Human resource informatization management system and method based on cloud platform

Abstract

The invention relates to the field of intelligent data analysis, and provides a human resource informatization management system and method based on a cloud platform, and the system employs a semantic analysis technology based on a natural language, carries out the semantic analysis of a project completion condition and a work quality condition of a to-be-analyzed employee, so as to directly know the work performance of the employee. Meanwhile, semantic analysis is performed on customer evaluation data to assist in understanding the work performance of the employees, and finally the performance levels of the employees are divided based on the work performance of the employees. In this way, the performance levels of the employees can be automatically divided, and automation and intelligentization of the evaluation process are achieved.

Description

Technical Field

[0001] The present application relates to the field of intelligent data analysis, and more specifically, to a human resources information management system and method based on a cloud platform. Background Art

[0002] Human resource information management is to achieve all-round management of human resources through a series of operations such as integration, storage, processing and analysis of various data in the human resource management process, with the help of information technology. It covers a wide range of management work, including personnel recruitment, employee training, performance appraisal, salary management, employee benefits and labor relations.

[0003] In today's reality, with the continuous expansion of enterprise scale and the increasing complexity of business, human resource management faces many challenges. The traditional human resource management model has revealed its shortcomings in many aspects. Take employee performance evaluation as an example. Under the traditional model, it often relies too much on subjective evaluation and regular performance interviews. This method has many disadvantages: on the one hand, subjective bias can easily lead to inaccurate and unfair performance evaluation results. Different evaluators may give very different evaluations to the same employee due to personal preferences, experience differences and other factors, which will not only affect the employee's work enthusiasm, but also may prevent truly excellent employees from receiving the recognition and rewards they deserve. On the other hand, the traditional performance evaluation method consumes a lot of time and resources. From organizing performance interviews, collecting evaluations from all parties to summarizing and analyzing the results, the entire process requires a lot of manpower and material resources, and is inefficient, which is difficult to meet the needs of the rapid development of modern enterprises.

[0004] Therefore, an optimized human resources information management solution is needed. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present application provides a human resources information management system and method based on a cloud platform.

[0006] According to one aspect of the present application, a human resources information management system based on a cloud platform is provided, which includes:

[0007] Employee performance-related data collection module, used to obtain project completion status, work quality and customer evaluation data of employees to be analyzed;

[0008] An employee performance-related data encoding module is used to encode the project completion status, work quality status and customer evaluation data of the employee to be analyzed to obtain an employee effectiveness global semantic feature vector and an employee performance auxiliary reference semantic feature vector;

[0009] An employee effectiveness global semantic feature optimization module, used for optimizing the employee effectiveness global semantic feature vector based on the employee performance auxiliary reference semantic feature vector by performing feature hierarchical expression optimization based on detail distribution difference position change to obtain an optimized employee effectiveness global semantic feature vector;

[0010] An employee performance semantic feature generation module, used for fusing the optimized employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector to obtain an employee performance semantic feature vector;

[0011] The employee performance grade result generating module is used to obtain the performance grade result of the employee to be analyzed based on the employee performance semantic feature vector.

[0012] According to another aspect of the present application, a human resources information management method based on a cloud platform is provided, which includes:

[0013] Obtain project completion status, work quality and customer evaluation data of the employees to be analyzed;

[0014] Data encoding is performed on the project completion status, work quality status and customer evaluation data of the employee to be analyzed to obtain a global semantic feature vector of employee effectiveness and an auxiliary reference semantic feature vector of employee performance;

[0015] Based on the employee performance auxiliary reference semantic feature vector, the employee effectiveness global semantic feature vector is optimized for feature hierarchical expression based on detail distribution difference position change to obtain an optimized employee effectiveness global semantic feature vector;

[0016] fusing the optimized employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector to obtain an employee performance semantic feature vector;

[0017] Based on the employee performance semantic feature vector, a performance grade result of the employee to be analyzed is obtained.

[0018] This application has significant technical effects due to the adoption of the above technical solutions:

[0019] The cloud-based human resources information management system and method provided by the present application adopts natural language-based semantic analysis technology, directly understands the employee's work performance by performing semantic analysis on the project completion and work quality of the employee to be analyzed, and performs semantic analysis on customer evaluation data to assist in understanding the employee's work performance, and finally divides the employee's performance level based on the employee's work performance. In this way, the employee's performance level can be automatically divided, realizing the automation and intelligence of the evaluation process. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0021] Figure 1 The present invention is a flowchart of a human resources information management method based on a cloud platform according to an embodiment of the present application.

