A remote human resources performance management system and method for qualification review

Abstract

The present invention relates to the technical field of qualification audit performance management, and proposes a remote human resources performance management system and method for qualification audit, the system comprising: a task requirement analysis module, a task generation and allocation module, an audit operation supervision module and a comprehensive evaluation module. The present invention introduces a task difficulty identification mechanism, considers factors such as data complexity and domain expertise, accurately evaluates task difficulty, generates personalized task lists for auditors according to the difficulty of different tasks in different qualification audit projects, and generates comprehensive performance scores by comprehensively considering task difficulty and operational behavior characteristics, thereby improving audit accuracy, realizing personalized performance evaluation, and optimizing task planning by estimating the review time, thereby improving overall work efficiency, and providing enterprises with a smarter and more effective remote human resources management and performance system.

Description

Technical Field

[0001] The present invention relates to the technical field of qualification review performance management, and in particular to a remote human resources performance management system and method for qualification review. Background Art

[0002] With the digital transformation of enterprises and the popularity of remote work, traditional work management methods can no longer adapt well to the increasingly decentralized work environment. In the field of qualification review, the content and difficulty of auditors' work vary, and a single working time standard often cannot accurately evaluate the quality of work. In addition, traditional attendance systems have difficulty capturing the subtle differences in work content, resulting in an inability to fully evaluate employees' actual contributions.

[0003] The problems existing in the existing technologies in remote human resources management and performance supervision are mainly reflected in the inability to comprehensively evaluate the quality of work content, the lack of a mechanism to identify "ineffective work", and the lack of a personalized supervision mechanism, including:

[0004] (1) Lack of work content evaluation mechanism

[0005] Existing technologies mainly use auditors’ working hours as the review standard, but they cannot effectively evaluate the quality of the work content. Since auditors handle different qualification review tasks when performing their work, their work content and difficulty vary. Therefore, it is easy to ignore the difference in work quality when relying solely on working hours for review. This problem may lead to inaccurate audit results because the actual work value of different auditors in the same time may vary, and existing technologies cannot solve this problem.

[0006] (2) Lack of a mechanism to identify “ineffective work”

[0007] Existing methods often fail to judge the effectiveness of work, that is, whether it is real and high-quality work. Since the review criteria are mainly based on working hours, it is difficult to identify whether the auditor has performed substantive tasks that are beneficial to the qualification review during working hours. This makes it impossible for existing technologies to effectively distinguish between effective and ineffective work, resulting in inaccurate assessments of auditor performance. In this case, it is difficult for companies to accurately manage and motivate work performance.

[0008] (3) Lack of personalized supervision mechanism

[0009] Existing technologies usually use a unified "maximum tolerable working time" to limit auditors' work, but this approach cannot fully take into account the differences in work content and difficulty. Since the nature of auditors' work varies, a fixed time limit cannot meet all situations. Therefore, existing methods are somewhat rigid in remote supervision and cannot achieve personalized supervision of different types of work. This limits the company's flexible management of remote work and causes a loss of management efficiency. Summary of the invention

[0010] In order to solve the above-mentioned problems of the prior art, the present invention provides a remote human resources performance management system and method for qualification review, aiming to solve the problems that the prior art cannot comprehensively evaluate the quality of the qualification review and evaluation work content, lacks a judgment mechanism for "invalid work" and lacks a personalized supervision mechanism.

[0011] A first aspect of the present invention provides a remote human resources performance management system for qualification review, the system comprising:

[0012] A task requirement analysis module, wherein the task requirement analysis module is configured to obtain the qualification review requirements of the target qualification review project, and determine the task elements of the target qualification review project according to the qualification review requirements;

[0013] A task generation and assignment module, wherein the task generation and assignment module is configured to establish a task list of a target qualification review project based on the task elements, generate a standard review duration for each task according to a difficulty score of each task in the task list, and assign the tasks in the task list to reviewers based on the standard review duration;

[0014] An audit operation supervision module, wherein the audit operation supervision module is configured to collect the operational behavior characteristics of the auditors in the audit system, associate a timestamp with each operational behavior characteristic, and generate a performance score for each auditor based on the operational behavior characteristics;

[0015] A comprehensive evaluation module is configured to generate a comprehensive performance score based on the difficulty score and performance score of each auditor performing the audit task during the target period.

