Artificial intelligence-based work log generation method, device, equipment and medium

Abstract

The embodiments of the present application provide a work log generation method, apparatus, equipment and medium based on artificial intelligence. The artificial intelligence technology automatically analyzes and obtains a first data subset as document type data to obtain a first work log data subset, obtains a second data subset as attendance type data to obtain a second work log data subset, and obtains a third data subset as picture type data to obtain a third work log data subset. Finally, the first work log data subset, the second work log data subset and the third work log data subset are automatically combined to obtain work log report data without manual sorting, thereby improving the efficiency of obtaining work log data and improving data accuracy.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of artificial intelligence, and specifically to a method, device, equipment and medium for generating a work log based on artificial intelligence. Background Art

[0002] Currently, when corporate employees record their work content for this week, this month or this year to form a work report, they generally use manual compilation to count and organize the data, which may result in omissions. This leads to low accuracy and efficiency in the work report data obtained based on manual compilation. Summary of the invention

[0003] The embodiments of the present application provide a work log generation method, device and storage medium based on artificial intelligence, aiming to solve the problems of low accuracy and low efficiency of work report data obtained by manual sorting in the prior art.

[0004] In a first aspect, the embodiment of the present application provides a method for generating a work log based on artificial intelligence from the perspective of a cloud server, the method comprising:

[0005] When it is determined that the time interval between the current system time and the last work log generation time is equal to the preset first time period, a work log data screening time interval is generated according to the current system time and the last work log generation time, and user identity information is obtained;

[0006] Obtain a user work log required data type set according to the user position type in the user identity information; wherein the user work log required data type set includes at least document type data, attendance type data and picture type data;

[0007] Generate a current screening condition according to the user work log required data type set and the work log data screening time interval, and obtain a target screening data set corresponding to the user identity information according to the current screening condition;

[0008] Acquire a first data subset of document type data in the target screening data set, perform keyword extraction on each document data included in the first data subset according to a preset keyword screening strategy, and obtain task data corresponding to each document data to form a first work log data subset;

[0009] Acquire a second data subset of the target screening data set, which is attendance type data, and merge the attendance data included in the second data subset in ascending order of date to obtain a second work log data subset;

[0010] Acquire a third data subset of picture type data in the target screening data set, and acquire an image recognition result set corresponding to the third data subset according to a pre-trained image recognition model to form a third work log data subset;

[0011] The first work log data subset, the second work log data subset and the third work log data subset are combined to obtain work log report data.

[0012] In a second aspect, an embodiment of the present application further provides a work log generation device based on artificial intelligence, and the work log generation device based on artificial intelligence includes:

[0013] A first acquisition module is used to generate a work log data screening time interval according to the current system time and the last work log generation time when it is determined that the time interval between the current system time and the last work log generation time is equal to a preset first time period, and to acquire user identity information;

[0014] A second acquisition module is used to acquire a user work log required data type set according to the user position type in the user identity information; wherein the user work log required data type set includes at least document type data, attendance type data and picture type data;

[0015] A data screening module, used to generate a current screening condition according to the user work log required data type set and the work log data screening time interval, and obtain a target screening data set corresponding to the user identity information according to the current screening condition;

[0016] A first extraction module is used to obtain a first data subset of document type data in the target screening data set, perform keyword extraction on each document data included in the first data subset according to a preset keyword screening strategy, and obtain task data corresponding to each document data to form a first work log data subset;

[0017] A second extraction module is used to obtain a second data subset of attendance type data in the target screening data set, and merge the attendance data included in the second data subset in ascending order of date to obtain a second work log data subset;

[0018] A third extraction module is used to obtain a third data subset of picture type data in the target screening data set, and obtain an image recognition result set corresponding to the third data subset according to a pre-trained image recognition model to form a third work log data subset;

[0019] The log report generating module is used to combine the first work log data subset, the second work log data subset and the third work log data subset to obtain work log report data.

[0020] In a third aspect, an embodiment of the present application further provides a processing device, including a processor and a memory, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps of any one of the artificial intelligence-based work log generation methods provided in the embodiments of the present application are executed.

[0021] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores a plurality of instructions, and the instructions are suitable for a processor to load to execute the steps in any one of the artificial intelligence-based work log generation methods provided in the embodiments of the present application.