[0022] Figure 2 In the human resources information management method based on a cloud platform according to an embodiment of the present application, the project completion status, work quality status and the customer evaluation data of the employee to be analyzed are respectively encoded to obtain a flowchart of the employee's global semantic feature vector of effectiveness and the employee's performance auxiliary reference semantic feature vector.

[0023] Figure 3 The present invention is a flowchart of extracting semantic features of the project completion status and work quality of the employee to be analyzed in the human resources information management method based on the cloud platform according to the embodiment of the present application to obtain the global semantic feature vector of the employee's effectiveness.

[0024] Figure 4 The present invention is a flowchart of performing semantic feature extraction on the customer evaluation data to obtain the auxiliary reference semantic feature vector of the employee performance in the human resources information management method based on the cloud platform according to an embodiment of the present application.

[0025] Figure 5 This is a system block diagram of a human resources information management system based on a cloud platform according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0027] Human resource information management refers to the integration, storage, processing and analysis of various data in the human resource management process, and the comprehensive management of human resources through information technology. It includes management work in personnel recruitment, employee training, performance appraisal, salary management, employee benefits and labor relations. In traditional human resource management, employee performance evaluation often relies on subjective evaluation and regular performance interviews, which can easily lead to inaccurate and unfair performance evaluation due to subjective bias, and this method consumes a lot of time and resources. Therefore, an optimized human resource information management solution is expected.

[0028] In recent years, deep learning and neural networks have been widely used in computer vision, natural language processing, text signal processing and other fields. In addition, deep learning and neural networks have also demonstrated a level close to or even beyond that of humans in image classification, object detection, semantic segmentation, text translation and other fields. The development of deep learning and neural networks has provided new solutions and solutions for the information management of human resources.

[0029] Figure 1 FIG. 1 is a flow chart of a human resources information management method based on a cloud platform according to an embodiment of the present application. Figure 1 As shown, according to the cloud platform-based human resources information management method of the embodiment of the present application, the method includes: S110, obtaining the project completion status, work quality status and customer evaluation data of the employee to be analyzed; S120, respectively encoding the project completion status, work quality status and the customer evaluation data of the employee to be analyzed to obtain a global semantic feature vector of employee effectiveness and an auxiliary reference semantic feature vector of employee performance; S130, based on the auxiliary reference semantic feature vector of employee performance, performing feature hierarchical expression optimization on the global semantic feature vector of employee effectiveness based on the position change of detail distribution differences to obtain an optimized global semantic feature vector of employee effectiveness; S140, fusing the optimized global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance to obtain an employee performance semantic feature vector; S150, based on the employee performance semantic feature vector, obtaining the performance grade result of the employee to be analyzed.

[0030] In step S110, the project completion, work quality and customer evaluation data of the employee to be analyzed are obtained. It should be understood that the project completion and work quality of the employee to be analyzed respectively record the specific tasks assigned to the employee in the different projects in which the employee participates, the task completion, and the employee's work efficiency, work attitude and other information. The project completion directly reflects the work results of the employee and is one of the important indicators for evaluating employee performance. The successful completion of the project, high-quality task execution, effective problem solving and significant results display will have a positive impact on the performance results of the employee, and vice versa, it may lead to a decline in performance. The work quality situation comprehensively reflects the work ability and professional quality of the employee and is an important basis for evaluating the performance of the employee. High accuracy, high efficiency and positive work attitude will add points to the performance results of the employee, while low-quality work performance may lead to a decline in performance. The customer's evaluation data includes customer satisfaction with the service provided by the employee, and customer evaluation of the employee's professional ability and communication ability. Customer evaluation data is an indispensable part of evaluating employee performance, which directly reflects the performance of the employee in the external environment and the degree of customer recognition of his work. Highly satisfied customer evaluation, professional service ability, good communication performance and positive customer feedback will have a positive impact on the performance results of the employee. In particular, in a specific implementable embodiment of the present application, the project completion status, work quality status and customer evaluation data of the employees to be analyzed can be obtained respectively through a project management system, a work quality evaluation form and a customer feedback questionnaire.

