Multi-mode report generation method for attendance software and related device thereof

Abstract

The invention discloses a multi-mode report generation method for attendance checking software and a related device thereof, and the method comprises the steps: obtaining attendance checking data from the attendance checking software, and extracting attendance checking features from the attendance checking data; training a language model through the attendance data and the attendance characteristics to obtain a generative model; and generating text content of the report according to the attendance data through the generative model, and generating a chart and an image corresponding to the report through a data visualization method. According to the method, the text content of the report is generated according to the attendance data through the generative model, and the chart and the image are generated through a visualization technology, so that the report generation efficiency is greatly improved, the time for manually compiling the report is shortened, and the report generation efficiency is improved; the report generated by the method comprises a plurality of forms such as texts, charts and images, the readability and intuition of the report are improved, and through multi-dimensional analysis and visual display of the attendance data, enterprises are helped to better make human resource management decisions, and the working efficiency is improved.

Description

Technical Field

[0001] This application relates to the technical field of attendance data processing, and in particular, to a multi-modal report generation method for attendance software and its related devices. Background Art

[0002] With the acceleration of the digital transformation of enterprises, attendance software has become an important tool for enterprise human resource management. However, the functions of traditional attendance software are relatively single, which can only record the attendance of employees and cannot conduct in-depth analysis and visual display of attendance data. In addition, when making human resource management decisions, enterprises need to consider multiple factors comprehensively, and traditional attendance software cannot meet this requirement. Moreover, most of the existing technologies write reports manually, resulting in low efficiency. Therefore, there is an urgent need for a method that can perform multi-dimensional analysis and visual display of attendance data to help enterprises make better human resource management decisions and improve efficiency. Summary of the Invention

[0003] This application provides a multi-modal report generation method for attendance software and its related devices, which is used to perform multi-dimensional analysis and visual display of attendance data to help enterprises make better human resource management decisions and improve efficiency.

[0004] In view of this, in the first aspect of this application, a multi-modal report generation method for attendance software is provided, including:

[0005] Obtain attendance data from the attendance software, and extract attendance features from the attendance data;

[0006] Train a language model with the attendance data and the attendance features to obtain a generative model;

[0007] Generate the text content of the report according to the attendance data through the generative model, and generate the corresponding charts and images of the report through data visualization methods.

[0008] Optionally, the extracting attendance features from the attendance data includes:

[0009] Calculate the work output of each employee within a unit time according to the attendance data to obtain the work efficiency of each employee;

[0010] Based on the attendance data and the shift schedule data as basic information, analyze the attendance situation, time characteristics and abnormal situations of each employee through a model to obtain the work quality of each employee.

[0011] Optionally, the training a language model with the attendance data and the attendance features to obtain a generative model includes:

[0012] Obtain report text samples according to the attendance features corresponding to the attendance data;

[0013] Pair the reported text sample with the attendance data to construct a data set;

[0014] Train a language model with the data set so that the language model learns the context relationship of the input sequence to generate text content, obtaining the predicted text content;

[0015] Calculate the difference value between the predicted text content and the corresponding reported text sample through a loss function, and reversely update the parameters of the language model through the difference value until the language model converges, obtaining a generative model.

[0016] Optionally, the method further includes:

[0017] Calculate the correlation between each attendance feature and the performance metric of the language model to obtain the importance score of each attendance feature; the performance metric of the language model is the BLEU score or ROUGE score of the predicted text content and the corresponding reported text sample;

[0018] Sort each attendance feature according to the importance score of each attendance feature, and remove one or more attendance features with the smallest importance score to obtain the filtered attendance features;

[0019] Train the language model according to the filtered attendance features to obtain a generative model.

[0020] Optionally, the method further includes:

[0021] Typeset the generated text content, charts and images to generate a final multi-modal report;

[0022] Output the multi-modal report in the form of PDF, Word and / or PPT.