[0022] From the above content, it can be concluded that the present application implements an artificial intelligence-based technology to automatically analyze the first data subset obtained as document type data to obtain the first work log data subset, obtain the second data subset obtained as attendance type data to obtain the second work log data subset, and obtain the third data subset obtained as picture type data to obtain the third work log data subset. Finally, the first work log data subset, the second work log data subset and the third work log data subset are automatically combined to obtain work log report data without manual sorting, thereby improving the efficiency of work log data acquisition and improving data accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 This is a flowchart of a method for generating a work log based on artificial intelligence in this application;

[0025] Figure 2 It is a structural schematic diagram of a work log generating device based on artificial intelligence in this application;

[0026] Figure 3 It is a structural schematic diagram of the processing equipment of this application. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0028] In the following description, the specific embodiments of the present application will be described with reference to the steps and symbols performed by one or more computers, unless otherwise stated. Therefore, these steps and operations will be mentioned several times as being performed by a computer, and the computer execution referred to in the embodiments of the present application includes the operation of a computer processing unit by an electronic signal representing data in a structured form. This operation converts the data or maintains it at a location in the memory system of the computer, which can be reconfigured or otherwise changed in a manner familiar to testers in the field. The data structure maintained by the data is the physical location of the memory, which has specific characteristics defined by the data format. However, the principle of the present application is described with the above text, which does not represent a limitation, and testers in the field will understand that the various steps and operations described below can also be implemented in hardware.

[0029] The principles of the present application may be operated using many other general or specific purpose computing, communication environments or configurations. Examples of well-known computing systems, environments and configurations suitable for use with the present application may include, but are not limited to, handheld phones, personal computers, servers, multiprocessor systems, microcomputer-based systems, mainframe computers, and distributed computing environments, including any of the above systems or devices.

[0030] The terms "first", "second", and "third" in this application are used to distinguish different objects rather than to describe a specific order. In addition, the terms "include", "have", and any variations thereof are intended to cover non-exclusive inclusions.

[0031] Next, we will begin to introduce the work log generation method based on artificial intelligence provided by this application.

[0032] See also Figure 1 , Figure 1 A flow chart of a method for generating a work log based on artificial intelligence is shown in the present application, and the method is applied to a cloud server. The method provided in the present application may specifically include the following steps:

[0033] S110. When it is determined that the time interval between the current system time and the last work log generation time is equal to a preset first time period, a work log data screening time interval is generated according to the current system time and the last work log generation time, and user identity information is obtained.

[0034] In the embodiment of the present application, the execution entities involved are cloud servers, first-type user terminals, second-type user terminals, and punch-in terminals. An office automation system is deployed on the cloud server, which can obtain the data uploaded by the first-type user terminals, the second-type user terminals, and the punch-in terminals for centralized processing to obtain work log report data. The first-type server is a portable mobile intelligent terminal (such as a smart phone, a laptop, a tablet computer, etc.), which can store a first data subset of document-type data and a third data subset of picture-type data, and regularly upload them to the cloud server for further data processing. The second-type server is a non-portable mobile intelligent terminal (such as a desktop computer, etc.), which can also store a first data subset of document-type data and a third data subset of picture-type data, and regularly upload them to the cloud server for further data processing. The punch-in terminal (also understood as an attendance terminal) is a face recognition punch-in device, a fingerprint recognition punch-in device, etc., which can collect the user's attendance data and upload it to the cloud server for further data processing.

[0035] This application describes the technical solution with the cloud server as the execution entity. Therefore, in order to automatically generate work log report data regularly in the cloud server, it is necessary to first determine whether the time interval between the current system time and the previous work log generation time (the previous work log generation time is such as 24:00 last Friday, or 24:00 at the end of last month, etc.) is equal to the preset first time period (for example, the first time period is set to 7 days, then the generated work log report data corresponds to the weekly work data report; for example, the first time period is set to 28, 29, 30 or 31 days, then the generated work log report data corresponds to the monthly work data report; of course, the first time period can also be set to one year or even a longer time value, that is, the first time period can be set according to the actual usage needs of the user).

[0036] For example, this application takes the first time period set to 7 days as an example. When it is determined that the time interval between the current system time and the last work log generation time is equal to the preset first time period, it means that the time point at which the work log report data can be generated has arrived. At this time, it is necessary to obtain which user generates the work log report data in this period, so it is necessary to generate the work log data screening time interval according to the current system time and the last work log generation time, and obtain the user identity information. For example, in the cloud server, exclusive storage areas are allocated for different users, and the data uploaded by each user based on the first type of user terminal, the second type of user terminal and the punch-in terminal is stored in the private storage space corresponding to the user. It should be noted that the acquisition, storage and application of personal information involved in this application are in compliance with the provisions of relevant laws, necessary confidentiality measures have been taken, and they do not violate public order and good customs. Moreover, the private storage space corresponding to each user is the user identity information of the known user. It can be seen that the acquisition of the initial work log data screening time interval and user identity information based on the cloud server can effectively generate data screening conditions to obtain data more accurately.

[0037] In one embodiment, step S110 includes:

[0038] The last work log generation time is used as the starting time point, the current system time is used as the ending time point, and the time period between the starting time point and the ending time point is used as the work log data screening time interval;

[0039] Obtain user login information corresponding to the current data storage area, and obtain user identity information according to the user information set included in the user login information; wherein the user information set includes at least user account data and user identity information; the user identity information includes at least user name, user gender, user age, user work number, user position type and position responsibility remarks information.

