binding responsibility fields in the structure path, the connection order change degree, and the expression span difference according to the cross-node field connection density, using the formula: ; Calculate the responsibility expression order offset density in the job path, construct the structure change description label between the reconstructed path structure and the original path structure, and generate path evolution feature information; Wherein, represents the responsibility expression order offset density in the job path, which is used to measure the intensive degree of the difference between the original node and the imported node responsibility field expression relative to the path structure span, is the normalized expression order value of the first responsibility field in the upstream node of the original job, which is obtained by obtaining the ordering number of the field in the path node list, and dividing the number by the total number of job path nodes, is the normalized expression order value of the first responsibility field in the downstream node of the imported job, which is obtained by obtaining the new ordering number of the field in the reorganized path structure, and dividing the number by the number of reorganized path nodes, is the normalized expression order value of the first The normalized span value of the mapping field in the path is obtained by calculating the index difference between the starting position and the ending position of the node where the field is located, and dividing the total length of the nodes in the entire path, The normalized maximum order value in the mapping field is obtained by selecting the maximum order number of the field in the path of all structure nodes and dividing the total number of nodes, The normalized minimum order value in the mapping field is obtained by selecting the minimum order number of the field in the path of all structure nodes and dividing the total number of nodes, The total number of items identified in the path node structure for the responsibility field indicates the number of types of the responsibility field, The field sequence index currently being processed in the responsibility field set is used to locate the upstream and downstream order of the corresponding field in the structure, The total number of entries in the path structure to be processed indicates the number of path segments for which the span is calculated, The path number index currently being processed is used to extract the normalized value of the node span in the path; According to the cross-node field connection density, the difference between the upstream node binding responsibility field and the downstream node binding field in the original post path is calculated. First, the normalized order number of all responsibility fields in the original path and the new path is counted, for example, a responsibility field "report review" ranks 5th in the original path, and the total path length is 10, then its If its position in the new path is the second, and the number of reorganized path nodes is 8, then After performing this operation on all items of responsibility field, the square sum of the order difference is calculated and summed, for example, there are 3 items of field, and the upstream / downstream normalized values are [0.5, 0.6, 0.75] and [0.25, 0.4, 0.5] respectively, then the square sum of the offset is: The normalized order and difference value of the field are shown in the following table: Table 3 Example of responsibility path structure reconstruction parameter table ; As shown in Table 3, the expression order of the responsibility field between the original path and the new path changes little, and the square of the normalized difference value of each field remains between 0.04 and 0.0625, indicating that the post reconstruction belongs to local structure adjustment rather than overall rearrangement.

[0029] Next, the path span denominator part is calculated, assuming that there are 3 path segments involved, and the mapping field span in each path is 2, 3 and 2 respectively, and the total length of the path nodes is 8, then the normalized span is The sum is At the same time, the maximum value of the normalized position of all fields is 0.75, and the minimum value is 0.25, then the expression span difference is . Enter the formula to calculate the job path responsibility expression order offset density: ; In the job path, the responsibility expression order offset density refers to the density of the responsibility field expression order offset between the original job and the transferred job within the responsibility path structure span. This parameter is used to measure the relative relationship between the concentration of expression position offset and the carrying capacity of the path in the responsibility structure adjustment process. The larger the value, the more concentrated and significant the change in the expression order of the responsibility field, and the change occurs in the path with limited carrying capacity of the responsibility span structure, indicating that there is a risk of structural overlap compression in the path reconstruction area. This index can be used as an important basis for job transition risk identification and path adaptability evaluation, supporting dynamic balance analysis between node structure carrying capacity and expression disturbance in the job change process, and has practical data discrimination ability. The value represents the reconstruction density of the path responsibility expression order in the job change process. Since the offset density threshold is set to 0.2, a value below this indicates that the responsibility expression structure remains continuous, so the degree of path reconstruction is low, and the change label is marked as "mild path reconstruction", and the job evolution feature information table is recorded.

[0030] Please refer to Figure 5 , the fracture identification module includes: The empty post section identification submodule obtains the path evolution feature information, detects the node group in the post connection path that has no continuous personnel binding field configured, calculates the length of the node group, and compares it with the total path length to obtain the empty post continuous ratio; Obtaining path evolution feature information refers to reading the path structure change record and post connection link label generated by the previous module, and judging the personnel configuration state of the node segment in the marked reconstructed path. The judgment method is to search whether each path segment contains a specific employee ID or personnel identification code bound to the post node, and if two or more unbound personnel post nodes appear continuously, the system records it as an "empty post node group". Taking a typical path as an example, it contains node sequence [N1, N2, N3, N4, N5, N6, N7], among which N2, N3 and N4 are not configured with personnel binding fields, and the remaining nodes are bound with personnel fields. The system will mark N2-N4 as an empty post node group, and the node group length is 3. Calculate the ratio of the length of the empty post node group to the total number of path nodes to obtain the empty post continuous ratio, for example, the total path length is 7, the empty post node group length is 3, and the empty post continuous ratio is If the ratio exceeds the set continuous vacancy judgment threshold of 0.4, the node group is identified as a "continuous empty post section", and its position range in the path is recorded as N2-N4, and is written into the post structure risk candidate section set for further verification.