[0016] Optionally, the task requirement analysis module specifically includes:

[0017] A qualification review requirement input unit, wherein the qualification review requirement input unit is configured to obtain the qualification review requirement of the target qualification review project input by the user; wherein the qualification review requirement is natural language information describing the target qualification review project;

[0018] A task element extraction unit, wherein the task element extraction unit is configured to extract the task elements in the qualification review requirements by using keyword matching and / or semantic matching; wherein the task elements include the field to which the enterprise belongs, the qualification review type and the task characteristics.

[0019] Optionally, the task generation and allocation module specifically includes:

[0020] A task list establishing unit, wherein the task list establishing unit is configured to match a number of tasks corresponding to the target qualification review project in a comparison table of task elements and task types according to the task elements of the target qualification review project, and establish a task list;

[0021] A task attribute labeling unit, wherein the task attribute labeling unit is configured to extract document attributes of the relevant review materials of each task in the task list, and label the document attributes as task attributes of the corresponding task; wherein the document attributes include data difficulty and task domain;

[0022] A task allocation unit, wherein the task allocation unit is configured to calculate a difficulty score of each task according to the task attributes of each task, generate a standard review duration for each task according to the difficulty score of each task, and allocate the tasks in the task list to auditors based on the standard review duration.

[0023] Optionally, the task allocation unit specifically includes:

[0024] a task difficulty score calculation subunit, wherein the task difficulty score calculation subunit is configured to calculate a task difficulty score according to the data difficulty and the task domain of each task;

[0025] The standard review time calculation subunit is configured to calculate the standard review time of each task according to the preset basic review time and the task difficulty score; wherein the expression of the standard review time is specifically:

[0026] T 审核 =T 基础 +k·S 任务难度

[0027] Among them, k is a constant, which represents the coefficient of the impact of task difficulty on the review time;

[0028] A task allocation subunit, wherein the task allocation subunit is configured to calculate the audit time allocated to each auditor according to the work efficiency of each auditor and the standard audit time of each person, and evenly allocate tasks according to the audit time of each auditor;

[0029] The expression of the audit time allocated to each auditor is specifically:

[0030]

[0031] Among them, the work efficiency of each auditor i is E i It is a parameter indicating the speed of its review.

[0032] Optionally, an expression to calculate the task difficulty score based on the data difficulty and task domain of each task, specifically:

[0033] S 任务难度 =w1×S 数据难度 +w2×S 领域难度

[0034] S 数据难度 =u1×S 复杂度 +u2×S 数据量 +u3×S 数据类型

[0035]

[0036]

[0037] Among them, w1 is the data difficulty score S 数据难度 The weight of w2 is the domain difficulty score S 领域难度 The weight of u1 is the complexity score S 复杂度 The weight of u2 is the data volume score S 数据量 u3 is the data type score S 数据类型 The weight, S i is the score of the i-th expert on the difficulty of the field, n is the number of experts, and F ij Score the domain features used to assess domain difficulty in the expert evaluation model, v j Score the domain feature F ij The weight of , m is the number of domain features.

[0038] Optional, complexity score S 复杂度 , Data volume score S 数据量 And the data type score S 数据类型 The expression is:

[0039]

[0040]

[0041]

[0042] Where: Complexity score S 复杂度 , Data volume score S数据量 And the data type score S 数据类型 The score is calculated by natural language processing tools.

[0043] Optionally, the audit operation supervision module specifically includes:

[0044] An operation behavior feature collection unit, wherein the operation behavior feature collection unit is configured to collect the auditor's operation behavior features in the audit system to form an operation behavior feature record set; wherein the operation behavior features include one or a combination of mouse click position, click frequency, operation path, and operation time;

[0045] A timestamp marking unit, the timestamp marking unit being configured to obtain a timestamp of each operation behavior feature and mark the timestamp to the corresponding operation behavior feature;

[0046] A performance scoring unit is configured to use a deep learning model based on a long short-term memory network (LSTM) to perform learning and training on the temporal relationship and performance scoring of operational behavior features marked with timestamps, and use the scoring model obtained after the training to obtain the performance score of each auditor when reviewing each task.