[0040] In an embodiment of the present application, for example, the first time period is set to 7 days, the current system time corresponds to 24:00 on XX2 January 20XX and corresponds to a Friday, and the previous work log generation time corresponds to 24:00 on XX4 March 20XX. If it is determined that the time interval between 24:00 on XX2 January 20XX and 24:00 on XX4 March 20XX is 7 days, then 24:00 on XX4 March 20XX is taken as the starting time point, and 24:00 on XX2 January 20XX is taken as the ending time point, and the time period between the starting time point and the ending time point is used as the work log data screening time interval, i.e., 24:00 on XX4 March 20XX to 24:00 on XX2 January 20XX.

[0041] After that, the work log report data generated by the cloud server in this period can be obtained, and the user login information corresponding to the current data storage area for extracting data from the cloud server can be obtained, and the user identity information can be obtained based on the user information set included in the user login information. Specifically, the user information set includes at least user account data (for example, a character string corresponding to the user account, i.e., user nickname and other information) and user identity information, and the user identity information includes at least user name, user gender, user age, user work number, user position type and position responsibility notes; wherein, the user position type includes the first position type corresponding to the business position type (generally understood as the role of a salesperson or salesperson in an enterprise, and the main responsibility of such employees is to sell products), and also includes the second position type corresponding to the technical support position type (generally understood as the role of a research and development personnel in an enterprise, and the main responsibility of such employees is to develop research and development projects), and the main work task items of the position type are described in the position responsibility notes corresponding to each user position type. It can be seen that through the above method, data screening conditions can be quickly generated.

[0042] S120. Obtain a user work log required data type set according to the user position type in the user identity information; wherein the user work log required data type set includes at least document type data, attendance type data and picture type data.

[0043] In the embodiments of the present application, since the job responsibilities and notes information corresponding to each user position type is pre-stored in the cloud server, the user work log demand data type set can be obtained based on the user position type, specifically, the user work log demand data type set is composed of user daily work editing documents (such as project requirements, etc.), contract documents, attendance data, work photos and other work log demand data types. It can be understood that the user's daily work editing documents and contract documents are document type data, attendance data is attendance type data, and work photos are picture type data. It can be seen that by generating a corresponding user work log demand data type set based on the user position type, the required data can be filtered out from the user data stored in the cloud server.

[0044] In one embodiment, step S120 includes:

[0045] When it is determined that the user position type is a first position type, obtaining a user work log required data type set corresponding to the first position type; wherein the first position type is a business position type;

[0046] When it is determined that the user position type is the second position type, a user work log requirement data type set corresponding to the second position type is obtained; wherein the second position type is a technical support position type.

[0047] In an embodiment of the present application, when it is determined that the user position type is the first position type corresponding to the business position type, the user work log demand data types included in the general user work log demand data type set include contract documents, work photos, and attendance data; when it is determined that the user position type is the second position type corresponding to the technical support position type, the user work log demand data types included in the general user work log demand data type set include user daily work editing documents, work photos, and attendance data. It can be seen that in the cloud server, different user work log demand data type sets are generated based on users of different position types, and data can be more accurately filtered for processing based on the user's position type, without processing all data, which can effectively improve data processing efficiency.

[0048] S130: Generate a current screening condition according to the user work log required data type set and the work log data screening time interval, and obtain a target screening data set corresponding to the user identity information according to the current screening condition.

[0049] In an embodiment of the present application, when the target user targeted by the user identity information (such as recording it as user A) obtains the corresponding user work log required data type set and the work log data screening time interval, they are combined together to generate the current screening condition. Then, based on user A, data retrieval is performed in the corresponding exclusive storage area of ​​the cloud server with the current screening condition to obtain the target screening data set of user A. Among them, the target screening data set includes a first data subset of document type data, a second data subset of attendance type data, and a third data subset of picture type data. It can be seen that the current screening conditions composed of time and user feature dimensions can more accurately screen the user's target data to assist in generating work log report data.

[0050] S140. Obtain a first data subset of document type data from the target screening data set, perform keyword extraction on each document data included in the first data subset according to a preset keyword screening strategy, and obtain task data corresponding to each document data to form a first work log data subset.

[0051] In an embodiment of the present application, since the target screening data set includes at least a first data subset of document type data, a second data subset of attendance type data and a third data subset of picture type data, at this time, the first data subset of document type data can be first subjected to keyword extraction according to a preset keyword screening strategy to obtain task data corresponding to each document data to form a first work log data subset.