[0031] The responsibility field comparison submodule acquires a set of responsibility fields of the upstream and downstream connected nodes of the vacancy section based on the vacancy continuity ratio, analyzes the number of items in the field set, judges the difference between the number and the responsibility configuration benchmark value, and obtains the responsibility configuration offset degree; After confirming that the node section has a vacancy continuity ratio exceeding the threshold value, the system will extract a set of responsibility fields from the upstream and downstream connected nodes of the vacancy section. The specific method is as follows: locate the previous node and the next node of the vacancy section, read the set of job responsibility configuration fields, for example, node N1 contains responsibility items [budget formulation, cost verification], and node N5 contains responsibility items [reimbursement approval, supplier cooperation], and combine the two to form a set of upstream and downstream responsibility fields with 4 items. Then the system reads the responsibility configuration benchmark value corresponding to the post path, that is, the number of responsibility items that should be covered by the path type under complete configuration, for example, a certain type of procurement path should cover 6 core responsibility items. The system compares the number of items in the set of upstream and downstream nodes of the vacancy section (4 items here) with the benchmark value 6, and obtains that the number of missing responsibility items is 2, and the configuration offset degree is . If the offset degree is greater than the set offset threshold 0.25, it is determined that the upstream and downstream responsibility configuration of the vacancy section is incomplete, and there is a risk of responsibility fracture. The following table shows the field difference of the node section: Table 4 Comparison of upstream and downstream responsibility fields of vacancy section and configuration benchmark ; As shown in Table 4, the set of responsibility fields of the vacancy section has obvious missing compared with the benchmark value, and the responsibility configuration offset degree reaches 33.3%. Mark this section as "responsibility coverage anomaly" and enter the fracture structure identification process.

[0032] The fracture interval determination submodule filters the position where the responsibility field coverage is broken according to the responsibility configuration offset degree, combined with the node distribution in the post connection path and the responsibility field coverage, marks the fracture section range in the node sequence, and establishes the structure risk identification result; The system further traverses the structure of the post connection path and analyzes the field coverage according to the calculated responsibility configuration offset. First, the system checks the distribution of the responsibility fields of each node in the path to determine whether the responsibility fields can be continuously transmitted from the upstream to the downstream due to the structural fracture of the nodes. In the specific analysis, the system splits the post path into a node sequence in order and marks the responsibility fields of each node. If a field does not appear in any node, it is determined that the field is in a responsibility fracture state in the path. The node sequence is marked with a fracture section based on the position label of the empty post section and the mapping result of the responsibility field. For example, in the path nodes N1-N7, it is found that the responsibility field "project approval" is completely missing during N2-N4, and there is no inheritance field in the upstream and downstream nodes. Therefore, the responsibility field is marked as a fracture item. The system marks the N2-N4 nodes as a fracture section range and establishes a structural risk identification result, which includes the path ID, the fracture start and end node number, the missing responsibility item list, and the fracture severity level. If there are more than two fracture fields in a path, and the fracture section length is more than 30% of the total node number, the risk level is determined as "medium-high risk", and the record is entered into the structural warning module for operation and maintenance decision reference.

[0033] Referring to Figure 6 , the access control module comprises: The field cumulative statistics submodule obtains the structural risk identification result, analyzes the salary fields, evaluation fields and organizational function fields in the access path, and counts the cumulative number of each type of field in the path to obtain the total number of sensitive fields in the path; After the system obtains the structural risk identification result, it locates all post path nodes with fracture sections or responsibility coverage abnormalities, and regards these paths as key areas for access control, entering the sensitive field identification process. The fields involved in the access path are divided into three categories according to the content type: salary fields, evaluation fields and organizational function fields. The system traverses all nodes bound to the archive record or task field in these paths in turn, extracts the name and field code of each type of field. For example, the fields involved in path P101 include "basic salary", "quarterly evaluation level", "department affiliation", "salary adjustment coefficient", "post level" and "year-end award budget". Through the field type mapping rule, the system identifies 3 salary fields, 2 evaluation fields and 1 organizational function field. After counting the field types at the path level, the system obtains the cumulative number of each type of field in the path, and sums up the total number of sensitive fields in the path. For example, the cumulative number of salary fields in a path is 4, the cumulative number of evaluation fields is 3, and the cumulative number of organizational function fields is 2. Therefore, the total number of sensitive fields in the path is The distribution structure of each type of field is recorded as follows: Table 5: Path Sensitive Field Type Distribution Table ; As shown in Table 5, the sensitive field distribution in path P101 is relatively concentrated, especially in the salary field, which appears frequently. Subsequent comparison with user access permissions is required to determine whether the corresponding field visibility permission is available.