[0031] In step S120, the project completion status, work quality status and customer evaluation data of the employee to be analyzed are respectively encoded to obtain the employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector. Specifically, Figure 2 The present invention is a flowchart of a human resource information management method based on a cloud platform according to an embodiment of the present application, in which the project completion status, work quality status and customer evaluation data of the employee to be analyzed are respectively encoded to obtain a global semantic feature vector of employee effectiveness and an auxiliary reference semantic feature vector of employee performance. Figure 2 As shown, the project completion status, work quality status and customer evaluation data of the employee to be analyzed are respectively data encoded to obtain a global semantic feature vector of employee effectiveness and an auxiliary reference semantic feature vector of employee performance, including: S121, performing semantic feature extraction on the project completion status and work quality status of the employee to be analyzed to obtain the global semantic feature vector of employee effectiveness; S122, performing semantic feature extraction on the customer evaluation data to obtain the auxiliary reference semantic feature vector of employee performance.

[0032] More specifically, Figure 3The present invention is a flowchart of extracting semantic features of the project completion and work quality of the employee to be analyzed to obtain the global semantic feature vector of the employee's effectiveness in the human resources information management method based on the cloud platform according to the embodiment of the present application. Figure 3 As shown, semantic features are extracted from the project completion status and work quality status of the employee to be analyzed to obtain the global semantic feature vector of the employee effectiveness, including: S1211, preprocessing the project completion status and work quality status of the employee to be analyzed to obtain preprocessed employee effectiveness data; S1212, segmenting the preprocessed employee effectiveness data and passing it through an employee effectiveness semantic encoder to obtain a plurality of employee effectiveness-related word semantic feature vectors; S1213, arranging the plurality of employee effectiveness-related word semantic feature vectors into an employee effectiveness-related word semantic input matrix and passing it through an employee work performance understander based on a global transformation matrix to obtain the employee effectiveness global semantic feature vector.

[0033] The original data on the project completion and work quality of the employees to be analyzed often contain irrelevant information. These noise data will interfere with the accuracy of the employee performance analysis results. In addition, there may be inconsistent data formats in the original data. In order to deal with the noise data and inconsistent data formats, the project completion and work quality of the employees to be analyzed need to be preprocessed to obtain the preprocessed employee effectiveness data. The preprocessed employee effectiveness data can more accurately reflect the actual performance and work results of the employees, which provides a more scientific and accurate basis for subsequent performance evaluation.

[0034] The pre-processed employee effectiveness data is a continuous Chinese text. Since the text sentences have no obvious boundaries, it is difficult to directly perform semantic analysis on the text. In order to extract useful semantic information from the pre-processed employee effectiveness data, the pre-processed employee effectiveness data needs to be segmented first. Word segmentation is the process of dividing a continuous text into a series of independent words or phrases. After word segmentation, each word or phrase represents a specific semantic unit, which enables subsequent analysis to focus more accurately on key information, such as specific aspects of project completion, specific indicators of work quality, etc. Then, the pre-processed employee effectiveness data after word segmentation is passed through an employee effectiveness semantic encoder to obtain multiple semantic feature vectors of employee effectiveness-related words. The employee effectiveness semantic encoder can extract the semantic information of the employee effectiveness data after word segmentation. The semantic feature vector extracted by the semantic encoder can convert the employee effectiveness data that was originally difficult to quantify into a quantifiable data form, which makes the evaluation of employee performance more objective and scientific. The employee effectiveness semantic encoder described in the present application is a converter-based semantic encoder that includes a word embedding layer. The word embedding layer can map words into a continuous vector space, so that semantically similar words are close in distance in the vector space. This representation method helps capture the semantic relationship between words, thereby improving the understanding ability of the semantic encoder. The semantic encoder based on the transformer has a strong ability to process long-distance dependencies. People of ordinary skill in the art should understand that the Transformer model uses a self-attention mechanism, which allows the model to calculate the attention weight of each position in the sequence in parallel when processing input sequence data. This parallel calculation method greatly improves the training speed and reasoning efficiency of the model, allowing the employee effectiveness semantic encoder to quickly respond to the processing needs of large amounts of data.