[0052] For example, when user A uses the first type of user terminal to work in an office or an off-site location (such as the location of the visited customer), or when user A uses the second type of user terminal to work in an office, the document type data edited can be understood as a cloud document editing data, which can be synchronized with the current data storage area in the cloud server (that is, the exclusive storage area corresponding to user A) in a timely manner, that is, the specific content of the document, the document name, the document modification time (the document modification time refers to the time when a document was last modified and saved) and other data of the document type data edited by the user using the first type of user terminal or the second type of user terminal can be uploaded to the cloud server synchronously. In this way, the first data subset in the target screening data set corresponding to user A in this week can be used as a data sample to quickly generate the document work content of user A in this week.

[0053] Since the specific content of each document does not need to be reflected when generating the document work content of user A this week, in order to quickly extract the document work content of user A this week, the document data included in the first data subset can be subjected to keyword extraction according to a preset keyword screening strategy, and the task data corresponding to each document data is obtained to form the first work log data subset. The keyword screening strategy is used to extract the Chinese document name of the document data to form the document task data, or to extract several core keywords of the document content to form the document task data.

[0054] In one embodiment, step S140 includes:

[0055] Obtaining the i-th document data in the first data subset, and obtaining the document modification time of the i-th document data; wherein the initial value of i is 1, and the value range of i is [1, N1], and N1 represents the total number of document data included in the first data subset;

[0056] If it is determined that the document name of the i-th document data is a Chinese name, the document name of the i-th document data is obtained as the keyword of the i-th document data according to the keyword screening strategy, and the i-th task data is composed of the keyword of the i-th document data and the document modification time of the i-th document data;

[0057] If it is determined that the document name of the i-th document data is a digital name, the body text of the i-th document data is obtained, and the i-th core keyword combination corresponding to the body text of the i-th document data is obtained according to the keyword screening strategy, and the i-th task data is composed of the i-th core keyword combination and the document modification time of the i-th document data;

[0058] Update the value of i by adding 1 to the value of i;

[0059] If it is determined that i does not exceed N1, returning to the step of obtaining the i-th document data in the first data subset and obtaining the document modification time of the i-th document data;

[0060] If it is determined that i exceeds N1, task data No. 1 to task data No. N1 are obtained to form a first work log data subset.

[0061] In the embodiment of the present application, the complete acquisition process of task data No. 1 is taken as an example to illustrate. Specifically, the first document data No. 1 in the first data subset is first acquired, and the document modification time of the document data No. 1 is acquired (for example, continuing to refer to the above example, the document modification time should be between 24:00 on XX3 / XX4 / 20XX and 24:00 on XX1 / XX2 / 20XX).

[0062] Then, it is determined whether the document name of the No. 1 document data is a Chinese name (such as XXX project specification) or a numerical name (such as 12). If it is determined that the document name of the No. 1 document data is a Chinese name, it means that user A has standardized the name of the document based on the preset task specification, so the document name of the No. 1 document data can be directly used as the keyword of the No. 1 document data based on the keyword screening strategy, without extracting the content of the specific document of the No. 1 document data to extract the keyword, and directly obtaining the Chinese name of the No. 1 document data can be more efficient and accurate.

[0063] If it is determined that the document name of the No. 1 document data is a digital name, it means that user A did not name the document based on the preset task specification. If the document name of the No. 1 document data is directly used as the keyword of the No. 1 document data, the task content cannot be intuitively reflected. At this time, the No. 1 core keyword combination corresponding to the body text of the No. 1 document data is obtained according to the keyword screening strategy, and the No. 1 core keyword combination and the document modification time of the No. 1 document data form the No. 1 task data. Wherein, when obtaining the No. 1 core keyword combination corresponding to the body text of the No. 1 document data according to the keyword screening strategy, the body text of the No. 1 document data can be sequentially subjected to word segmentation processing, keyword extraction processing based on the TF-IDF model (i.e., word frequency-inverse text frequency index model), and the top 3 keywords with TF-IDF values ​​are selected from the keyword set according to the preset keyword screening number (such as set to 3) to form the No. 1 core keyword combination corresponding to the body text of the No. 1 document data, so that the No. 1 core keyword combination and the document modification time of the No. 1 document data form the No. 1 task data. Or when obtaining the No. 1 core keyword combination corresponding to the body text of the No. 1 document data according to the keyword screening strategy, the text summary (generally within 20 words) of the body text of the No. 1 document data can be directly extracted based on the LDA model (i.e., implicit Dirichlet distribution model, which is also a topic model) as the No. 1 core keyword combination, and finally the No. 1 core keyword combination and the document modification time of the No. 1 document data form the No. 1 task data. By analogy, continue to obtain the No. 2 task data of the No. 2 document data in the first data subset, until the No. N1 task data of the No. N1 document data is obtained, thereby forming the first work log data subset.

[0064] S150, obtaining a second data subset of the target screening data set, which is attendance type data, and merging the attendance data included in the second data subset in ascending order of date to obtain a second work log data subset.

