Human efficiency evaluation model training method and device

Abstract

The invention discloses a human effect evaluation model training method and device, and relates to the technical field of computers. A specific embodiment of the method comprises the steps of obtaining production system data, historical performance data and historical evaluation data; determining work content feature data based on the production system data; based on the work content feature data, the historical performance data and the historical evaluation data, training to obtain a human effect evaluation model; the richness of human efficiency evaluation dimensions can be increased, and the fairness and accuracy of human efficiency evaluation results are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for training a human efficiency evaluation model. Background Art

[0002] Human efficiency evaluation refers to a systematic evaluation of personnel's work efficiency. In the existing technology, the evaluation dimensions used in human efficiency evaluation are relatively single, and most of the data required for the evaluation are data from manual offline interactions. Online data only accounts for a small part of the required data, and the quality of online data is poor, which can easily lead to inaccurate evaluation data and affect the accuracy of human efficiency evaluation.

[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the related art:

[0004] The evaluation dimensions used in human efficiency evaluation are relatively single, resulting in a lack of fairness in the evaluation; the data required for the evaluation is inaccurate, and the accuracy of the human efficiency evaluation results is poor. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a method and device for training a human efficiency evaluation model, which can increase the richness of human efficiency evaluation dimensions and improve the fairness and accuracy of human efficiency evaluation results.

[0006] To achieve the above object, according to one aspect of an embodiment of the present invention, a method for training a human efficiency evaluation model is provided, comprising:

[0007] Obtain production system data, historical performance data, and historical evaluation data;

[0008] Determining work content characteristic data based on the production system data;

[0009] Based on the work content feature data, historical performance data and historical evaluation data, a human efficiency evaluation model is trained.

[0010] Further, based on the production system data, work content characteristic data is determined, including:

[0011] Extracting basic feature data from the production system data; wherein the basic feature data includes item features, delivery area features, operation features, and service type features;

[0012] The basic feature data is subjected to feature clustering to obtain the work content feature data; wherein the work content feature data includes work quality, work attitude and work environment.

[0013] Furthermore, based on the work content feature data, historical performance data and historical evaluation data, a human efficiency evaluation model is trained, including:

[0014] A training sample set is constructed using the work content feature data, historical performance data and historical evaluation data. The work content feature data, historical performance data and historical evaluation data are used as input, and a pre-set classification result is used as a target classification result to train a human efficiency evaluation model.

[0015] Furthermore, after the human efficiency evaluation model is obtained through training, the method further includes:

[0016] Respond to requests for human efficiency evaluation and obtain information to be evaluated;

[0017] Inputting the information to be evaluated into a human efficiency evaluation model for evaluation to obtain a prediction result of the information to be evaluated;

[0018] The prediction result is pushed to the investigation and review platform for verification, and the verification result fed back by the investigation and review platform is received, and the final evaluation result of the information to be evaluated is determined based on the verification result.

[0019] Furthermore, after determining the final evaluation result of the information to be evaluated based on the verification result, the method further includes:

[0020] Determine target users corresponding to final evaluation results below a preset evaluation threshold;

[0021] Tasks are assigned to the target users based on a preset incentive strategy.

[0022] Furthermore, after determining the final evaluation result of the information to be evaluated based on the verification result, the method further includes:

[0023] The information to be evaluated and the final evaluation result of the information to be evaluated are input into the human efficiency evaluation model to perform iterative optimization training on the human efficiency evaluation model.

[0024] According to a second aspect of an embodiment of the present invention, a human efficiency evaluation model training device is provided, comprising:

[0025] Data acquisition module, which acquires production system data, historical performance data, and historical evaluation data;

[0026] A feature extraction module, based on the production system data, determines work content feature data;

[0027] The model training module trains a human efficiency evaluation model based on the work content feature data, historical performance data and historical evaluation data.

[0028] According to a third aspect of an embodiment of the present invention, there is provided an electronic device, including:

[0029] one or more processors;

[0030] a storage device for storing one or more programs,

[0031] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any of the above embodiments.

[0032] According to a fourth aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method described in any one of the above embodiments is implemented.

[0033] According to a fifth aspect of an embodiment of the present invention, there is provided a computer program product, including a computer program, which implements the method of any of the above embodiments when the computer program is executed by a processor.