[0035] Considering that there is a semantic association between the semantic feature vectors of different employee effectiveness-related words, this semantic association is crucial for understanding employee work performance. In order to maintain the semantic association between different vectors, the multiple employee effectiveness-related word semantic feature vectors are first arranged into an employee effectiveness-related word semantic input matrix. By arranging the semantic feature vectors into a matrix, it can be ensured that the input data meets the requirements of the model, so that subsequent calculations and analyses can be carried out smoothly. Then, the arranged employee effectiveness-related word semantic input matrix is ​​passed through an employee work performance understander based on a global transformation matrix to obtain an employee effectiveness global semantic feature vector. The employee work performance understander based on a global transformation matrix described in this application is a convolutional neural network model that uses a global transformation matrix as a filter. It should be understood that the global transformation matrix can capture the global relationship between all elements in the input matrix, and this global information helps the employee performance classification model to better grasp employee performance. And through the role of the global transformation matrix, the model can fuse and abstract the features in the input matrix. This fusion and abstraction process helps to extract more representative employee effectiveness features, thereby improving the ability to understand employee work performance.

[0036] More specifically, Figure 4 The present invention is a flowchart of extracting semantic features from the customer evaluation data to obtain the employee performance auxiliary reference semantic feature vector in the human resource information management method based on the cloud platform according to the embodiment of the present application. Figure 4 As shown, semantic feature extraction is performed on the customer evaluation data to obtain the auxiliary reference semantic feature vector of employee performance, including: S1221, preprocessing the customer evaluation data to obtain preprocessed customer evaluation data; S1222, segmenting the preprocessed customer evaluation data and passing it through a customer evaluation semantic encoder to obtain the auxiliary reference semantic feature vector of employee performance.

[0037] Customer evaluation data mainly comes from customer feedback questionnaire surveys. In order to extract relevant important evaluation data from the original customer feedback questionnaire survey, it is necessary to preprocess the customer evaluation data to obtain preprocessed customer evaluation data. Through preprocessing, some redundant information that is not important to the employee performance classification can be removed, and the original customer evaluation data can be converted into a more concise expression form, which is conducive to improving the efficiency of subsequent customer evaluation data analysis.

[0038] In order to facilitate the semantic analysis of subsequent customer evaluation data, the pre-processed customer evaluation data needs to be segmented first. Word segmentation helps to decompose the pre-processed customer evaluation data into smaller and more meaningful units, avoiding misunderstandings or confusions that may occur when the entire pre-processed customer evaluation data is analyzed as a single unit. Then, the pre-processed customer evaluation data after word segmentation is passed through the customer evaluation semantic encoder to obtain an auxiliary reference semantic feature vector for employee performance. Through the processing of the customer evaluation semantic encoder, features related to employee service quality and service attitude can be extracted. The customer evaluation semantic encoder described in this application is a bidirectional long short-term memory neural network model including a word embedding layer. The word embedding layer can convert the segmented vocabulary into a dense vector representation. The bidirectional long short-term memory neural network can simultaneously capture the forward and reverse information in the input sequence, so as to more comprehensively understand the contextual information of the pre-processed customer evaluation data, which helps to more accurately capture the semantic information and potential logic in the customer evaluation, thereby extracting features that are more relevant to employee performance.

[0039] In step S130, the employee effectiveness global semantic feature vector is optimized based on the feature hierarchical expression of the position change of the detail distribution difference based on the employee performance auxiliary reference semantic feature vector to obtain an optimized employee effectiveness global semantic feature vector. Accordingly, considering that the employee's project completion, work quality and customer evaluation data may come from different data sources, the quality of these data sources is different, resulting in differences in the initial distribution of the data, and the employee effectiveness semantic encoder and the customer evaluation semantic encoder process the data based on different models to extract features of different granularities. In addition, customer evaluation data usually contains subjective opinions, while the employee's project completion and work quality tend to be objective quantitative indicators. This difference between the subjective and objective will lead to inconsistent feature distribution. Due to the fine-grained inconsistency of feature distribution between the employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector, this will cause the fused feature vector to have abnormal stacking on the boundary of the feature manifold. Based on this, in the technical solution of the present application, before fusing the global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance to obtain the semantic feature vector of employee performance, the global semantic feature vector of employee effectiveness is optimized for feature hierarchical expression based on the position changes of detail distribution differences based on the auxiliary reference semantic feature vector of employee performance.