[0023] The second aspect of this application provides a multi-modal report generation device for attendance software, including:

[0024] A data acquisition unit for acquiring attendance data from the attendance software;

[0025] A feature extraction unit for extracting attendance features from the attendance data;

[0026] A model training unit for training a language model with the attendance data and the attendance features to obtain a generative model;

[0027] A report generation unit for generating the text content of the report according to the attendance data through the generative model, and generating the charts and images corresponding to the report through a data visualization method.

[0028] Optionally, the feature extraction unit is specifically configured to calculate the work output of each employee per unit time based on the attendance data, and obtain the work efficiency of each employee;

[0029] Based on the attendance data and the shift schedule data as basic information, analyze the attendance situation, time characteristics, and abnormal situations of each employee through a model to obtain the work quality of each employee.

[0030] Optionally, the model training unit is specifically configured to obtain a report text sample according to the attendance characteristics corresponding to the attendance data;

[0031] Pair the report text sample with the attendance data to construct a data set;

[0032] Train a language model through the data set, so that the language model learns the context relationship of the input sequence to generate text content, and obtain the predicted text content;

[0033] Calculate the difference value between the predicted text content and the corresponding report text sample through a loss function, and update the parameters of the language model in reverse through the difference value until the language model converges to obtain a generative model.

[0034] A third aspect of the present application provides an electronic device, which includes a processor and a memory;

[0035] The memory is used to store program code and transmit the program code to the processor;

[0036] The processor is configured to execute any one of the multi-modal report generation methods for attendance software described in the first aspect according to the instructions in the program code.

[0037] A fourth aspect of the present application provides a computer-readable storage medium, which is characterized in that the computer-readable storage medium is used to store program code, and when the program code is executed by a processor, it implements any one of the multi-modal report generation methods for attendance software described in the first aspect.

[0038] It can be seen from the above technical solutions that the present application has the following advantages:

[0039] The present application generates the text content of the report according to the attendance data through a generative model, and generates charts and images of the attendance data through visualization technology, which greatly improves the report generation efficiency, reduces the time for manually writing reports, and improves the report generation efficiency; the reports generated by the present application include various forms such as text, charts, and images, which improve the readability and intuitiveness of the reports. Through multi-dimensional analysis and visualization display of the attendance data, it helps the enterprise to make better human resource management decisions and improve work efficiency. Description of the Drawings

[0040] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0041] Figure 1 It is a flowchart of a multi-modal report generation method for an attendance software provided by an embodiment of the present application;

[0042] Figure 2 It is a structural diagram of a multi-modal report generation device for an attendance software provided by an embodiment of the present application. Detailed implementation manners

[0043] In order to enable those skilled in the art to better understand the solution of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0044] For ease of understanding, please refer to Figure 1 , an embodiment of the present application provides a multi-modal report generation method for an attendance software, including:

[0045] Step 110: Obtain attendance data from the attendance software and extract attendance features from the attendance data;

[0046] The attendance data of each employee can be read from the attendance software (such as DingTalk, Enterprise WeChat, etc.), including clock-in time, late / early leave situation, leave records, etc. The collected attendance data can be subjected to data cleaning and preprocessing to remove outliers and noise data to ensure the accuracy and integrity of the data. For example, for the attendance data with abnormal clock-in time, it can be compared with the actual working hours of the employee to determine whether there are abnormal situations.

[0047] Then, attendance features related to employees' work performance are extracted from the preprocessed attendance data, such as working hours, work efficiency, and work quality. Specifically, the work efficiency of employees can be obtained by calculating the amount of work output per unit time; based on the attendance data and the scheduling data as basic information, the attendance situation of each employee (such as attendance rate, number of late / early leave / absenteeism, etc.), time characteristics (daily working hours, overtime situation, implementation of special schedules, overtime / late / early leave / absenteeism / absence, etc., stability of clock-in time, regularity of rest time, etc.) and abnormal situations (abnormal clock-in records / special clock-in records, etc.) are analyzed through a model to obtain the work quality of each employee.