[0034] An embodiment of the above invention has the following advantages or beneficial effects: by acquiring production system data, historical performance data and historical evaluation data; based on the production system data, determining the work content characteristic data; based on the work content characteristic data, historical performance data and historical evaluation data, training a human efficiency evaluation model; being able to increase the richness of human efficiency evaluation dimensions and improve the fairness and accuracy of human efficiency evaluation results.

[0035] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0037] Figure 1 is a schematic diagram of the main process of the human efficiency evaluation model training method according to an embodiment of the present invention;

[0038] Figure 2 It is a schematic diagram of the main process of a human efficiency evaluation model training method according to a reference embodiment of the present invention;

[0039] Figure 3 is a schematic diagram of main modules of a human efficiency evaluation model training device according to an embodiment of the present invention;

[0040] Figure 4 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0041] Figure 5 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0043] It should be noted that the acquisition, storage and application of personal information etc. involved in the embodiments of the present invention are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0044] At present, the fairness and accuracy of human efficiency evaluation results are poor. For example, in the front-line operation process of logistics terminals, managers need to evaluate the workload and work efficiency of each courier. Fair and reasonable work efficiency evaluation methods and evaluation systems can maximize the enthusiasm and subjective initiative of employees, thereby improving the overall production and operation efficiency of the enterprise and creating more value for the enterprise. However, the main work content of employees in different positions is quite different. A single evaluation standard and evaluation dimension will cause the lack of fairness in human efficiency evaluation. For example, the main responsibilities of couriers in logistics terminals are collection and delivery. The main evaluation dimension of their work efficiency is the number of operations (collection or delivery). Collection has many more links with customers than delivery. Therefore, collection operations may need to be given a higher evaluation weight than delivery operations. Although positions such as couriers with cars, on-site couriers, and heavy cargo couriers are all courier positions, and have collection and delivery operations, they are targeted at different scenarios and have different operation volumes. For example, on-site couriers do not need to collect goods door-to-door, heavy cargo couriers have larger cargo bubble counts, and it is difficult to deliver goods to the building. Couriers with cars also have expenses such as fuel costs. In the front-line operation scenarios of logistics, many data interactions are offline, and only a small number of system processes are online. Therefore, online data only accounts for a small part of the required data, and the quality of online data is poor. The data may be inaccurate due to certain objective reasons (such as equipment failure, equipment delay, equipment failure to upgrade, etc.) or subjective reasons (such as brushing orders, brushing indicators, etc.), affecting the accuracy of personnel efficiency evaluation results.

[0045] In view of this, according to one aspect of an embodiment of the present invention, a method for training a human efficiency evaluation model is provided.

[0046] Figure 1 FIG. 1 is a schematic diagram of the main process of the human efficiency evaluation model training method according to an embodiment of the present invention. Figure 1 As shown, according to an embodiment of the present invention, the following embodiments are included:

[0047] Step S101, obtaining production system data, historical performance data and historical evaluation data;

[0048] Among them, the production system data is data that records various key parameters, indicators or events in the production process. Specifically, the production system can be a waybill full-process tracking production system, a personnel information production system or a site basic information system. The waybill full-process tracking production system can count the work operation data such as the number of collected orders, the number of dispatched orders, and the number of split orders per day, and can also provide data such as the weight, volume, whether there is a time limit requirement, whether it is heavy goods, whether there is a special quotation, and whether it is a preferential product. The personnel information production system can provide the basic entry information and department information of all front-line staff, such as the unique number of personnel, position, station, province and other administrative divisions. The site basic information system can provide daily updated collection and dispatch area data, for example, whether it is a residential area, an office building, a merchant's shipping place, area, whether there is an elevator, etc.

[0049] Historical performance data is the performance data of personnel in work or tasks that is recorded and stored regularly. It can be daily, monthly, quarterly or annual performance data, reflecting the performance of personnel in completing work tasks, achieving goals or achieving results. The historical performance data includes: key performance indicators, workload, quality indicators, customer satisfaction, project completion, goal achievement, etc.; historical performance data can be presented in quantitative form, such as numbers, percentages, scores, etc., and can be in the form of tables, reports, database records, etc.; historical performance data can come from system records, data statistics, performance reports, etc., such as sales, production output, customer feedback, etc.