[0040] Specifically, in an embodiment of the present application, based on the employee performance auxiliary reference semantic feature vector, the employee effectiveness global semantic feature vector is optimized for feature hierarchical expression based on detail distribution difference position change to obtain an optimized employee effectiveness global semantic feature vector, including: calculating the position difference between the employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector to obtain an employee performance shift position difference representation vector; after cascading the employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector into an employee performance joint feature vector, the employee performance joint feature vector is input into a one-dimensional convolutional layer to obtain an employee performance shift convolutional coding representation vector; after linear embedding modulation encoding is performed on the employee effectiveness global semantic feature vector using a first embedding matrix into an employee effectiveness global semantic feature linear modulation vector, Calculate the positional difference between the employee effectiveness global semantic feature linear modulation vector and the employee performance auxiliary reference semantic feature vector to obtain an employee performance shift modulation difference representation vector; use the employee performance shifted positional difference representation vector as the value vector, the employee performance shifted convolutional coding representation vector and the employee performance shift modulation difference representation vector as the query vector and key vector, input the employee performance shifted positional difference representation vector, the employee performance shifted convolutional coding representation vector and the employee performance shift modulation difference representation vector into a shift corrector based on an analog converter structure to obtain an employee performance multi-scale hierarchical fusion shift modulation weight vector; calculate the positional point multiplication between the employee performance multi-scale hierarchical fusion shift modulation weight vector and the employee effectiveness global semantic feature vector to obtain the optimized employee effectiveness global semantic feature vector.

[0041] In an embodiment of the present application, the employee performance auxiliary reference semantic feature vector is optimized based on the feature hierarchical expression of the employee effectiveness global semantic feature vector based on the detail distribution difference position change to obtain an optimized employee effectiveness global semantic feature vector, including: processing the employee performance auxiliary reference semantic feature vector with the employee effectiveness global semantic feature vector using the following formula to obtain an optimized employee effectiveness global semantic feature vector; wherein the formula is:

[0042]

[0043] V 2d =Conv 1D ([V1; V2])

[0044]

[0045] V1'=V1⊙V d

[0046] Wherein, V1 represents the global semantic feature vector of employee effectiveness, V2 represents the auxiliary reference semantic feature vector of employee performance, Indicates point-by-point subtraction, V 1d represents the employee performance shift by position difference representation vector, [;] represents the cascade operation, Conv1D(·) represents one-dimensional convolutional coding, V 2d represents the employee performance shift convolutional coding representation vector, W1 represents the first embedding matrix, b1 represents the first modulation bias vector, V 3d represents the employee performance shift modulation differential representation vector, represents vector multiplication, Softmax represents the normalized exponential function, T represents the transposed vector of the feature vector, d represents the length of the feature vector, Relu represents the activation function, V d represents the multi-scale hierarchical fusion shift modulation weight vector of employee performance, ⊙ represents the position point multiplication, and V1' represents the optimized global semantic feature vector of employee effectiveness.

[0047] In the technical solution of the present application, in the process of fusing the global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance, a feature hierarchical expression optimization method based on the position change of detail distribution difference is proposed to address the problem of fine-grained inconsistency in feature distribution that arises in the process of fusing the global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance. The method aims to reduce the abnormal stacking phenomenon of the fused feature vector on the boundary of the feature manifold, so as to improve the robustness and classification performance of the model.

[0048] Specifically, the method first calculates the element-by-element difference between the global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance to obtain the position difference representation vector of employee performance displacement, thereby capturing the difference between the two feature vectors. Then, the global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance are cascaded into a joint feature vector and input into a one-dimensional convolution layer to generate an employee performance displacement convolution coding representation vector. That is, the deep-level displacement implicit information between the global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance is captured by one-dimensional convolution coding. At the same time, the global semantic feature vector of employee effectiveness is linearly embedded and modulated to be a linear modulation vector of employee effectiveness global semantic features using a first embedding matrix, and the element-by-element difference between the linear modulation vector of employee effectiveness global semantic features and the auxiliary reference semantic feature vector of employee performance is calculated to obtain an employee performance displacement modulation difference representation vector. The employee performance displacement modulation difference representation vector is used to represent the displacement information position expression of the global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance in a linear space. That is, in the technical solution of the present application, the employee performance shift position differential representation vector, the employee performance shift convolutional coding representation vector and the employee performance shift modulation differential representation vector respectively represent the explicit expression of the fine-grained distribution difference shift information between the employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector in different feature spaces.