[0047] Optionally, the audit operation supervision module further includes:

[0048] Performance rating adjustment unit;

[0049] The performance score adjustment unit is configured to obtain the auditor's performance score influencing factors, generate the auditor's performance impact score using fuzzy reasoning, and generate the auditor's comprehensive performance score based on the performance impact score and the original performance score;

[0050] Among them, fuzzy reasoning is used to generate the expression of the auditor's performance impact score, specifically: P' i =F'(P i , performance score influencing factors); among them, P i is the original performance score, F' represents fuzzy reasoning, and P' i It represents a performance impact score, wherein the performance impact factor includes a factor representation of each task difficulty score.

[0051] Optionally, the comprehensive evaluation module specifically includes:

[0052] Comprehensive performance rating unit;

[0053] Wherein, the comprehensive performance scoring unit is configured to generate a comprehensive performance score of the auditor based on the performance impact score and the original performance score;

[0054] Among them, the expression for generating the auditor's comprehensive performance score is: 绩效 =α·P i +(1-α)·P' i ; Where α is the weight coefficient of the performance impact score and the original performance score.

[0055] A second aspect of the present invention provides a remote human resources performance management method for qualification review, the method comprising:

[0056] S1: Obtain the qualification review requirements of the target qualification review project, and determine the task elements of the target qualification review project based on the qualification review requirements;

[0057] S2: Based on the task elements, a task list of the target qualification review project is established, a standard review duration is generated for each task according to the difficulty score of each task in the task list, and the tasks in the task list are assigned to the reviewers based on the standard review duration;

[0058] S3: Collect the auditor's operational behavior characteristics in the audit system, associate a timestamp with each operational behavior characteristic, and generate a performance score for each auditor based on the operational behavior characteristics;

[0059] S4: Generate a comprehensive performance score based on the difficulty score and performance score of each auditor performing the audit tasks during the target period.

[0060] The beneficial effects of the present invention are as follows: a remote human resources performance management system and method for qualification review is proposed, which introduces a task difficulty identification mechanism, considers factors such as data complexity and domain expertise, accurately evaluates task difficulty, generates personalized task lists for auditors according to the difficulty of different tasks in different qualification review projects, and generates comprehensive performance scores by comprehensively considering task difficulty and operational behavior characteristics, thereby improving review accuracy, realizing personalized performance evaluation, and optimizing task planning by estimating review time, thereby improving overall work efficiency and providing enterprises with a smarter and more effective remote human resources management and performance system. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a schematic diagram of the structure of the remote human resources performance management system for qualification review provided by the present invention;

[0062] Figure 2 This is a flow chart of the remote human resources performance management method for qualification review provided by the present invention.

[0063] Reference numerals:

[0064] 10-Task requirement analysis module; 20-Task generation and allocation module; 30-Audit operation supervision module; 40-Comprehensive evaluation module. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] Embodiment 1:

[0067] Reference Figure 1 , Figure 1 A schematic diagram of the structure of a remote human resources performance management system for qualification review provided by an embodiment of the present invention.

[0068] like Figure 1 As shown, a remote human resources performance management system for qualification review, the system includes: a task requirement analysis module 10, the task requirement analysis module 10 is configured to obtain the qualification review requirements of the target qualification review project, and determine the task elements of the target qualification review project based on the qualification review requirements; a task generation and allocation module 20, the task generation and allocation module 20 is configured to establish a task list of the target qualification review project based on the task elements, generate a standard review time for each task according to the difficulty score of each task in the task list, and allocate the tasks in the task list to auditors based on the standard review time; an audit operation supervision module 30, the audit operation supervision module 30 is configured to collect the operating behavior characteristics of the auditor in the audit system, associate a timestamp with each operating behavior characteristic, and generate a performance score for each auditor based on the operating behavior characteristics; a comprehensive evaluation module 40, the comprehensive evaluation module 40 is configured to generate a comprehensive performance score based on the difficulty score and performance score of each auditor performing the audit task in the target time period.