[0050] Historical evaluation data is data that is recorded and stored regularly to evaluate personnel, and can be daily, monthly, quarterly or annual evaluation data. The historical evaluation data includes: work performance evaluation, ability evaluation, teamwork evaluation, customer evaluation, self-evaluation, development suggestions, etc., which can be presented in qualitative form, such as text description, rating, feedback, etc., and can be in the form of evaluation form, evaluation report, feedback meeting minutes, etc.; the historical evaluation data comes from direct superiors, colleagues, customers or self-evaluation, such as performance appraisal, customer satisfaction survey, etc.

[0051] Acquire production system data, historical performance data, and historical evaluation data. By acquiring multiple data sources, increase the data required for personal performance evaluation, more comprehensively evaluate personnel's work performance and performance level, and improve the objectivity and accuracy of the evaluation.

[0052] Step S102, determining work content characteristic data based on the production system data;

[0053] Specifically, the work content characteristic data refers to the characteristic information related to the work content contained in the production system data, and the characteristic information can reflect different aspects of the work content. The work content characteristic data includes: work type, work tasks, workload, work quality, work timeliness, work scenes, work skill requirements, special work environment, etc. The work type refers to the specific work type or position that the person is engaged in, such as a collector, a deliveryman, a warehouse manager, etc.; the work tasks refer to the specific tasks or work content that the person needs to complete at work, such as collecting express deliveries, sorting goods, delivering parcels, etc.; the workload refers to the number of works completed by the person within a certain period of time, such as the number of collection orders, the number of delivery orders, the number of goods processed, and other indicators; the work quality refers to the quality level of the work completed by the person, such as the damage rate of goods, the wrong delivery rate, the customer complaint rate, etc.; the work timeliness refers to the timeliness of the work completed by the person, such as the delivery timeliness, the collection timeliness, etc.; the special working environment refers to the special working environment or situation that the person may face at work, such as weather conditions, traffic conditions, etc.; the work skill requirements refer to the skills or ability requirements that the person must possess at work, such as driving skills, communication skills, etc.; the work scene refers to the specific scene or working environment of the employee, such as residential areas, commercial areas, office buildings, etc.

[0054] By collecting and analyzing job content characteristic data and incorporating them into the model, we can conduct a more comprehensive and systematic human effectiveness evaluation and more accurately evaluate the work performance and performance levels of personnel in different positions and working environments.

[0055] Step S103: training a human efficiency evaluation model based on the work content feature data, historical performance data, and historical evaluation data.

[0056] Among them, the human efficiency evaluation model refers to a mathematical model for predicting personnel work performance based on work content characteristic data, historical performance data and historical evaluation data, using machine learning or statistical modeling technology. The human efficiency evaluation model can predict personnel's performance and performance level in future work by analyzing personnel's work content characteristics and their past performance and evaluation data.

[0057] Specifically, the human efficiency evaluation model can be a regression model, a classification model, an integrated learning model or a deep learning model, etc. When training the human efficiency evaluation model, it is necessary to use the work content feature data, historical performance data and historical evaluation data as the samples of the model, and to train the machine learning algorithm on these samples to establish a model that can accurately predict the work performance of personnel.

[0058] The embodiment of the present invention obtains production system data, historical performance data and historical evaluation data; determines work content feature data based on the production system data; and trains a human efficiency evaluation model based on the work content feature data, historical performance data and historical evaluation data; thereby increasing the richness of human efficiency evaluation dimensions and improving the fairness and accuracy of human efficiency evaluation results.

[0059] Optionally, step S102 determines work content characteristic data based on the production system data, including:

[0060] Extracting basic feature data from the production system data; wherein the basic feature data includes item features, delivery area features, operation features, and service type features;

[0061] The basic feature data is subjected to feature clustering to obtain the work content feature data; wherein the work content feature data includes work quality, work attitude and work environment.

[0062] Specifically, basic feature data related to the work content is extracted from the above production system data, and the basic feature data includes item features, collection and delivery area features, operation features, and service type features. Item features include information such as the weight, volume, whether the delivery address is the same, and quantity of the package; collection and delivery area features include information such as whether it is a residential area, whether it is an office building, whether it is the merchant's shipping address, population density, personnel type, and whether there is an elevator; operation features include information such as the number of collection orders and the number of delivery orders; service type features include information such as whether it is a small item, whether a specific discount is used, and whether timeliness is required.