[0065] In the embodiment of the present application, the attendance type data of user A stored in the general cloud server is obtained by the punch-in terminal and uploaded to the current data storage area corresponding to user A in the cloud server, and the time of the punch-in data corresponding to the obtained second data subset is between 24:00 on March XX4, 20XX and 24:00 on January XX2, 20XX. At this time, the attendance data included in the second data subset can be merged in ascending order of date to form a second work log data subset composed of attendance data. It can be seen that based on this method, the target attendance data can be quickly extracted.

[0066] In one embodiment, step S150 includes:

[0067] Obtain the j-th attendance data subset in the second data subset; wherein the initial value of j is 1, and the value range of j is [1, N2], and N2 represents the total number of days in the first time period;

[0068] Based on a preset daily working time period, obtaining the j-day attendance mark corresponding to the j-day attendance data subset; wherein the j-day attendance mark is one of a normal attendance mark and an abnormal attendance mark;

[0069] Combine the j-th day's on-duty attendance time point and the j-th day's off-duty attendance time point of the j-th day's attendance data subset with the j-th day's attendance identifier to obtain the j-th attendance data;

[0070] Update the value of j by adding 1 to it.

[0071] If it is determined that j does not exceed N2, return to the step of obtaining the j-th attendance data subset in the second data subset;

[0072] If it is determined that j exceeds N2, the attendance data No. 1 to No. N2 are obtained to form a second work log data subset.

[0073] In the embodiment of the present application, the complete acquisition process of the first attendance data is taken as an example to illustrate. Specifically, the first-day attendance data subset in the second data subset (which includes the first-day attendance time point and the first-day off-duty attendance time point) is first acquired, and then the first-day attendance identifier corresponding to the first-day attendance data subset is acquired based on the preset daily working time period. The preset daily working time period is 9:00-18:00. When obtaining the attendance mark of the first day, it is specifically determined whether the attendance time point of the first day is before 9 o'clock and whether the attendance time point of the first day is after 18:00. If it is determined that the attendance time point of the first day is before 9 o'clock and the attendance time point of the first day is after 18:00, the attendance mark of the first day corresponding to the attendance data subset of the first day is determined to be a normal attendance mark; if it is determined that the attendance time point of the first day is after 9 o'clock or the attendance time point of the first day is before 18:00, the attendance mark of the first day corresponding to the attendance data subset of the first day is determined to be an abnormal attendance mark. Since the attendance time point of the 1st day, the attendance time point of the 1st day and the attendance mark of the 1st day corresponding to the attendance data subset of the 1st day are known at this time, the attendance time point of the 1st day, the attendance time point of the 1st day and the attendance mark of the 1st day corresponding to the attendance data subset of the 1st day can be combined to obtain the attendance data No. 1. By analogy, continue to obtain the attendance data No. 2 of the attendance data subset of the 2nd day in the second data subset, until the attendance data No. N2 of the attendance data subset of the N2th day is obtained, thereby forming the second work log data subset. It can be seen that based on this method of obtaining attendance data day by day, it is convenient to display the attendance data more clearly later.

[0074] S160. Obtain a third data subset of picture type data in the target screening data set, and obtain an image recognition result set corresponding to the third data subset according to a pre-trained image recognition model to form a third work log data subset.

[0075] In the embodiment of the present application, the picture type data of user A stored in the cloud server generally refers to the work photos collected by the user through the first type of user terminal when the user is out visiting customers (taking photos with customers or taking photos of the front desk of the customer company, etc.) or visiting some products that can be authorized to be photographed, and uploaded to the current data storage area corresponding to user A in the cloud server, and the shooting time of the picture type data corresponding to the third data subset obtained is between 24:00 on March XX4, 20XX and 24:00 on January XX2, 20XX. At this time, the picture recognition results of each picture included in the third data subset can be extracted based on a pre-trained image recognition model (such as a convolutional neural network, a target detection network, and other image recognition models) to form a third work log data subset. It can be seen that based on this method, the specific work items of the work pictures can be quickly extracted.

[0076] In one embodiment, step S160 includes:

[0077] Obtaining the kth image in the third data subset; wherein the initial value of k is 1, and the value range of k is [1, N3], and N3 represents the total number of images in the third data subset;

[0078] Acquire the image recognition type of the kth image based on the image recognition model; wherein the image recognition type is one of a location picture type, a person picture type, a document picture type, or an object picture type;

[0079] If it is determined that the image recognition type of the k-th image is a location picture type, obtaining the shooting location information and shooting time of the k-th image, and forming the k-th picture information by the shooting location information and shooting time of the k-th image;

[0080] If it is determined that the image recognition type of the k-th image is a person picture type, obtaining the person unique number and shooting time of the k-th image, and forming the k-th picture information by the person unique number and shooting time of the k-th image;

[0081] If it is determined that the image recognition type of the k-th image is a document image type, obtaining the text recognition result and shooting time of the k-th image, and forming the k-th image information by the text recognition result and shooting time of the k-th image;

[0082] If it is determined that the image recognition type of the k-th image is an object picture type, obtaining the object recognition result and shooting time of the k-th image, and forming the k-th picture information by the object recognition result and shooting time of the k-th image;

[0083] Update the value of k by adding 1 to the value of k;

[0084] If it is determined that k does not exceed N3, returning to the step of obtaining the kth image in the third data subset;

[0085] If it is determined that j exceeds N3, the picture information No. 1 to the picture information No. N3 are obtained to form a third work log data subset.