[0069] The problems existing in the prior art in remote human resources management and performance supervision are mainly reflected in the inability to comprehensively evaluate the quality of work content, the lack of a mechanism to distinguish "ineffective work", and the lack of a personalized supervision mechanism. In order to solve the above problems, this embodiment extracts task elements from the qualification review requirements of the target qualification review project, and then generates a task list based on the task elements, and introduces a task difficulty identification mechanism, taking into account factors such as data complexity and domain expertise, accurately assesses the difficulty of the task, and generates a personalized task list for the auditor according to the difficulty of different tasks in different qualification review projects, ensuring that the tasks faced by the auditor have a consistent difficulty level, and can more reasonably allocate tasks, thereby helping to improve the accuracy of the audit and ensure that the auditor's ability to handle tasks of different difficulty levels is more balanced. At the same time, by comprehensively considering the difficulty of the task and the characteristics of the operation behavior, a comprehensive performance score is generated to improve the accuracy of the audit. Through real-time supervision and deep learning methods, the system can comprehensively analyze the auditor's operation behavior, generate a personalized performance evaluation for it, and optimize the task planning by estimating the review time, thereby improving the overall work efficiency and providing enterprises with a smarter and more effective remote human resources management and performance system.

[0070] In a preferred embodiment, the task requirement analysis module specifically includes: a qualification review requirement input unit, which is configured to obtain the qualification review requirements of the target qualification review project input by the user; wherein the qualification review requirements are natural language information describing the target qualification review project; a task element extraction unit, which is configured to extract the task elements in the qualification review requirements by keyword matching and / or semantic matching; wherein the task elements include the field to which the enterprise belongs, the qualification review type and the task characteristics.

[0071] Considering that different qualification audit projects have different task elements, this embodiment obtains the detailed qualification audit requirements input by the qualification audit department, which records information such as the field to which it belongs, the qualification audit type (for example, quality management system certification, environmental management system certification, etc.), and task characteristics (for example, document type, data complexity, etc.), and extracts the task elements through in-depth analysis of the qualification audit requirements to generate the audit tasks of the target qualification audit project. Specifically, the system deeply analyzes the detailed qualification audit requirements provided by the enterprise qualification audit department. First, the system has a detailed understanding of the field in which the enterprise is located, taking into account the laws and regulations and industry specificities that may exist in different fields. Secondly, the system clearly identifies and understands in detail the specific audit types that the enterprise wants to conduct, such as ISO 9001 quality management system certification or ISO 14001 environmental management system certification, to ensure that the system meets the corresponding regulations and standards. Further, the system deeply analyzes the task characteristics, including document type and data complexity, to establish a basis for document management and assessment of task difficulty. The system also pays attention to the correlation between tasks and the matching of tasks and audit types to ensure the logic and consistency of the audit process. Finally, the system takes into account regulatory compliance and establishes a mechanism to monitor and adapt to regulatory changes to ensure that the audit always meets the latest regulatory requirements. Based on this, the system can accurately understand the company's qualification audit needs, extract task elements from the qualification audit needs, and use task elements to provide a basis for subsequent task list generation and task difficulty assessment, so that the generated task list fits the current qualification audit project and improves the accuracy of task difficulty assessment.

[0072] In a preferred embodiment, the task generation and assignment module specifically includes: a task list establishment unit, which is configured to match a number of tasks corresponding to the target qualification review project in a comparison table of task elements and task types according to the task elements of the target qualification review project, and establish a task list; a task attribute labeling unit, which is configured to extract document attributes of relevant review materials for each task in the task list, and label the document attributes as task attributes for the corresponding task; wherein the document attributes include data difficulty and task domain; a task assignment unit, which is configured to calculate the difficulty score of the task according to the task attributes of each task, generate a standard review time for each task according to the difficulty score of each task, and assign the tasks in the task list to the auditor based on the standard review time.