[0063] After extracting these basic feature data, clustering is performed on these basic feature data, and data with similar features are classified to form new feature groups. Pattern recognition technology can be used to analyze the relationship between different basic features, thereby effectively simplifying the data structure and extracting feature groups that can better describe the work content. The work content feature data obtained through feature clustering mainly includes indicators such as work quality, work attitude, and work environment; work quality is used to indicate the accuracy and efficiency of operation execution, which is directly related to the completion of production tasks; work attitude is used to reflect the enthusiasm and sense of responsibility of staff in the production process; work environment is used to indicate factors that affect the staff's operating experience and output level, such as the physical environment of the workplace, equipment status, and other factors.

[0064] Specifically, a feature clustering algorithm is used to perform cluster analysis on the above basic feature data, and data points with similar features are classified into the same category; the data set obtained after feature clustering is regarded as work content feature data, and each cluster represents a group of data points with similar work content features, reflecting different types of work content; the obtained work content feature data is analyzed and interpreted to understand the work content features represented by each cluster, as well as the relationships and differences between different features.

[0065] Among them, the feature clustering algorithm is an algorithm for grouping data points with similar features into the same cluster or category, for example, the K-means clustering algorithm: by iteratively assigning data points to K clusters and updating the center points of the clusters to minimize the square error within the cluster; hierarchical clustering algorithm: a bottom-up or top-down clustering method, which builds a clustering tree based on the similarity between data points. In bottom-up agglomerative clustering, each data point starts as a separate cluster, and then the clusters are gradually merged based on similarity. In top-down divisive clustering, all data points start as a cluster, and then the clusters are gradually divided based on similarity; density clustering algorithm: clusters are identified by discovering high-density areas without pre-specifying the number of clusters, core points, boundary points and noise points are identified based on the density of data points, and core points are connected together to form clusters; spectral clustering algorithm: a clustering method based on graph theory, which can effectively process non-convex clusters and high-dimensional data by converting data into the Laplacian matrix of the graph and using eigenvectors to reduce the dimension and cluster the data.

[0066] The embodiment of the present invention extracts basic feature data from the production system data and obtains work content feature data through feature clustering, so as to better understand and describe the work content characteristics of personnel and provide a data basis for the subsequent establishment of a human efficiency evaluation model.

[0067] Before the basic feature data is subjected to feature clustering, the method further includes: preprocessing the basic feature data; wherein the preprocessing includes at least one of the following: data deduplication processing, missing data completion processing, outlier processing, invalid data processing, and data normalization processing.

[0068] Specifically, before the basic feature data is clustered, it is usually necessary to pre-process the data. The data deduplication process is to delete the existing identical data according to different dimensions. For example, according to the minimum granularity (i.e., the personnel granularity), the personnel information stored at different times in the database is deduplicated according to the latest valid state. The missing data completion process is to process null values, which can be filled with 0 values, the most recent data, or the average value within a certain time range. The outlier processing needs to be selected in combination with the business scenario to be retained or eliminated. Usually, the risk control scenario focuses more on outliers, so they are usually retained or even analyzed in depth. However, for machine learning, outliers will affect the results of model training. Outliers can be processed by deleting outliers, replacing them with specific values, truncating them, etc. to ensure the accuracy and robustness of the data. In addition to the above-mentioned duplicate values, missing values, and outliers, there will also be some invalid data caused by data import, data entry, storage format, etc. For example, it should be 1.04, but it is stored as 1.o4. This part of invalid data is manually processed. Normalization is another data processing method used to compress the range of data to the range of 0 to 1, including 0 and 1. Normalization can keep the data in the same range.

[0069] In addition, in the logistics terminal business scenario, due to the unstable business of newly opened sites, you can also select appropriate business departments to be included in the model based on the site's operating time and operating status. For example, sites with an operating time of less than one month will not be included in the model for the time being, or sites with extremely scarce staff and in the peak recruitment period will not be included in the model.

[0070] Before performing feature clustering, the embodiment of the present invention pre-processes the basic feature data by at least one of data deduplication, missing data completion, outlier processing, invalid data processing, and data normalization processing, so as to improve the quality and accuracy of the clustering results and ensure the reliability and robustness of the data.

[0071] Optionally, step S103 trains a human efficiency evaluation model based on the work content feature data, historical performance data and historical evaluation data, including: using the work content feature data, historical performance data and historical evaluation data to construct a training sample set, taking the work content feature data, historical performance data and historical evaluation data as input, and taking a pre-set classification result as a target classification result, to train a human efficiency evaluation model.