[0034] The role permission comparison submodule is based on the total number of path sensitive fields, calls the access level field group corresponding to the user role, compares the coverage ratio of the number of sensitive fields and the number of accessible fields in the role field set, and obtains the permission coverage ratio. After obtaining the total number of path sensitive fields, the system calls the role type of the current user and extracts the field access level group corresponding to the role from the access permission control table. Taking "first-level management position" as an example, its role access field group includes 2 accessible salary fields, 1 evaluation field, and 2 organizational function fields. The system matches by field type and counts the number of coincidences between the current role accessible field set and the path sensitive field. Taking path P101 as an example, the number of sensitive fields is 9, of which the fields consistent with the role field set are "basic salary", "annual bonus budget", "department affiliation", and "post level", a total of 4. The system calculates the permission coverage ratio as: If the coverage ratio is less than the set threshold of 0.6, the system marks the field access ability of the current role for the path as "insufficient permission coverage" and records the ratio and field details into the permission matching report as the basis for path permission state judgment.

[0035] The path permission judgment submodule judges the matching relationship between the sensitive fields in the path node and the role field set according to the permission coverage ratio, marks the access permission state of the path, and establishes the file access management result. According to the permission coverage ratio, the system enters the access permission state judgment process. The judgment logic is as follows: if the permission coverage ratio is greater than or equal to 0.8, it is considered as "complete permission"; if it is between 0.6 and 0.8, it is marked as "restricted access"; if it is less than 0.6, it is marked as "not accessed". Taking path P101 as an example, the number of accessible fields corresponding to the user role is 4, and the total number of sensitive fields is 9, the coverage ratio is 0.444, which is less than 0.6, meeting the "not accessed" rule, the system generates the path access permission state as "rejected" accordingly, and generates the access audit record, the record content includes user ID, access path number, sensitive field list, accessible field list and final judgment state. The system writes the permission state into the file access control table and updates the path control strategy to ensure that the user's subsequent access to the rejected fields in the path will trigger the access rejection or desensitization prompt mechanism.

[0036] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application in other forms. Any skilled person in the art can modify or change the disclosed technical content into equivalent embodiments with equivalent changes, and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application, without departing from the technical solution content of the present application, still falls within the protection scope of the present application.

Claims

1. A knowledge graph-based human resource record storage and management system, characterized in that, The system includes: The job identification module calls employee file information, analyzes the combination of job description phrases and skill keywords, compares the differences in field order and word usage between the job phrases and the standard template, performs expression mapping between job terminology and the standard template, and generates job terminology normalization results. Based on the job terminology normalization results, the path modeling module calculates the jump frequency and node span differences of path nodes in employee task records at multiple stages, analyzes the alignment of path connections with job responsibilities and processes, filters node reconstruction segments, establishes stage labels, and generates job cycle division results. The node linkage module uses the job cycle division results to determine the intersection of responsibility fields between job nodes, filter fields with cross-path connection, adjust the mapping order, calculate the crossover of upstream and downstream responsibilities, construct structural change description tags, and generate path evolution feature information. The fracture identification module combines the path evolution feature information to construct the job connection path, analyze the length of the continuous vacant job node group and the path ratio, determine the difference in responsibility fields of upstream and downstream nodes, delineate the scope of responsibility interruption, and generate structural risk identification results.

2. The knowledge graph-based human resource file storage and management system according to claim 1, characterized in that, The job terminology normalization result includes standard job fields, skill attribution tags, and expression mapping structure. The job cycle division result includes stage behavior identifiers, task span range, and path segment numbers. The path evolution feature information includes responsibility migration tags, node connection order, and evolution path structure. The structural risk identification result includes vacant job connection segments, responsibility breakpoint locations, and link interruption range.