[0073] In this embodiment, after obtaining the task elements of the target qualification review project, a task list can be established by extracting several tasks that best match the target qualification review project from a comparison table of task elements and task types. Thereafter, the task attributes are annotated by extracting the document attributes of the relevant review materials of each task, and then the difficulty score and standard review time of each task are calculated based on the annotated task attributes. Thus, by obtaining the standard review time for each task, it can be ensured that when allocating tasks and generating a task list for each auditor, the different time required for auditors to process tasks of different difficulty levels is taken into consideration, and a personalized list with relatively balanced review time is generated. By estimating the review time based on the task difficulty score, the system can plan task time more accurately. The adaptive task allocation and performance evaluation mechanism helps to optimize the review process, improve overall work efficiency, save time and resources for the enterprise, make task allocation more flexible and reasonable, and improve overall review efficiency and the scientific nature of human resource management.

[0074] Specifically, the task allocation unit includes: a task difficulty score calculation subunit, which is configured to calculate the task difficulty score according to the data difficulty and task field of each task; a standard review time calculation subunit, which is configured to calculate the standard review time of each task according to the preset basic review time and the task difficulty score; wherein the expression of the standard review time is specifically:

[0075] T 审核 =T 基础 +k·S 任务难度

[0076] Among them, k is a constant, which represents the coefficient of the impact of task difficulty on the review time;

[0077] A task allocation subunit, wherein the task allocation subunit is configured to calculate the audit time allocated to each auditor according to the work efficiency of each auditor and the standard audit time of each person, and evenly allocate tasks according to the audit time of each auditor;

[0078] The expression of the audit time allocated to each auditor is specifically:

[0079]

[0080] Among them, the work efficiency of each auditor i is E i It is a parameter indicating the speed of its review.

[0081] Among them, the expression for calculating the task difficulty score based on the data difficulty and task domain of each task is as follows:

[0082] S任务难度 =w1×S 数据难度 +w2×S 领域难度

[0083] S 数据难度 =u1×S 复杂度 +u2×S 数据量 +u3×S 数据类型

[0084]

[0085]

[0086] Among them, w1 is the data difficulty score S 数据难度 The weight of w2 is the domain difficulty score S 领域难度 The weight of u1 is the complexity score S 复杂度 The weight of u2 is the data volume score S 数据量 u3 is the data type score S 数据类型 The weight, S i is the score of the i-th expert on the difficulty of the field, n is the number of experts, and F ij Score the domain features used to assess domain difficulty in the expert evaluation model, v j Score the domain feature F ij The weight of , m is the number of domain features.

[0087] Among them, the complexity score S 复杂度 , Data volume score S 数据量 And the data type score S 数据类型 The expression is:

[0088]

[0089]

[0090]

[0091] Where: Complexity score S 复杂度 , Data volume score S 数据量 And the data type score S 数据类型 The score is calculated by natural language processing tools.

[0092] In this embodiment, for the scoring of data difficulty, the TF-IDF (Term Frequency-Inverse Document Frequency) algorithm based on natural language processing tools is used to process semantic information, and the average sentence length, vocabulary diversity, actual number of words, actual number of paragraphs, number of each data type and the corresponding maximum score of the relevant review document corresponding to each task are calculated through syntactic analysis, and the complexity score S is calculated respectively. 复杂度 , Data volume score S 数据量 And the data type score S 数据类型 , and then calculate the data difficulty score by using weighted average; for the score of the data field, a comprehensive evaluation model based on expert knowledge is used for evaluation. In practical applications, the comprehensive evaluation model is based on the subjective judgment of experts, and the field difficulty evaluation is achieved by scoring the key features and weighted summing them; assuming that there are n experts, each expert i has m key features F in the field ij The scoring range is between [0,1]. At the same time, each feature F ij Assign a weight w j , indicating the relative importance of the feature to the domain difficulty. Expert i’s assessment of the domain difficulty S i It can be obtained by weighted summing up the scores of each feature to obtain the overall domain difficulty assessment S 领域 , the evaluation results of all experts can be averaged or weighted averaged, with weight v j It can be determined through expert discussion, data analysis or other methods; in the preferred solution, different weights can also be considered for different experts to improve the accuracy of the assessment; for the calculation of the standard audit time, it is determined by comprehensively calculating factors such as the basic audit time, the difficulty of the task, and the auditor's work efficiency, to ensure that in the task allocation process, the tasks of different difficulty levels are processed by the auditors, and the required audit time is relatively balanced. By adjusting the value k of the coefficient of the impact of task difficulty on the audit time, the degree of impact of task difficulty on the audit time can be controlled, making task allocation more flexible and reasonable, thereby ensuring that the tasks faced by auditors have a consistent difficulty level. This helps to improve the accuracy of the audit and ensure that the auditors have a more balanced ability to handle tasks of different difficulty levels.