[0072] Specifically, the work content feature data, historical performance data and historical evaluation data are integrated into a training sample set, and the training sample set is preprocessed. The data preprocessing can be feature scaling, data normalization, processing missing values, etc., to ensure the quality and consistency of the data; according to the nature of the task and the characteristics of the data, a suitable machine learning model or deep learning model is selected, such as the ML-KNN model (Maximum Likelihood k-Nearest Neighbor, multi-label K nearest neighbor algorithm model), logistic regression, support vector machine, random forest, deep neural network, etc.; use the training sample set to train the selected model, take the work content feature data, historical performance data and historical evaluation data as input, and use the pre-set classification result as the target classification result; update the model parameters by optimizing the algorithm to make the model's prediction result as close to the actual classification result as possible; use the validation set or cross-validation method to evaluate the trained model, evaluate the model's performance and accuracy, for example, use various indicators such as accuracy, precision, recall rate, F1 score, etc. to evaluate the performance of the model; tune the model according to the evaluation results, adjust the model's hyperparameters, change the model structure, or use ensemble learning and other methods to improve the model's performance; use the trained model to predict and classify new data and evaluate the work performance of personnel.

[0073] The embodiments of the present invention construct a training sample set by utilizing work content feature data, historical performance data, and historical evaluation data, and train a human effectiveness evaluation model, thereby being able to accurately predict the results of personnel work performance. By comprehensively considering work content characteristics, past work performance, and evaluation data, it is possible to more comprehensively evaluate personnel work performance. In addition, the management party provides targeted management and training suggestions based on the evaluation results, thereby improving the work efficiency and overall productivity of the personnel, and further achieving the management party's long-term development goals.

[0074] Optionally, the human efficiency evaluation model adopts a multi-label K nearest neighbor algorithm model.

[0075] Among them, the multi-label K nearest neighbor algorithm model (ML-KNN algorithm) is a multi-label classification algorithm that uses the K nearest neighbor algorithm to classify multi-label data. Specifically, work content feature data, historical performance data, and historical evaluation data will be used as input features, and the pre-set classification results will be used as target classification results. Through the ML-KNN algorithm, the model can predict multiple labels of new samples based on the labels of K nearest neighbor samples according to the input feature data, and can take into account labels of different dimensions and diverse work content characteristics to more accurately evaluate the work performance of personnel.

[0076] As shown in Table 1 below, the features of three different dimensions can be arranged and combined to form a total of 3×3×3=27 categories, including {a,d,g}, {a,e,g}, {a,f,g}, {b,d,g}, {c,f,g}, {c,d,h}, {a,d,i}, etc.

[0077]

[0078] The embodiment of the present invention adopts a multi-label K nearest neighbor algorithm model to improve the accuracy and efficiency of multi-label classification tasks. It combines the intuitiveness of the K nearest neighbor algorithm and the needs of multi-label classification, can effectively use the similarities between samples to predict multiple labels, and can more comprehensively understand and describe data by comprehensively considering the relationship between multiple features and multiple labels, thereby improving the accuracy and robustness of classification.

[0079] Figure 2 FIG. 1 is a schematic diagram of the main process of a human efficiency evaluation model training method according to a reference embodiment of the present invention. Figure 2 As shown, after the training to obtain the human efficiency evaluation model in step S103, the following steps are also included:

[0080] Step S201, in response to a request for human efficiency evaluation, obtaining information to be evaluated;

[0081] Step S202, inputting the information to be evaluated into a human efficiency evaluation model for evaluation to obtain a prediction result of the information to be evaluated;

[0082] Step S203: Push the prediction result to the investigation and review platform for verification, receive the verification result fed back by the investigation and review platform, and determine the final evaluation result of the information to be evaluated based on the verification result.

[0083] The request for human efficiency evaluation is a request initiated by the human resources department, managers or other relevant departments to evaluate the work performance of personnel (such as collectors, delivery personnel, managers, etc.). The information to be evaluated refers to the relevant data or information of the person to be evaluated, including data on the person's work performance, work attitude, work efficiency, etc. within a certain period of time.