[0049] Next, the employee performance shifted positional differential representation vector is used as the value vector, the employee performance shifted convolutional coding representation vector and the employee performance shifted modulation differential representation vector are used as the query vector and the key vector respectively, and these vectors are input into the shift corrector based on the simulated transformer structure to generate the employee performance multi-scale hierarchical fusion shift modulation weight vector. In this way, the importance of features at different scales is evaluated by the attention mechanism, that is, the attention mechanism is used to simulate the hierarchical precipitation expression of the fine-grained difference features between the employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector in different feature spaces. Finally, the element-by-element product of the employee performance multi-scale hierarchical fusion shift modulation weight vector and the employee effectiveness global semantic feature vector is calculated to obtain the optimized employee effectiveness global semantic feature vector, that is, the weighted operation is used to assign weights to each element in the employee effectiveness global semantic feature vector, which reduces the influence of local disturbances and improves the quality of the feature vector.

[0050] In summary, in the method, the employee performance shift position difference representation vector, the employee performance shift convolutional coding representation vector and the employee performance shift modulation difference representation vector respectively represent the explicit expression of the fine-grained distribution difference shift information between the employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector in different feature spaces. The generation of the employee performance multi-scale hierarchical fusion shift modulation weight vector utilizes the attention mechanism to evaluate the importance of features at different scales and provide weight information for feature shift correction. Ultimately, the generation of the optimized employee effectiveness global semantic feature vector reduces the impact of disturbances, improves the model's robustness to local disturbances in the input data, enhances the model's adaptability to noisy or variant data sets, improves the fusion quality of feature vectors, and improves the overall classification performance and training efficiency of the model.

[0051] In step S140, the optimized global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance are fused to obtain an employee performance semantic feature vector. It should be understood that the optimized global semantic feature vector of employee effectiveness usually reflects the overall work effectiveness of employees in multiple aspects and over a long period of time, including but not limited to work efficiency, work quality, etc., and it provides a macroscopic perspective of the comprehensive performance of employees in the organization; while the auxiliary reference semantic feature vector of employee performance focuses on the customer's evaluation of the specific work performance of employees, especially the service quality, service attitude, problem-solving ability, etc. perceived from the customer's perspective, and it provides a microscopic perspective of employee work performance and direct information of customer feedback. In order to make up for the shortcomings of a single feature vector in performance evaluation, the technical solution of the present application needs to fuse the optimized global semantic feature vector of employee effectiveness and the auxiliary reference semantic feature vector of employee performance to obtain an employee performance semantic feature vector. It provides a microscopic perspective of employee work performance and direct information of customer feedback. The fused semantic feature vector of employee performance contains richer and more comprehensive information, which helps to classify performance more accurately.

[0052] In step S150, based on the employee performance semantic feature vector, the performance grade result of the employee to be analyzed is obtained. Specifically, in an embodiment of the present application, based on the employee performance semantic feature vector, the performance grade result of the employee to be analyzed is obtained, including: passing the employee performance semantic feature vector through a classifier-based employee performance grade generator to obtain a classification result, and the classification result is the performance grade label of the employee to be analyzed. It should be understood that the classifier will be trained using the project completion status, work quality status and customer evaluation data of the labeled employees. During the training process, the classifier will learn how to predict the performance grade based on the performance characteristics of the relevant employees in the input data. According to the output of the classifier, the performance grade label of the employee to be analyzed can be determined. The obtained performance grade label can provide important decision support for organizational managers, and it can also enable managers to formulate corresponding incentive measures based on performance grades to improve the enthusiasm of employees at work.

[0053] In summary, the human resource information management method based on the cloud platform of the embodiment of the present application is explained, which adopts the semantic analysis technology based on natural language, directly understands the work performance of employees by performing semantic analysis on the project completion and work quality of the employees to be analyzed, and assists in understanding the work performance of employees by performing semantic analysis on customer evaluation data, and finally divides the performance level of employees based on their work performance. In this way, the performance level of employees can be automatically divided, realizing the automation and intelligence of the evaluation process.