[0086] In the embodiment of the present application, the complete acquisition process of the first image is taken as an example for explanation. Specifically, the first image in the third data subset is first acquired, and then the image type and specific image recognition result of the first image are acquired based on the image recognition model, so as to realize the extraction of work content from the work image. Since the image recognition type of the first image is one of the location image type, the person image type, the document image type or the thing image type, if it is determined that the image recognition type of the first image is the location image type (generally, if it is detected that there are no human bodies, documents, products and other categories in the image, it can be determined as a location image type), the shooting location information and shooting time of the first image are obtained, and the shooting location information and shooting time of the first image constitute the first picture information; wherein, when the first type of user end with a camera generally takes a photo to obtain the taken photo, the image attribute data (such as including shooting time, shooting location and other information) is known, and the shooting location information and shooting time can be quickly obtained by reading the image attribute data.

[0087] If it is determined that the image recognition type of the first image is a person picture type (generally, if a human body is detected in the picture, and there are no location scenes, documents, or products, it can be determined to be a person picture type), the person unique number and shooting time of the first image are obtained, and the first picture information is composed of the person unique number and shooting time of the first image. When generating the person unique number of the first image, the person unique number can be composed of the date corresponding to the shooting time plus the shooting sequence number of the current photo (for example, the first image is the first photo taken today), so that the person photo No. 01 on March XX5, 20XX is used as the person unique number, and then combined with the shooting time of March XX5, 20XX HH1:MM1:SS1 to obtain the first picture information.

[0088] If it is determined that the person picture type of the first image is a document picture type (generally, if it is detected that there is a document in the picture, but no human body, real location scene, or product exists, it can be determined to be a document picture type), the text recognition result of the first image is obtained based on the image recognition model (such as the convolutional recurrent neural network model, also known as the CRNN model), and the shooting time is obtained from the picture attribute data. The text recognition result of the first image and the shooting time constitute the first picture information.

[0089] If the image recognition type of the first image is determined to be a physical image type (generally, if a physical product is detected in the image, and there are no human bodies, documents, or real scene locations, it can be determined to be a physical image type), the object recognition result of the first image is obtained based on the image recognition model (such as the target detection network included therein, more specifically, the single-lens multi-box detector model, i.e., the SSD model) (e.g., the object recognition result is product 1), and the shooting time is obtained from the image attribute data. The object recognition result of the first image and the shooting time form the first image information. It can be seen that based on the specific identification of the work photo type and the specific photo content, it is convenient to more clearly display other non-text editing work items in the future.

[0090] S170: Combine the first work log data subset, the second work log data subset and the third work log data subset to obtain work log report data.

[0091] In an embodiment of the present application, after the first work log data subset, the second work log data subset and the third work log data subset of the user (such as user A in the above example) are obtained in the cloud server, the first work log data subset, the second work log data subset and the third work log data subset are automatically combined to obtain work log report data. It can be seen that it is obtained after analyzing and summarizing various types of data based on the user, without the need for the user to collect work materials and then organize them, which improves data processing efficiency.

[0092] In one embodiment, step S170 includes:

[0093] Acquire image information of the document image type in the first work log data subset and the third work log data subset to form a work content item report data set;

[0094] Acquire the picture information of the location picture type in the second work log data subset and the third work log data subset to form an attendance item report data set;

[0095] Acquire picture information of a person picture type in the third work log data subset, and acquire picture information of a thing picture type in the third work log data subset to form an auxiliary work content item report data set;

[0096] The work content item report data set, the attendance item report data set and the auxiliary work content item report data set are combined to obtain work log report data.

[0097] In an embodiment of the present application, since the work pictures included in the third work log data subset may be identified as pictures of customer visits, pictures of real objects, pictures of documents (such as minutes of a meeting) or pictures of real-life locations of the user, after further analyzing the work pictures included in the third work log data subset, the picture information of the location picture type in the third work log data subset may be further combined with the second work log data subset as an auxiliary attendance data to obtain an attendance item report data set; the picture information of the person picture type and the picture information of the thing picture type in the third work log data subset are combined to form an auxiliary work content item report data set; the picture information of the document picture type in the third work log data subset is combined with the first work log data subset to form a work content item report data set. Finally, the work content item report data set, the attendance item report data set and the auxiliary work content item report data set are combined, more specifically, filled into the corresponding filling areas in the preset work log report template (there are work content item report data set filling areas, attendance item report data set filling areas and auxiliary work content item report data set filling areas in the work log report template) to obtain work log report data. It can be seen that based on this method of automatically filling and generating report data, there is no need for manual sorting, and not only are there more data dimensions, but the efficiency is also higher.