[0084] In response to the above-mentioned request for human efficiency evaluation, the information to be evaluated of the personnel to be evaluated is collected from relevant data sources. After the information to be evaluated is collected, the information to be evaluated is input into the above-mentioned pre-trained human efficiency evaluation model for evaluation. The human efficiency evaluation model analyzes the input information and generates a prediction result, that is, a prediction result of the personnel's work performance; in order to ensure the accuracy and credibility of the evaluation result, the above-mentioned prediction result is pushed to the investigation and review platform for verification, and the investigation and review platform further verifies, compares, reviews and other operations on the prediction result to ensure the authenticity and accuracy of the data; the final evaluation result of the information to be evaluated is determined based on the verification result fed back by the investigation and review platform.

[0085] Among them, the front-line investigation and review personnel will view and download the account clue information that needs to be investigated and verified on the investigation and review platform, and conduct offline investigation and verification. The front-line investigators will feedback the actual investigation results to the headquarters data personnel through offline feedback, and incorporate the data results as input into the basic feature data.

[0086] In addition, you can also view the final evaluation results for different personnel according to the time series, that is, according to the unique number of the personnel, their work efficiency is arranged in ascending or descending order according to the time dimension (hours, days, weeks, months, quarters, years, etc.), so as to view the work efficiency results of each employee, which is convenient for managers to evaluate and see the historical work efficiency of each person, which can be used as an evaluation reference.

[0087] For example, as shown in Table 2, the output of the above ML-KNN model is pushed to the investigation and review platform, and the following verification results are obtained:

[0088]

[0089] The embodiment of the present invention establishes a comprehensive and objective human efficiency evaluation system. By responding to evaluation requests, the information to be evaluated is obtained and input into the evaluation model for analysis. It can quickly and accurately evaluate the work performance of personnel and push the predicted results to the investigation and review platform for verification. The credibility and fairness of the evaluation results can be improved, and the final evaluation results are determined based on the verification results, ensuring the accuracy and objectivity of the evaluation process.

[0090] Optionally, after determining the final evaluation result of the information to be evaluated based on the verification result, the method further includes: determining a target user corresponding to the final evaluation result that is lower than a preset evaluation threshold; and assigning tasks to the target user based on a preset incentive strategy.

[0091] The evaluation threshold is a standard set in advance to determine whether the final evaluation result meets the expected goal. When a final evaluation result is lower than the preset evaluation threshold, it means that the evaluation object has failed to meet the expected standards in some aspects, such as indicators such as production efficiency, work quality or service satisfaction are lower than the threshold. The target users corresponding to these low evaluation results are identified and classified as user groups that need further intervention or attention. Tasks are assigned to these target users based on the preset incentive strategy. The incentive strategy can be set according to business needs and employee performance, including reward mechanisms, training opportunities, work adjustments, etc.; tasks that match the current capabilities and needs of the target users can be assigned in combination with their specific circumstances. For example, for target users with poor performance, training tasks to improve skills can be assigned, or their workload can be reduced to help them gradually improve their efficiency. Through this targeted task allocation, incentives can be implemented more effectively, which helps to improve the performance and overall work quality of target users. Managers can track the completion of these incentive tasks in real time, and further adjust the incentive strategy based on user feedback and task execution results to ensure continuous optimization and improvement. At the same time, for some target users who need emotional motivation, team activities can be arranged or psychological support can be provided to help relieve work pressure and improve job satisfaction. By paying attention to the emotions and psychological state of target users, timely measures can be taken to relieve employees' negative emotions and improve work enthusiasm and satisfaction.

[0092] The embodiments of the present invention can help the system continuously optimize user performance by analyzing evaluation results, identifying target users and implementing incentives, thereby improving overall operational efficiency and service quality.

[0093] Optionally, after determining the final evaluation result of the information to be evaluated based on the verification result, it also includes: inputting the information to be evaluated and the final evaluation result of the information to be evaluated into the human efficiency evaluation model to perform iterative optimization training on the human efficiency evaluation model.

[0094] Specifically, first, collect the information to be evaluated and the corresponding final evaluation results, pre-process the collected data, and divide the data set consisting of the above data into a training set and a test set; input the training set into the human efficiency evaluation model to train the model, and the model will learn the relationship between work performance based on the input features and the corresponding evaluation results; use the test set to verify the trained model and evaluate the performance of the model; optimize the model based on the verification results; repeat the above steps, iterate the model multiple times, and each iteration will adjust and optimize the model based on the new data and verification results until the required performance is achieved.