3. The knowledge graph-based human resource file storage and management system according to claim 1, characterized in that, The job identification module includes: The job description phrase extraction submodule obtains employee file information, extracts job description phrases and skill keywords, analyzes the combination of phrases and keywords, establishes a job skill dataset, and generates job phrase data. The structure format comparison submodule compares the differences in field order and word usage between the job term data and the standard term template, based on the amount of job term data. It determines the order in which keywords appear in the template, filters fields that match the template structure, and generates the job structure difference quantity. The terminology mapping adjustment submodule calls the job structure difference quantity, adjusts the original job phrase descriptions according to the combination matching results between jobs and skills, maps them to standard template expressions, and obtains job terminology normalization results.

4. The knowledge graph-based human resource file storage and management system according to claim 3, characterized in that, The path modeling module includes: The node frequency analysis submodule obtains the normalization result of the job terminology, counts the frequency of occurrence of path nodes in each stage of the employee task record and the number of stage nodes, analyzes the sequential distribution of nodes in each stage, establishes the node occurrence change range, and generates the stage jump change amount. The task path evaluation submodule calls the stage jump change amount, compares the connection order between task path nodes and key nodes of the job standard task process, filters the continuity of node distribution and missing segments, judges the co-occurrence ratio of actual task nodes and standard path nodes, and obtains the critical path fit coefficient. The behavior segmentation construction submodule monitors the distribution of node reconstruction behavior in the time series based on the critical path fit coefficient, collects the cumulative frequency, number of segment nodes, node time interval and span of each behavior type in the periodic time period, calculates the node behavior change index, groups the behavior types and divides them into periodic segments, forms a structured job cycle segmentation identifier, and establishes the job cycle division result.

5. The knowledge graph-based human resource file storage and management system according to claim 4, characterized in that, The node linkage module includes: The responsibility intersection extraction submodule, based on the job cycle division results, calls the responsibility field set in the original job node and the transferred job node, calculates the overlap ratio between the number of matching content items of the responsibility field in the text expression and the semantic range expression vector of the field, constructs the intersection expression vector based on the coverage rate of the responsibility items and the content similarity score, obtains the overlapping expression set and judges the consistency of information structure, and obtains the job responsibility intersection ratio. The cross-job mapping submodule filters field groups with path continuity markers based on the job responsibility intersection ratio, detects the original position labels of the target field group in the job path structure, adjusts the mapping order and judges the degree of overlap of field positions after remapping, extracts overlapping structure segments and counts the range of connected paths, and obtains the cross-node field connection density. The path change identification submodule calculates the number of differences, the degree of change in connection order, and the difference in expression span between the binding responsibility fields of upstream and downstream nodes in the structural path based on the cross-node field connection density. It also calculates the responsibility expression order offset density in the job path, constructs structural change description tags between the reconstructed path structure and the original path structure, and generates path evolution feature information.

6. The knowledge graph-based human resource file storage and management system according to claim 5, characterized in that, The fracture detection module includes: The vacant post section identification submodule obtains the path evolution feature information, detects the continuous node groups without configured personnel binding fields in the post connection path, calculates the length of the node group, and compares it with the total path length to obtain the vacant post continuity ratio. The responsibility field comparison submodule collects the responsibility field set of upstream and downstream connecting nodes of the vacant post section based on the vacant post continuity ratio, analyzes the number of field set items, judges the difference between the number and the responsibility configuration benchmark value, and obtains the responsibility configuration offset. The fracture interval determination submodule, based on the responsibility configuration offset, combined with the node distribution and responsibility field coverage in the job connection path, filters out locations where there are fractures in the responsibility field coverage, marks the fracture segment range in the node sequence, and establishes structural risk identification results.

7. The knowledge graph-based human resource file storage and management system according to claim 1, characterized in that, The system also includes: Based on the structural risk identification results, the access control module analyzes the distribution of sensitive fields in the access path, calls the role field group to compare the field matching coverage ratio, determines the path access permission status, and generates file access management results. The file access management results include permission path labels, sensitive field density, and access judgment level.

8. The knowledge graph-based human resource file storage and management system according to claim 7, characterized in that, The access control module includes: The cumulative statistics submodule obtains the structural risk identification results, analyzes the salary field, performance evaluation field and organizational function field in the access path, counts the cumulative number of occurrences of each type of field in the path, and obtains the total number of path-sensitive fields. The role permission comparison submodule, based on the total number of path sensitive fields, calls the access level field group corresponding to the user role, compares the coverage ratio of the number of sensitive fields with the number of accessible fields in the role field set, and obtains the permission coverage ratio. The path permission determination submodule determines the matching relationship between sensitive fields and role field sets in the path node based on the permission coverage ratio, marks the access permission status of the path, and establishes the file access management result.

Images