Search

CN-121981692-A - Human resource comprehensive management big data supervision service system

CN121981692ACN 121981692 ACN121981692 ACN 121981692ACN-121981692-A

Abstract

The invention discloses a human resource comprehensive management big data supervision service system, which relates to the technical field of human resources. Judging whether the quality is qualified or not, if the quality is not qualified, triggering a cleaning flow, if the quality is qualified, carrying out data quality ranking, breaking data barriers of each stage, constructing a personal digital file penetrating through a professional life cycle, ensuring the integrity and relevance of the periodic data of a target object, greatly improving the detection efficiency and accuracy, reducing the manual error, defining the weight distribution of key indexes and auxiliary indexes, highlighting the data quality core dimension, providing a priority basis for a subsequent intelligent screening module, evaluating the on-job performance and off-job risk dimension evaluation predicted value of the target object, carrying out grading early warning judgment, assisting a manager to check the recruitment risk in advance, avoiding loss, reducing the subjectivity error of the recruitment decision, improving the recruitment efficiency and quality, and reducing the recruitment cost.

Inventors

  • WANG MINDONG
  • ZHANG LUNQI

Assignees

  • 上海梧桐范式数字科技有限公司

Dates

Publication Date
20260505
Application Date
20260126

Claims (10)

  1. 1. The utility model provides a manpower resources integrated management big data supervision service system which characterized in that includes: the basic data module is used for acquiring the mapping relation of the multi-source heterogeneous data of the target object in recruitment, incumbent and off-job stages, defining the association rule of key data fields of each stage and acquiring the periodic data of the target object; The intelligent supervision module is used for carrying out time sequence sequencing on the periodic data of the target object according to the time dimension, constructing a dynamic data management system, obtaining time dimension full-period data of the target object, automatically identifying data quality coincidence indexes of all time periods through an intelligent algorithm, judging whether the quality is qualified or not, triggering a cleaning flow if the quality is unqualified, and carrying out data quality ranking if the quality is qualified; The intelligent screening module analyzes the on-job information and off-job information of the human resource platform based on the data quality ranking through NLP, and comprises target object evaluation, a target object satisfaction questionnaire and off-job interview records, extracts candidate information, evaluates the on-job performance and off-job risk dimension evaluation predicted value of the target object, performs hierarchical early warning judgment, and judges the target object adaptation degree by combining computer vision analysis of the target object expression and limb language in video interview.
  2. 2. The human resource comprehensive management big data supervision service system according to claim 1, wherein the specific analysis method for defining the association rule of the key data field of each stage is as follows: And logically associating resume information, interview records, performance assessment in job phase, training records, rewards and punishment information, salary change, off-job application in off-job phase, off-job interview records and work handover list in recruitment phase by taking a unique identifier of a target object as a main key, constructing a personal digital file penetrating through a professional life cycle, and acquiring target object cycle data.
  3. 3. The human resource comprehensive management big data supervision service system according to claim 2, wherein the data quality compliance index of each time period is automatically identified by an intelligent algorithm, and the specific analysis method is as follows: Based on the obtained time dimension full period data of the target object, converting the time dimension full period data into a time sequence data set, extracting each data factor in the time sequence data set, wherein each time sequence fluctuation value, data repetition value, format specification value and field missing quantity of each time period are included, calculating the mean value and standard deviation of each data factor by adopting a moving average method, carrying out normalization processing, and carrying out weighted summation to obtain the data quality coincidence index of each time period.
  4. 4. The human resource comprehensive management big data supervision service system according to claim 1, wherein the judging quality is qualified, and the specific analysis method comprises the following steps: Based on the obtained data quality coincidence indexes of all time periods, comparing the data quality coincidence indexes of all time periods with the data quality coincidence index safety regions stored in the database, if the data quality coincidence indexes of a certain time period are in the data quality coincidence index safety regions, indicating that the data quality of the time period is in a qualified state, recording the data quality coincidence indexes of the time period as the data quality coincidence indexes of the qualified state of the time period, further obtaining the data quality coincidence indexes of the qualified state of all time periods, and if the data quality coincidence indexes of the certain time period are not in the data quality coincidence index safety regions, indicating that the data quality of the time period is unqualified.
  5. 5. The human resource comprehensive management big data supervision service system according to claim 1, wherein the data quality ranking is performed by a specific analysis method comprising: Based on the data quality coincidence index of the qualified state of each time period, constructing a multi-dimensional ranking index system, wherein the index system takes the data quality coincidence index as a key index, simultaneously takes the data integrity ratio, the data consistency deviation rate and the data timeliness achievement rate as auxiliary correction indexes, respectively endows the key index and each auxiliary correction index with weight values, calculates the comprehensive data quality score of each time period by adopting a weighted summation algorithm, sorts the target object periodic data of each time period according to the order of the comprehensive data quality score from high to low, generates a data quality ranking list, and marks the key index and the auxiliary index detail corresponding to each time period ranking.