[0095] The embodiment of the present invention inputs the information to be evaluated and the final evaluation results into the human efficiency evaluation model for iterative optimization training, which can continuously improve and optimize the performance and accuracy of the model. By feeding back the actual information to be evaluated and the corresponding final evaluation results to the model, the model can be helped to learn and correct possible deviations or errors, thereby improving the prediction and generalization capabilities of the model. Through iterative optimization training, the model can gradually adapt to the ever-changing data and environment, better reflect the actual situation, and provide more accurate and reliable results for the evaluation.

[0096] According to a second aspect of the embodiment of the present invention, Figure 3 As shown, a human efficiency evaluation model training device 300 includes:

[0097] Data acquisition module 301, acquiring production system data, historical performance data and historical evaluation data;

[0098] A feature extraction module 302 determines work content feature data based on the production system data;

[0099] The model training module 303 trains a human efficiency evaluation model based on the work content feature data, historical performance data and historical evaluation data.

[0100] Optionally, the feature extraction module 302 is further used for:

[0101] Basic feature data is extracted from the production system data, and feature clustering is performed on the basic feature data to obtain the work content feature data.

[0102] Optionally, the feature extraction module 302 is further used for:

[0103] The basic feature data is preprocessed; wherein the preprocessing includes at least one of the following: data deduplication processing, missing data completion processing, outlier processing, invalid data processing, and data normalization processing.

[0104] Optionally, the model training module 303 is further used to:

[0105] A training sample set is constructed using the work content feature data, historical performance data and historical evaluation data. The work content feature data, historical performance data and historical evaluation data are used as input, and a pre-set classification result is used as a target classification result to train a human efficiency evaluation model.

[0106] Optionally, the human efficiency evaluation model in the model training module 303 adopts a multi-label K nearest neighbor algorithm model.

[0107] Optionally, the human efficiency evaluation model training device 300 further includes a human efficiency evaluation module, and the human efficiency evaluation module is used to:

[0108] Respond to requests for human efficiency evaluation and obtain information to be evaluated;

[0109] Inputting the information to be evaluated into a human efficiency evaluation model for evaluation to obtain a prediction result of the information to be evaluated;

[0110] The prediction result is pushed to the investigation and review platform for verification, and the verification result fed back by the investigation and review platform is received, and the final evaluation result of the information to be evaluated is determined based on the verification result.

[0111] Optionally, the human efficiency evaluation module is further used to:

[0112] The information to be evaluated and the final evaluation result of the information to be evaluated are input into the human efficiency evaluation model to perform iterative optimization training on the human efficiency evaluation model.

[0113] It should be noted that the specific implementation content of the device for training the human efficiency evaluation model described in the present invention has been described in detail in the method for training the human efficiency evaluation model described above, so the repeated content will not be described here.

[0114] According to a third aspect of an embodiment of the present invention, there is provided an electronic device, including:

[0115] one or more processors;

[0116] a storage device for storing one or more programs,

[0117] When the one or more programs are executed by the one or more processors, the one or more processors implement the method provided in the first aspect of the embodiment of the present invention.

[0118] According to a fourth aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored, and when the program is executed by a processor, the method provided by the first aspect of the embodiment of the present invention is implemented.

[0119] According to a fifth aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the method provided by the first aspect of the embodiment of the present invention is implemented.

[0120] Figure 4 An exemplary system architecture 400 is shown to which the human efficiency evaluation model training method or human efficiency evaluation model training device according to the embodiment of the present invention can be applied.

[0121] like Figure 4As shown, system architecture 400 may include terminal devices 401, 402, 403, network 404 and server 405. Network 404 is used to provide a medium for communication links between terminal devices 401, 402, 403 and server 405. Network 404 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0122] Users can use terminal devices 401, 402, 403 to interact with server 405 through network 404 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 401, 402, 403, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0123] The terminal devices 401 , 402 , and 403 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers.

[0124] The server 405 may be a server that provides various services, such as a backend management server (only an example) that provides support for shopping websites browsed by users using the terminal devices 401, 402, and 403. The backend management server may analyze and process the received data such as the labor efficiency evaluation request, and feed back the processing results (such as the training results - only an example) to the terminal device.

[0125] It should be noted that the human efficiency evaluation model training method provided in the embodiment of the present invention is generally executed by the server 405 , and accordingly, the human efficiency evaluation model training device is generally arranged in the server 405 .