[0048] Step 120: Train a language model with the attendance data and attendance features to obtain a generative model;

[0049] In one embodiment, the report text sample can be directly obtained according to the attendance features corresponding to the attendance data; the report text sample can be generated according to a report template, and the report text sample is used to describe the attendance situation and work performance of employees. Alternatively, the report text sample can be manually written according to the attendance features corresponding to the attendance data;

[0050] Pair the report text sample with the attendance data to construct a data set; the data set can be divided proportionally according to different types of data to obtain a training set, a validation set, and a test set; the training set is used to train the model, the validation set is used to adjust the hyperparameters of the model, and the test set is used to evaluate the performance of the model.

[0051] Train the language model with the data set so that the language model learns the context relationship of the input sequence to generate text content, and obtain the predicted text content; the language model can be a language model with a Transformer structure. Calculate the difference value between the predicted text content and the corresponding report text sample through a loss function, and update the parameters of the language model in reverse according to the difference value until the language model converges (the performance of the model on the validation set no longer improves or reaches the preset number of training rounds), and obtain a generative model. An optimizer can be used to update the parameters of the language model, enabling the language model to adaptively adjust the learning rate and automatically optimize the data processing and model training.

[0052] In another embodiment, after training the language model through the above embodiments, the attendance features can be further screened to select the features that are most valuable for report generation. Specifically, first, the correlation between each attendance feature and the performance metrics of the language model can be calculated to evaluate the actual contribution of each attendance feature to the optimization goal, and this correlation can be used as the importance score of each attendance feature; the performance metrics of the language model can be the BLEU score or ROUGE score of the predicted text content and the corresponding report text sample, or the keyword field matching rate of the predicted text content and the corresponding report text sample; the specific calculation processes of the BLEU score and ROUGE score belong to the prior art and will not be elaborated here.

[0053] Then, sort the attendance features according to the importance scores of the attendance features, and remove one or more attendance features with the smallest importance scores (i.e., the least important one or more attendance features) to obtain the screened attendance features;

[0054] Finally, train the language model according to the screened attendance features to obtain a generative model. Obtain the report text sample according to the screened attendance features, then configure the report text sample with the attendance data to construct a dataset, and further train the language model to obtain the final generative model.

[0055] Obtaining the attendance features that are most valuable for the report through feature screening helps to improve the accuracy of report generation.

[0056] Step 130: Generate the text content of the report according to the attendance data through the generative model, and generate the corresponding charts and images of the report through the data visualization method;

[0057] After training the generative model, the generative model can be embedded in the attendance software to automatically retrieve the attendance data of each employee in the attendance software, and then generate the text content of the report. The text content is used to describe the attendance situation (attendance, number of late arrivals and early departures, number of leave days, etc.) and work performance (working hours, work efficiency, work quality, etc.) of each employee; the corresponding charts and images of the report can also be generated through the data visualization method. The chart part is used to display the statistical analysis results of the attendance data, which can include the distribution of attendance data, trend analysis, etc.; the image part is used to display the work scenes and work achievements of employees, which can include work photos of employees, work achievement displays, etc.

[0058] Furthermore, the generated text content, charts and images can be typeset through a typesetting software to generate the final multi-modal report, and the multi-modal report can be output in the forms of PDF, Word, PPT, etc., which is convenient for enterprise managers to view and share.

[0059] It should be noted that the template of the multi-modal report can be designed according to the actual needs of the enterprise. The template can include various forms such as text, charts, images, etc. to meet the needs of different users. For example, for the text part, different paragraphs and headings can be designed to clearly display the attendance situation and work performance of employees.

[0060] In this application, the text content of the report is generated by a generative model based on attendance data, and charts and images of the attendance data are generated by visualization technology, which greatly improves the report generation efficiency, reduces the time for manual report writing, and improves the report generation efficiency; the report generated by this application includes various forms such as text, charts and images, which improves the readability and intuitiveness of the report. Through multi-dimensional analysis and visual display of the attendance data, it helps the enterprise to make better human resource management decisions and improve work efficiency.