[0054] Figure 5 : is a system block diagram of a human resources information management system based on a cloud platform according to an embodiment of the present application. Figure 5As shown, according to the embodiment of the present application, the cloud platform-based human resources information management system 100 includes: an employee performance-related data collection module 110, which is used to obtain the project completion status, work quality status and customer evaluation data of the employee to be analyzed; an employee performance-related data encoding module 120, which is used to respectively encode the project completion status, work quality status and the customer evaluation data of the employee to be analyzed to obtain an employee effectiveness global semantic feature vector and an employee performance auxiliary reference semantic feature vector; an employee effectiveness global semantic feature optimization module 130, which is used to optimize the feature hierarchical expression of the employee effectiveness global semantic feature vector based on the detail distribution difference position change based on the employee performance auxiliary reference semantic feature vector to obtain an optimized employee effectiveness global semantic feature vector; an employee performance semantic feature generation module 140, which is used to fuse the optimized employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector to obtain an employee performance semantic feature vector; an employee performance grade result generation module 150, which is used to obtain the performance grade result of the employee to be analyzed based on the employee performance semantic feature vector.

[0055] Here, those skilled in the art can understand that the specific functions and operations of each unit and module in the above-mentioned cloud platform-based human resources information management system 100 have been referred to above. Figures 1 to 4 The description of the human resources information management method based on the cloud platform has been introduced in detail, and therefore, its repeated description will be omitted.

[0056] In summary, the cloud platform-based human resource information management system 100 according to the embodiment of the present application is explained, which adopts the semantic analysis technology based on natural language, directly understands the work performance of employees by performing semantic analysis on the project completion and work quality of the employees to be analyzed, and assists in understanding the work performance of employees by performing semantic analysis on customer evaluation data, and finally divides the performance level of employees based on their work performance. In this way, the performance level of employees can be automatically divided, realizing the automation and intelligence of the evaluation process.

[0057] As described above, the human resource information management system 100 based on the cloud platform according to the embodiment of the present application can be implemented in various wireless terminals, such as a server for human resource information management based on the cloud platform. In one example, the human resource information management system 100 based on the cloud platform according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the human resource information management system 100 based on the cloud platform can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the human resource information management system 100 based on the cloud platform can also be one of the many hardware modules of the wireless terminal.

[0058] Alternatively, in another example, the cloud platform-based human resources information management system 100 and the wireless terminal may also be separate devices, and the cloud platform-based human resources information management system 100 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

Claims

1. A human resources information management method based on a cloud platform, characterized in that: include: Obtain project completion status, work quality and customer evaluation data of the employees to be analyzed; Data encoding is performed on the project completion status, work quality status and customer evaluation data of the employee to be analyzed to obtain a global semantic feature vector of employee effectiveness and an auxiliary reference semantic feature vector of employee performance; Based on the employee performance auxiliary reference semantic feature vector, the employee effectiveness global semantic feature vector is optimized for feature hierarchical expression based on detail distribution difference position change to obtain an optimized employee effectiveness global semantic feature vector; fusing the optimized employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector to obtain an employee performance semantic feature vector; Based on the employee performance semantic feature vector, a performance grade result of the employee to be analyzed is obtained.

2. The human resources information management method based on the cloud platform according to claim 1 is characterized in that: The project completion status, work quality status and customer evaluation data of the employee to be analyzed are respectively encoded to obtain a global semantic feature vector of employee effectiveness and an auxiliary reference semantic feature vector of employee performance, including: Extracting semantic features of the project completion and work quality of the employee to be analyzed to obtain a global semantic feature vector of the employee's effectiveness; Semantic feature extraction is performed on the customer evaluation data to obtain the employee performance auxiliary reference semantic feature vector.

3. The human resources information management method based on cloud platform according to claim 2 is characterized in that: The semantic features of the project completion and work quality of the employee to be analyzed are extracted to obtain the global semantic feature vector of the employee's effectiveness, including: Preprocessing the project completion status and work quality status of the employee to be analyzed to obtain preprocessed employee efficiency data; The pre-processed employee effectiveness data is segmented and passed through an employee effectiveness semantic encoder to obtain a plurality of employee effectiveness related word semantic feature vectors; The employee effectiveness global semantic feature vector is obtained by arranging the plurality of employee effectiveness related word semantic feature vectors into an employee effectiveness related word semantic input matrix and then passing the matrix through an employee work performance understander based on a global conversion matrix.