[0126] It should be understood that Figure 4 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0127] Reference below Figure 5 , which shows a schematic diagram of the structure of a computer system 500 of a terminal device suitable for implementing an embodiment of the present invention. Figure 5 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0128] like Figure 5As shown, the computer system 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 502 or a program loaded from a storage part 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the system 500 are also stored. The CPU 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0129] The following components are connected to the I / O interface 505: an input section 506 including a keyboard, a mouse, etc.; an output section 507 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN card, a modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to the I / O interface 505 as needed. A removable medium 511, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 510 as needed, so that a computer program read therefrom is installed into the storage section 508 as needed.

[0130] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 509, and / or installed from the removable medium 511. When the computer program is executed by the central processing unit (CPU) 501, the above-mentioned functions defined in the system of the present invention are executed.

[0131] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0132] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0133] The modules involved in the embodiments of the present invention may be implemented in software or hardware. The modules described may also be set in a processor. For example, they may be described as: a processor including: a data acquisition module, a model training module, and a feature extraction module. The names of these modules do not, in some cases, constitute limitations on the modules themselves. For example, the data acquisition module may also be described as a "module for acquiring production system data, historical performance data, and historical evaluation data."

[0134] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: obtaining production system data, historical performance data, and historical evaluation data; determining work content feature data based on the production system data; and training a human efficiency evaluation model based on the work content feature data, historical performance data, and historical evaluation data.

[0135] The computer program product provided by the embodiment of the present invention includes a computer program, and when the computer program is executed by a processor, the human efficiency evaluation model training method in the embodiment of the present invention is implemented.

[0136] According to the technical solution of the embodiment of the present invention, it has the following advantages or beneficial effects: by acquiring production system data, historical performance data and historical evaluation data; based on the production system data, determining the work content feature data; based on the work content feature data, historical performance data and historical evaluation data, training a human efficiency evaluation model; it can increase the richness of the human efficiency evaluation dimensions and improve the fairness and accuracy of the human efficiency evaluation results.

[0137] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0138] It should be noted that the acquisition, storage and application of user personal information involved in the technical solution of the present disclosure are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

Claims

1. A method for training a human efficiency evaluation model, characterized in that: include: Obtain production system data, historical performance data, and historical evaluation data; Determining work content characteristic data based on the production system data; Based on the work content feature data, historical performance data and historical evaluation data, a human efficiency evaluation model is trained.

2. The method according to claim 1, characterized in that Based on the production system data, work content characteristic data is determined, including: Extracting basic feature data from the production system data; wherein the basic feature data includes item features, delivery area features, operation features, and service type features; The basic feature data is subjected to feature clustering to obtain the work content feature data; wherein the work content feature data includes work quality, work attitude and work environment.

3. The method according to claim 1, characterized in that Based on the work content feature data, historical performance data and historical evaluation data, a human efficiency evaluation model is trained, including: A training sample set is constructed using the work content feature data, historical performance data and historical evaluation data. The work content feature data, historical performance data and historical evaluation data are used as input, and a pre-set classification result is used as a target classification result to train a human efficiency evaluation model.

4. The method according to claim 1, characterized in that After the human efficiency evaluation model is obtained through training, the method further includes: Respond to requests for human efficiency evaluation and obtain information to be evaluated; Inputting the information to be evaluated into a human efficiency evaluation model for evaluation to obtain a prediction result of the information to be evaluated; The prediction result is pushed to the investigation and review platform for verification, and the verification result fed back by the investigation and review platform is received, and the final evaluation result of the information to be evaluated is determined based on the verification result.

5. The method according to claim 4, characterized in that After determining the final evaluation result of the information to be evaluated based on the verification result, the method further includes: Determine target users corresponding to final evaluation results below a preset evaluation threshold; Tasks are assigned to the target users based on a preset incentive strategy.

6. The method according to claim 4, characterized in that After determining the final evaluation result of the information to be evaluated based on the verification result, the method further includes: The information to be evaluated and the final evaluation result of the information to be evaluated are input into the human efficiency evaluation model to perform iterative optimization training on the human efficiency evaluation model.

7. A human efficiency evaluation model training device, characterized in that: include: Data acquisition module, which acquires production system data, historical performance data, and historical evaluation data; A feature extraction module, based on the production system data, determines work content feature data; The model training module trains a human efficiency evaluation model based on the work content feature data, historical performance data and historical evaluation data.

8. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images