[0098] The method realizes automatic combination of the first work log data subset, the second work log data subset and the third work log data subset to obtain work log report data without manual sorting, thereby improving the efficiency of work log data acquisition and improving data accuracy.

[0099] In order to facilitate better implementation of the method of the present application, the embodiment of the present application also provides a work log generating device 100 based on artificial intelligence.

[0100] See also Figure 2 , Figure 2 This is a structural schematic diagram of the work log generation device 100 based on artificial intelligence in the present application, wherein the work log generation device 100 based on artificial intelligence may specifically include the following structures: a first acquisition module 110, a second acquisition module 120, a data screening module 130, a first extraction module 140, a second extraction module 150, a third extraction module 160 and a log report generation module 170.

[0101] Among them, the first acquisition module 110 is used to generate a work log data screening time interval according to the current system time and the last work log generation time when it is determined that the time interval between the current system time and the last work log generation time is equal to a preset first time period, and obtain user identity information.

[0102] In the embodiment of the present application, the terminals involved are cloud servers, first-type user terminals, second-type user terminals, and punch-in terminals. An office automation system is deployed on the cloud server, which can obtain the data uploaded by the first-type user terminals, the second-type user terminals, and the punch-in terminals for centralized processing to obtain work log report data. The first-type server is a portable mobile intelligent terminal (such as a smart phone, a laptop, a tablet computer, etc.), which can store a first data subset of document-type data and a third data subset of picture-type data, and regularly upload them to the cloud server for further data processing. The second-type server is a non-portable mobile intelligent terminal (such as a desktop computer, etc.), which can also store a first data subset of document-type data and a third data subset of picture-type data, and regularly upload them to the cloud server for further data processing. The punch-in terminal (also understood as an attendance terminal) is a face recognition punch-in device, a fingerprint recognition punch-in device, etc., which can collect the user's attendance data and upload it to the cloud server for further data processing.

[0103] This application describes the technical solution with the cloud server as the execution entity. Therefore, in order to automatically generate work log report data regularly in the cloud server, it is necessary to first determine whether the time interval between the current system time and the previous work log generation time (the previous work log generation time is such as 24:00 last Friday, or 24:00 at the end of last month, etc.) is equal to the preset first time period (for example, the first time period is set to 7 days, then the generated work log report data corresponds to the weekly work data report; for example, the first time period is set to 28, 29, 30 or 31 days, then the generated work log report data corresponds to the monthly work data report; of course, the first time period can also be set to one year or even a longer time value, that is, the first time period can be set according to the actual usage needs of the user).

[0104] For example, this application takes the first time period set to 7 days as an example. When it is determined that the time interval between the current system time and the last work log generation time is equal to the preset first time period, it means that the time point at which the work log report data can be generated has arrived. At this time, it is necessary to obtain which user generates the work log report data in this period, so it is necessary to generate the work log data screening time interval according to the current system time and the last work log generation time, and obtain the user identity information. For example, in the cloud server, exclusive storage areas are allocated for different users, and the data uploaded by each user based on the first type of user terminal, the second type of user terminal and the punch-in terminal is stored in the private storage space corresponding to the user. It should be noted that the acquisition, storage and application of personal information involved in this application are in compliance with the provisions of relevant laws, necessary confidentiality measures have been taken, and they do not violate public order and good customs. Moreover, the private storage space corresponding to each user is the user identity information of the known user. It can be seen that the acquisition of the initial work log data screening time interval and user identity information based on the cloud server can effectively generate data screening conditions to obtain data more accurately.

[0105] In one embodiment, the first acquisition module 110 is specifically used for:

[0106] The last work log generation time is used as the starting time point, the current system time is used as the ending time point, and the time period between the starting time point and the ending time point is used as the work log data screening time interval;

[0107] Obtain user login information corresponding to the current data storage area, and obtain user identity information according to the user information set included in the user login information; wherein the user information set includes at least user account data and user identity information; the user identity information includes at least user name, user gender, user age, user work number, user position type and position responsibility remarks information.

[0108] In an embodiment of the present application, for example, the first time period is set to 7 days, the current system time corresponds to 24:00 on XX2 January 20XX and corresponds to a Friday, and the previous work log generation time corresponds to 24:00 on XX4 March 20XX. If it is determined that the time interval between 24:00 on XX2 January 20XX and 24:00 on XX4 March 20XX is 7 days, then 24:00 on XX4 March 20XX is taken as the starting time point, and 24:00 on XX2 January 20XX is taken as the ending time point, and the time period between the starting time point and the ending time point is used as the work log data screening time interval, i.e., 24:00 on XX4 March 20XX to 24:00 on XX2 January 20XX.