[0061] The above is an embodiment of a multi-modal report generation method for attendance software provided by this application. The following is an embodiment of a multi-modal report generation device for attendance software provided by this application.

[0062] Please refer to Figure 2 , a multi-modal report generation device for attendance software provided by an embodiment of this application includes:

[0063] A data acquisition unit for acquiring attendance data from the attendance software;

[0064] A feature extraction unit for extracting attendance features from the attendance data;

[0065] A model training unit for training a language model through the attendance data and attendance features to obtain a generative model;

[0066] A report generation unit for generating the text content of the report by the generative model according to the attendance data, and generating the corresponding charts and images of the report by the data visualization method.

[0067] As a further improvement, the feature extraction unit is specifically used to calculate the work output of each employee within a unit time according to the attendance data to obtain the work efficiency of each employee;

[0068] Based on the attendance data and the shift schedule data as basic information, the attendance situation, time characteristics and abnormal situations of each employee are analyzed by the model to obtain the work quality of each employee.

[0069] As a further improvement, the model training unit is specifically used to obtain report text samples according to the attendance features corresponding to the attendance data;

[0070] Pair the report text samples with the attendance data to construct a data set;

[0071] Train a language model with a dataset so that the language model learns the context relationship of the input sequence to generate text content and obtain predicted text content;

[0072] Calculate the difference value between the predicted text content and the corresponding reported text sample through a loss function, and update the parameters of the language model in reverse according to the difference value until the language model converges to obtain a generative model.

[0073] As a further improvement, the device further includes: a feature screening unit for calculating the correlation between each attendance feature and the performance metric of the language model to obtain the importance score of each attendance feature; the performance metric of the language model is the BLEU score or ROUGE score of the predicted text content and the corresponding reported text sample;

[0074] Sort each attendance feature according to the importance score of each attendance feature, and remove one or more attendance features with the smallest importance score to obtain the screened attendance features;

[0075] Train the language model according to the screened attendance features to obtain a generative model.

[0076] As a further improvement, the device further includes: a layout unit for laying out the generated text content, charts and images to generate a final multimodal report; output the multimodal report in the form of PDF, Word and / or PPT.

[0077] This application uses a generative model to generate the text content of the report based on attendance data, and uses visualization technology to generate charts and images of attendance data, which greatly improves the report generation efficiency, reduces the time for manual report writing, and improves the report generation efficiency; the report generated by this application includes various forms such as text, charts and images, which improves the readability and intuitiveness of the report. Through multi-dimensional analysis and visualization display of attendance data, it helps enterprises make better human resource management decisions and improve work efficiency.

[0078] This application embodiment also provides an electronic device, which includes a processor and a memory;

[0079] The memory is used to store program code and transmit the program code to the processor;

[0080] The processor is used to execute the multimodal report generation method for attendance software in the foregoing method embodiment according to the instructions in the program code.

[0081] This application embodiment also provides a computer-readable storage medium, which is used to store program code. When the program code is executed by a processor, it implements the multimodal report generation method for attendance software in the foregoing method embodiment.

[0082] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the devices and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described herein again.

[0083] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of this application and the above-mentioned drawings are used to distinguish similar objects and do not necessarily describe a specific order or sequence. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that comprises a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0084] It should be understood that in this application, "at least one (item)" means one or more, and "a plurality" means two or more. "And / or" is used to describe the association relationship of associated objects and indicates that there can be three relationships. For example, "A and / or B" can mean: only A exists, only B exists, and both A and B exist at the same time. Among them, A and B can be singular or plural. The character " / " generally means that the associated objects before and after are in an "or" relationship. "At least one (one) of the following" or its similar expression means any combination of these items, including any combination of single item (one) or plural items (ones). For example, at least one (one) of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0085] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the devices or units can be in an electrical, mechanical or other form.