4. The human resources information management method based on cloud platform according to claim 3 is characterized in that: Extracting semantic features from the customer evaluation data to obtain the employee performance auxiliary reference semantic feature vector includes: Preprocessing the customer evaluation data to obtain preprocessed customer evaluation data; The pre-processed customer evaluation data is segmented and passed through a customer evaluation semantic encoder to obtain the employee performance auxiliary reference semantic feature vector.

5. The human resources information management method based on cloud platform according to claim 4 is characterized in that: The employee effectiveness semantic encoder is a converter-based semantic encoder including a word embedding layer, the employee work performance understander based on a global transformation matrix is ​​a convolutional neural network model that uses a global transformation matrix as a filter, and the customer evaluation semantic encoder is a bidirectional long short-term memory neural network model including a word embedding layer.

6. The human resources information management method based on cloud platform according to claim 5 is characterized in that: Based on the employee performance auxiliary reference semantic feature vector, the employee effectiveness global semantic feature vector is optimized for feature hierarchical expression based on detail distribution difference position change to obtain an optimized employee effectiveness global semantic feature vector, including: Calculating the positional difference between the employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector to obtain an employee performance shift positional difference representation vector; After concatenating the employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector into an employee performance joint feature vector, the employee performance joint feature vector is input into a one-dimensional convolutional layer to obtain an employee performance shift convolutional coding representation vector; After performing linear embedding modulation encoding on the employee effectiveness global semantic feature vector using the first embedding matrix to obtain an employee effectiveness global semantic feature linear modulation vector, a positional difference between the employee effectiveness global semantic feature linear modulation vector and the employee performance auxiliary reference semantic feature vector is calculated to obtain an employee performance shift modulation difference representation vector; Taking the employee performance shifted position difference representation vector as the value vector, the employee performance shifted convolutional coding representation vector and the employee performance shifted modulation difference representation vector as the query vector and key vector, the employee performance shifted position difference representation vector, the employee performance shifted convolutional coding representation vector and the employee performance shifted modulation difference representation vector are input into a shift corrector based on an imitation transformer structure to obtain a multi-scale hierarchical fusion shift modulation weight vector of employee performance; The optimized employee effectiveness global semantic feature vector is obtained by calculating the position point multiplication between the employee performance multi-scale hierarchical fusion shift modulation weight vector and the employee effectiveness global semantic feature vector.

7. The human resources information management method based on cloud platform according to claim 6 is characterized in that: Based on the employee performance semantic feature vector, a performance grade result of the employee to be analyzed is obtained, including: passing the employee performance semantic feature vector through a classifier-based employee performance grade generator to obtain a classification result, wherein the classification result is a performance grade label of the employee to be analyzed.

8. A human resources information management system based on a cloud platform, characterized in that: include: Employee performance-related data collection module, used to obtain project completion status, work quality and customer evaluation data of employees to be analyzed; An employee performance-related data encoding module is used to encode the project completion status, work quality status and customer evaluation data of the employee to be analyzed to obtain an employee effectiveness global semantic feature vector and an employee performance auxiliary reference semantic feature vector; An employee effectiveness global semantic feature optimization module, used for optimizing the employee effectiveness global semantic feature vector based on the employee performance auxiliary reference semantic feature vector by performing feature hierarchical expression optimization based on detail distribution difference position change to obtain an optimized employee effectiveness global semantic feature vector; An employee performance semantic feature generation module, used for fusing the optimized employee effectiveness global semantic feature vector and the employee performance auxiliary reference semantic feature vector to obtain an employee performance semantic feature vector; The employee performance grade result generating module is used to obtain the performance grade result of the employee to be analyzed based on the employee performance semantic feature vector.

9. The human resources information management system based on the cloud platform according to claim 8 is characterized in that: The employee performance related data coding module includes: An employee effectiveness global semantic feature generation unit, used for extracting semantic features of the project completion status and work quality status of the employee to be analyzed to obtain the employee effectiveness global semantic feature vector; The employee performance auxiliary reference semantic feature generation unit is used to extract semantic features from the customer evaluation data to obtain the employee performance auxiliary reference semantic feature vector.

10. The cloud platform-based human resources information management system according to claim 9, characterized in that: The employee performance grade result generating module is used to: pass the employee performance semantic feature vector through a classifier-based employee performance grade generator to obtain a classification result, and the classification result is the performance grade label of the employee to be analyzed.

Images