[0093] In a preferred embodiment, the audit operation supervision module specifically includes: an operation behavior feature collection unit, which is configured to collect the operation behavior features of the auditor in the audit system to form an operation behavior feature record set; wherein the operation behavior features include one or a combination of mouse click position, click frequency, operation path, and operation time; a timestamp marking unit, which is configured to obtain the timestamp of each operation behavior feature and mark the timestamp to the corresponding operation behavior feature; a performance scoring unit, which is configured to use a deep learning model based on a long short-term memory network LSTM to perform learning and training on the timing relationship and performance scoring of the operation behavior features marked with timestamps, and use the scoring model obtained after the training to obtain the performance score of each auditor when auditing each task.

[0094] The audit operation supervision module further includes: a performance score adjustment unit; wherein the performance score adjustment unit is configured to obtain the auditor's performance score influencing factors and generate the auditor's performance impact score using fuzzy reasoning; and generate a comprehensive performance score for each auditor based on the performance impact score and the original performance score; wherein the expression for generating the auditor's performance impact score using fuzzy reasoning is specifically: P' i =F'(P i , performance score influencing factors); among them, P i is the original performance score, F' represents fuzzy reasoning, and P' i It represents a performance impact score, wherein the performance impact factor includes a factor representation of each task difficulty score.

[0095] Specifically, the comprehensive evaluation module includes: a comprehensive performance scoring unit; wherein the comprehensive performance scoring unit is configured to generate a comprehensive performance score of the auditor according to the performance impact score and the original performance score; wherein the expression for generating the comprehensive performance score of the auditor is specifically: P 绩效 =α·P i +(1-α)·P' i ; Where α is the weight coefficient of the performance impact score and the original performance score.

[0096] In this embodiment, the audit operation supervision module generates operation records with timestamps by real-time monitoring of the auditor's operation behavior in the system, including mouse click frequency, input content, operation path, operation time, etc., to provide an analysis basis for deep learning operation scoring; specifically, each operation record is represented as a vector X i , including information such as mouse click location, click frequency, input content, etc., and at the same time, the timestamp of each operation record is associated: X i =[xi1 ,x i2 ,…,x in ,Timestamp i ], a deep learning model based on long short-term memory network LSTM is constructed to learn the temporal relationship between operation records. The output of the model is the performance score P of each operation record. i ; Among them, P i =LSTM(X i ). After that, the characterization of the performance score influencing factors considering the difficulty score of each task is introduced, and fuzzy reasoning P' is performed in combination with fuzzy rules. i =F'(P i , performance score influencing factors). Finally, in order to comprehensively consider the task difficulty and operation behavior, the task difficulty assessment and operation score are combined to comprehensively evaluate the auditor's performance, that is, the performance score P output by the deep learning model i The score P' after fuzzy rule adjustment i Combined, the final reviewer performance score P is obtained 绩效 As a result, more objective and comprehensive performance evaluation results are obtained, avoiding the problem of relying solely on simple quantitative indicators while ignoring quality, and providing enterprises with a more scientific management basis.

[0097] Reference Figure 2 , Figure 2 A flowchart of a remote human resources performance management method for qualification review provided by an embodiment of the present invention.