  6. 6. The human resource comprehensive management big data supervision service system according to claim 1, wherein the evaluation target object performs on-job performance and off-job risk dimension evaluation prediction values, and the specific analysis method thereof is as follows: And constructing a target object period data and NLP analysis extracted candidate information based on the qualified data quality ranking, respectively performing normalization calibration on the target object period data and the NLP analysis extracted candidate information, and further obtaining target object on-job performance and off-job risk dimension evaluation prediction values.
  7. 7. The human resource comprehensive management big data supervision service system according to claim 1, wherein the on-job performance evaluation index system comprises the following specific analysis methods: The on-job performance evaluation index system takes a quantization index as a main material and takes a qualitative index as an auxiliary material, wherein the quantization index comprises performance assessment average score, training completion rate, reward and punishment coefficient and salary fluctuation range of a target object in an on-job period, the qualitative index comprises colleague evaluation equipartition, leadership evaluation equipartition and target object self satisfaction degree scoring extracted through NLP analysis, weights are respectively assigned to the indexes, equal weight distribution of the qualitative indexes is carried out, and a weighted summation algorithm is adopted to calculate the on-job performance evaluation value of the target object.
  8. 8. The human resource comprehensive management big data supervision service system according to claim 1, wherein the off-office risk prediction index system comprises the following specific analysis methods: the off-job risk prediction index system comprises dominant indexes and recessive indexes, wherein the dominant indexes comprise satisfaction questionnaires of target objects, deviation rates, post abnormal frequencies and off-job related keyword occurrence frequencies, the recessive indexes comprise on-job performance evaluation value fluctuation amplitude of the target objects, industry on-post off-job rate reference values and work handover preparation perfection, after normalization processing is carried out on the indexes, an off-job risk prediction model is constructed by adopting a logistic regression and random forest fusion algorithm, normalized values of the indexes are input into the model, and the off-job risk prediction value of the target objects is output.
  9. 9. The human resource comprehensive management big data supervision service system according to claim 1, wherein the hierarchical early warning judgment is performed by the specific analysis method: When the off-job risk prediction value is smaller than or equal to the off-job risk prediction safety value and the on-job performance evaluation value is larger than or equal to the on-job performance evaluation safety value, determining that primary early warning is carried out, wherein the primary early warning does not need to trigger a special treatment process, only the target object is brought into a conventional attention list, and is synchronously updated to the human resource management platform; When the off-job risk prediction value is smaller than or equal to the off-job risk prediction safety value and the on-job performance evaluation value is smaller than or equal to the on-job performance evaluation safety value, judging as secondary early warning, triggering an important investigation flow, automatically extracting NLP analysis details, on-job performance index fluctuation data and off-job risk index details of the target object, pushing the NLP analysis details, the on-job performance index fluctuation data and the off-job risk index details to corresponding human resource management personnel, reminding the personnel to finish targeted communication in a specified workday, investigating risk causes and recording treatment results; When the off-job risk prediction value is larger than the off-job risk prediction safety value, and the on-job performance evaluation value is larger than the on-job performance evaluation safety value, or when the off-job risk prediction value is larger than the off-job risk prediction safety value, and the on-job performance evaluation value is smaller than the on-job performance evaluation safety value, the method judges that three-level early warning is performed, immediately triggers an emergency treatment process, synchronously pushes early warning notification to a responsible person of a department to which a target object belongs and a key manager of human resources, clearly requires to start special butt joint in a specified working day, and makes handover preparation in advance.
  10. 10. The human resource comprehensive management big data supervision service system according to claim 1, wherein the specific analysis method is that the expression and the language of the object in the video interview are analyzed by combining with the computer vision, and the adaptation degree of the object is judged, and the specific analysis method is that: The method comprises the steps of obtaining video interview original data of a target object recruitment stage, combining a data quality ranking result of an intelligent supervision module, screening video interview data with qualified data quality, extracting facial features and limb key features of the target object, converting the video interview data into facial feature vectors and limb key feature vectors of the target object, normalizing the facial feature vectors and the limb feature vectors to eliminate dimension influence, adopting a weighted summation algorithm, giving weights corresponding to expression features and limb features by combining a post adaptation standard, weighting and summing to obtain expression adaptation score and limb adaptation score of the target object, further obtaining comprehensive suitability score, comparing the comprehensive suitability score with a post adaptation threshold preset in a database, judging that the target object is adapted to the corresponding post if the comprehensive suitability score is more than or equal to the adaptation threshold, judging that the target object is not adapted to the corresponding post if the comprehensive suitability score is less than the adaptation threshold, and carrying out early warning prompt.