[0086] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0087] In addition, in each embodiment of the present application, each functional unit can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0088] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (English full name: Read-Only Memory, English abbreviation: ROM), random access memories (English full name: Random Access Memory, English abbreviation: RAM), magnetic disks, or optical discs that can store program codes.

[0089] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. A multimodal report generation method for attendance software, characterized in that: include: Acquire attendance data from the attendance software, and extract attendance features from the attendance data; Training a language model by using the attendance data and the attendance features to obtain a generative model; The generative model generates the text content of the report according to the attendance data, and the charts and images corresponding to the report are generated by the data visualization method.

2. The multimodal report generation method for attendance software according to claim 1, characterized in that: The step of extracting attendance features from the attendance data comprises: Calculate the work output of each employee per unit time based on the attendance data to obtain the work efficiency of each employee; Taking the attendance data and shift scheduling data as basic information, the model is used to analyze the attendance, time characteristics and abnormal situations of each employee to obtain the work quality of each employee.

3. The multimodal report generation method for attendance software according to claim 1, characterized in that: The step of training a language model by using the attendance data and the attendance features to obtain a generative model includes: Acquire a report text sample according to the attendance feature corresponding to the attendance data; Pairing the report text sample with the attendance data to construct a data set; Training a language model through the data set enables the language model to learn the contextual relationship of the input sequence to generate text content and obtain predicted text content; The difference value between the predicted text content and the corresponding report text sample is calculated by a loss function, and the parameters of the language model are reversely updated by the difference value until the language model converges to obtain a generative model.

4. The multimodal report generation method for attendance software according to claim 3, characterized in that: The method further comprises: Calculate the correlation between each attendance feature and the performance index of the language model to obtain the importance score of each attendance feature; the performance index of the language model is the BLEU score or ROUGE score of the predicted text content and the corresponding report text sample; sorting the attendance features according to the importance scores of the attendance features, removing one or more attendance features with the smallest importance scores, and obtaining screened attendance features; The language model is trained according to the screened attendance features to obtain a generative model.

5. The multimodal report generation method for attendance software according to claim 1, characterized in that: The method further comprises: Typeset the generated text content, charts and images to produce the final multimodal report; The multimodal report is output in PDF, Word and / or PPT format.

6. A multimodal report generating device for attendance software, characterized in that: include: A data acquisition unit, used for acquiring attendance data from the attendance software; A feature extraction unit, used for extracting attendance features from the attendance data; A model training unit, used for training a language model through the attendance data and the attendance features to obtain a generative model; The report generation unit is used to generate the text content of the report according to the attendance data through the generative model, and generate the charts and images corresponding to the report through the data visualization method.

7. The multimodal report generating device for attendance software according to claim 6, characterized in that: The feature extraction unit is specifically used to calculate the work output of each employee per unit time according to the attendance data to obtain the work efficiency of each employee; Taking the attendance data and shift scheduling data as basic information, the model is used to analyze the attendance, time characteristics and abnormal situations of each employee to obtain the work quality of each employee.

8. The multimodal report generating device for attendance software according to claim 6, characterized in that: The model training unit is specifically used to obtain a report text sample according to the attendance feature corresponding to the attendance data; Pairing the report text sample with the attendance data to construct a data set; Training a language model through the data set enables the language model to learn the contextual relationship of the input sequence to generate text content and obtain predicted text content; The difference value between the predicted text content and the corresponding report text sample is calculated by a loss function, and the parameters of the language model are reversely updated by the difference value until the language model converges to obtain a generative model.

9. An electronic device, characterized in that: The device comprises a processor and a memory; The memory is used to store program codes and transmit the program codes to the processor; The processor is used to execute the multimodal report generation method for attendance software according to any one of claims 1 to 5 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program codes, and when the program codes are executed by a processor, the multimodal report generation method for attendance software according to any one of claims 1 to 5 is implemented.

Images