[0098] like Figure 2 As shown, a remote human resource performance management method for qualification review includes the following steps:

[0099] S1: Obtain the qualification review requirements of the target qualification review project, and determine the task elements of the target qualification review project based on the qualification review requirements;

[0100] S2: Based on the task elements, a task list of the target qualification review project is established, a standard review duration is generated for each task according to the difficulty score of each task in the task list, and the tasks in the task list are assigned to the reviewers based on the standard review duration;

[0101] S3: Collect the auditor's operational behavior characteristics in the audit system, associate a timestamp with each operational behavior characteristic, and generate a performance score for each auditor based on the operational behavior characteristics;

[0102] S4: Generate a comprehensive performance score based on the difficulty score and performance score of each auditor performing the audit tasks during the target period.

[0103] In this embodiment, by introducing a task difficulty identification mechanism, factors such as data complexity and domain expertise are taken into consideration to accurately assess task difficulty. A personalized task list is generated for auditors based on the difficulty of different tasks in different qualification review projects. By comprehensively considering task difficulty and operational behavior characteristics, a comprehensive performance score is generated to improve review accuracy, achieve personalized performance evaluation, and optimize task planning by estimating review time, thereby improving overall work efficiency and providing enterprises with a smarter and more effective remote human resources management and performance system.

[0104] The specific implementation methods of the remote human resources performance management method for qualification review of the present application are basically the same as the above-mentioned embodiments of the remote human resources performance management system for qualification review, and will not be repeated here.

[0105] In the description of the embodiments of the present invention, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "center", "top", "bottom", "top", "bottom", "inside", "outside", "inner side", "outer side" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Among them, "inside" refers to an internal or enclosed area or space. "Periphery" refers to the area surrounding a specific component or a specific area.

[0106] In the description of the embodiments of the present invention, the terms "first", "second", "third", and "fourth" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", and "fourth" may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0107] In the description of the embodiments of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "install", "connect", "connect", and "assemble" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0108] In the description of the embodiments of the present invention, specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0109] In the description of the embodiments of the present invention, it is to be understood that "-" and "~" represent a range between two values, and the range includes the endpoints. For example: "AB" represents a range greater than or equal to A and less than or equal to B. "A~B" represents a range greater than or equal to A and less than or equal to B.

[0110] In the description of the embodiments of the present invention, the term "and / or" herein is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " herein generally indicates that the associated objects before and after are in an "or" relationship.

[0111] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A remote human resources performance management system for qualification review, characterized in that: The system comprises: A task requirement analysis module, wherein the task requirement analysis module is configured to obtain the qualification review requirements of the target qualification review project, and determine the task elements of the target qualification review project according to the qualification review requirements; A task generation and allocation module, wherein the task generation and allocation module is configured to establish a task list of a target qualification review project based on the task elements, generate a standard review duration for each task according to a difficulty score of each task in the task list, and allocate the tasks in the task list to the reviewers based on the standard review duration; An audit operation supervision module, wherein the audit operation supervision module is configured to collect the operational behavior characteristics of the auditors in the audit system, associate a timestamp with each operational behavior characteristic, and generate a performance score for each auditor based on the operational behavior characteristics; A comprehensive evaluation module, wherein the comprehensive evaluation module is configured to generate a comprehensive performance score according to the difficulty score and performance score of each auditor performing the audit task during the target period; The task generation and allocation module specifically includes: A task list establishing unit, wherein the task list establishing unit is configured to match a number of tasks corresponding to the target qualification review project in a comparison table of task elements and task types according to the task elements of the target qualification review project, and establish a task list; A task attribute labeling unit, wherein the task attribute labeling unit is configured to extract document attributes of the relevant review materials of each task in the task list, and label the document attributes as task attributes of the corresponding task; wherein the document attributes include data difficulty and task domain; A task allocation unit, the task allocation unit being configured to calculate a difficulty score of each task according to a task attribute of each task, generate a standard review duration for each task according to the difficulty score of each task, and allocate the tasks in the task list to the reviewers based on the standard review duration; The task allocation unit specifically includes: a task difficulty score calculation subunit, wherein the task difficulty score calculation subunit is configured to calculate a task difficulty score according to the data difficulty and the task domain of each task; The standard review time calculation subunit is configured to calculate the standard review time of each task according to the preset basic review time and the task difficulty score; wherein the expression of the standard review time is specifically: Among them, k is a constant, which represents the coefficient of the impact of task difficulty on the review time; A task allocation subunit, wherein the task allocation subunit is configured to calculate the audit time allocated to each auditor according to the work efficiency of each auditor and the standard audit time of each task, and evenly allocate tasks according to the audit time of each auditor; The expression of the audit time allocated to each auditor is specifically: ； Among them, the work efficiency of each auditor i is a parameter indicating its review speed; The audit operation supervision module specifically includes: An operation behavior feature collection unit, wherein the operation behavior feature collection unit is configured to collect the auditor's operation behavior features in the audit system to form an operation behavior feature record set; wherein the operation behavior features include one or a combination of mouse click position, click frequency, operation path, and operation time; A timestamp marking unit, the timestamp marking unit being configured to obtain a timestamp of each operation behavior feature and mark the timestamp to the corresponding operation behavior feature; A performance scoring unit, wherein the performance scoring unit is configured to use a deep learning model based on a long short-term memory network (LSTM) to perform learning training on the time series relationship and performance scoring of the operation behavior features marked with timestamps, and use the scoring model obtained after the training to obtain the performance score of each auditor when reviewing each task; Wherein, the audit operation supervision module also includes: Performance rating adjustment unit; Wherein, the performance score adjustment unit is configured to obtain the auditor's performance score influencing factors and generate the auditor's performance impact score using fuzzy reasoning; Among them, fuzzy reasoning is used to generate the expression of the auditor's performance impact score, specifically: ;in, For the raw performance score, represents fuzzy reasoning, represents a performance impact score, wherein the performance impact factor includes a factor representation of each task difficulty score; The comprehensive evaluation module specifically includes: Comprehensive performance rating unit; Wherein, the comprehensive performance scoring unit is configured to generate a comprehensive performance score of the auditor based on the performance impact score and the original performance score; The expression for generating the auditor's comprehensive performance score is: ; Where α is the weight coefficient of the performance impact score and the original performance score.