Description

Human resource comprehensive management big data supervision service system Technical Field The invention relates to the technical field of human resources, in particular to a human resource comprehensive management big data supervision service system. Background Along with the expansion of enterprise scale and the aggravation of personnel flow, multi-source heterogeneous data generated by target objects in recruitment, on-duty and off-duty full life cycles are increasingly complex, lack of unified association rules, difficulty in forming a complete data system penetrating through the professional life cycles, poor data interoperability, lack of dynamic management mechanisms, outstanding problems of fluctuation, repetition, deletion and the like of data time sequences, difficulty in guaranteeing data reliability, so that analysis of a human resource comprehensive management big data supervision service system is necessary. The invention patent application patent of CN115170051A discloses a human resource comprehensive management big data supervision service system, which relates to the technical field of information management and comprises a data acquisition module, a data storage module, a background server, an information release module, a main part preselection module and a training screening module; the data acquisition module acquires staff information through an administrator mode and a staff application mode, so that the enterprise information can be updated in time, meanwhile, when staff actively submits an information modification request, the staff information is statistically verified, irrelevant staff is prevented from tampering the information, the authenticity of modified content is guaranteed, the information release module is used for sequencing and releasing the staff information according to the business coefficients of the staff, the enterprise staff is subjected to comparison excitation, the training screening module is used for selecting staff limiting the number according to the promotion coefficients to conduct candidate trunk training, the maximum potential is exerted, the personal and the performance of the enterprise are improved, and the dual development of the enterprise and the individual is realized. The system for supervising and managing the large data of the human resources comprehensively can meet basic requirements, but has potential defects and challenges, and is particularly characterized in that the attention to sequential sequencing of the periodic data of the target object according to the time dimension is low in the prior art, so that the construction of a dynamic data governance system is influenced, the identification of the data quality coincidence index of each time period is influenced, the detection efficiency and accuracy are reduced, the manual error is improved, the data quality is reduced, and the problem of providing priority basis for a follow-up intelligent screening module is influenced. 2. In the prior art, the evaluation of the target object on-job performance and off-job risk dimension evaluation predicted value is not accurate enough, the grading of early warning judgment is influenced, the problem that the treatment flows and responsibility main bodies of different early warning levels are not clear enough is caused, the subjectivity error of recruitment decision is improved, the recruitment efficiency and quality are reduced, and the recruitment cost is further improved. Disclosure of Invention The invention aims to provide a human resource comprehensive management big data supervision service system, which solves the problems existing in the background technology. In order to solve the technical problems, the invention adopts the following technical scheme that the invention provides a human resource comprehensive management big data supervision service system, which comprises a basic data module, an intelligent supervision module and an intelligent screening module. And the basic data module is used for acquiring the mapping relation of the multi-source heterogeneous data of the target object in recruitment, incumbent and off-job stages, determining the association rule of the key data field of each stage and acquiring the cycle data of the target object. And the intelligent supervision module is used for carrying out time sequence sequencing on the periodic data of the target object according to the time dimension, constructing a dynamic data management system, obtaining time dimension full-period data of the target object, automatically identifying the data quality coincidence index of each time period through an intelligent algorithm, judging whether the quality is qualified or not, triggering a cleaning flow if the quality is unqualified, and carrying out data quality ranking if the quality is qualified. The intelligent screening module analyzes the on-job information and off-job information of the human resource platform based on the data