2. The remote human resources performance management system for qualification review according to claim 1 is characterized in that: The task requirement analysis module specifically includes: A qualification review requirement input unit, wherein the qualification review requirement input unit is configured to obtain the qualification review requirement of the target qualification review project input by the user; wherein the qualification review requirement is natural language information describing the target qualification review project; A task element extraction unit, wherein the task element extraction unit is configured to extract the task elements in the qualification review requirements by using keyword matching and / or semantic matching; wherein the task elements include the field to which the enterprise belongs, the qualification review type and the task characteristics.

3. The remote human resources performance management system for qualification review according to claim 1 is characterized in that: The expression for calculating the task difficulty score based on the data difficulty and task domain of each task is as follows: ； ； ； ； Among them, w1 is the data difficulty score The weight of w2 is the domain difficulty score. The weight of u1 is the complexity score The weight of u2 is the data volume score u3 is the data type score The weight, S i is the score of the i-th expert on the difficulty of the field, n is the number of experts, and F ij Score the domain features used to assess domain difficulty in the expert evaluation model, v j is the weight of the domain feature score Fij, and m is the number of domain features.

4. The remote human resources performance management system for qualification review according to claim 3 is characterized in that: Complexity score , Data volume scoring and data type scoring The expression is: ； ； ； Among them: Complexity score , Data volume scoring and data type scoring The score is calculated by natural language processing tools.

5. A remote human resources performance management method for qualification review, characterized in that: A remote human resources performance management system for qualification review as claimed in any one of claims 1 to 4, the method comprising: S1: Obtain the qualification review requirements of the target qualification review project, and determine the task elements of the target qualification review project based on the qualification review requirements; S2: Based on the task elements, a task list of the target qualification review project is established, a standard review duration is generated for each task according to the difficulty score of each task in the task list, and the tasks in the task list are assigned to the reviewers based on the standard review duration; S3: Collect the auditor's operational behavior characteristics in the audit system, associate a timestamp with each operational behavior characteristic, and generate a performance score for each auditor based on the operational behavior characteristics; S4: Generate a comprehensive performance score based on the difficulty score and performance score of each auditor performing the audit tasks during the target period.

Images