[0109] After that, the work log report data generated by the cloud server in this period can be obtained, and the user login information corresponding to the current data storage area for extracting data from the cloud server can be obtained, and the user identity information can be obtained based on the user information set included in the user login information. Specifically, the user information set includes at least user account data (for example, a character string corresponding to the user account, i.e., user nickname and other information) and user identity information, and the user identity information includes at least user name, user gender, user age, user work number, user position type and position responsibility notes; wherein, the user position type includes the first position type corresponding to the business position type (generally understood as the role of a salesperson or salesperson in an enterprise, and the main responsibility of such employees is to sell products), and also includes the second position type corresponding to the technical support position type (generally understood as the role of a research and development personnel in an enterprise, and the main responsibility of such employees is to develop research and development projects), and the main work task items of the position type are described in the position responsibility notes corresponding to each user position type. It can be seen that through the above method, data screening conditions can be quickly generated.

[0110] The second acquisition module 120 is used to acquire a user work log required data type set according to the user position type in the user identity information; wherein the user work log required data type set includes at least document type data, attendance type data and picture type data.

[0111] In the embodiments of the present application, since the job responsibilities and notes information corresponding to each user position type is pre-stored in the cloud server, the user work log demand data type set can be obtained based on the user position type, specifically, the user work log demand data type set is composed of user daily work editing documents (such as project requirements, etc.), contract documents, attendance data, work photos and other work log demand data types. It can be understood that the user's daily work editing documents and contract documents are document type data, attendance data is attendance type data, and work photos are picture type data. It can be seen that by generating a corresponding user work log demand data type set based on the user position type, the required data can be filtered out from the user data stored in the cloud server.

[0112] In one embodiment, the second acquisition module 120 is specifically used for:

[0113] When it is determined that the user position type is a first position type, obtaining a user work log required data type set corresponding to the first position type; wherein the first position type is a business position type;

[0114] When it is determined that the user position type is the second position type, a user work log requirement data type set corresponding to the second position type is obtained; wherein the second position type is a technical support position type.

[0115] In an embodiment of the present application, when it is determined that the user position type is the first position type corresponding to the business position type, the user work log demand data types included in the general user work log demand data type set include contract documents, work photos, and attendance data; when it is determined that the user position type is the second position type corresponding to the technical support position type, the user work log demand data types included in the general user work log demand data type set include user daily work editing documents, work photos, and attendance data. It can be seen that in the cloud server, different user work log demand data type sets are generated based on users of different position types, and data can be more accurately filtered for processing based on the user's position type, without processing all data, which can effectively improve data processing efficiency.

[0116] The data screening module 130 is used to generate a current screening condition according to the user work log required data type set and the work log data screening time interval, and obtain a target screening data set corresponding to the user identity information according to the current screening condition.

[0117] In an embodiment of the present application, when the target user targeted by the user identity information (such as recording it as user A) obtains the corresponding user work log required data type set and the work log data screening time interval, they are combined together to generate the current screening condition. Then, based on user A, data retrieval is performed in the corresponding exclusive storage area of ​​the cloud server with the current screening condition to obtain the target screening data set of user A. Among them, the target screening data set includes a first data subset of document type data, a second data subset of attendance type data, and a third data subset of picture type data. It can be seen that the current screening conditions composed of time and user feature dimensions can more accurately screen the user's target data to assist in generating work log report data.

[0118] The first extraction module 140 is used to obtain a first data subset of document type data in the target screening data set, perform keyword extraction on each document data included in the first data subset according to a preset keyword screening strategy, and obtain task data corresponding to each document data to form a first work log data subset.

[0119] In an embodiment of the present application, since the target screening data set includes at least a first data subset of document type data, a second data subset of attendance type data and a third data subset of picture type data, at this time, the first data subset of document type data can be first subjected to keyword extraction according to a preset keyword screening strategy to obtain task data corresponding to each document data to form a first work log data subset.

[0120] For example, when user A uses the first type of user terminal to work in an office or an off-site location (such as the location of the visited customer), or when user A uses the second type of user terminal to work in an office, the document type data edited can be understood as a cloud document editing data, which can be synchronized with the current data storage area in the cloud server (that is, the exclusive storage area corresponding to user A) in a timely manner, that is, the specific content of the document, the document name, the document modification time (the document modification time refers to the time when a document was last modified and saved) and other data of the document type data edited by the user using the first type of user terminal or the second type of user terminal can be uploaded to the cloud server synchronously. In this way, the first data subset in the target screening data set corresponding to user A in this week can be used as a data sample to quickly generate the document work content of user A in this week.

[0121] Since the specific content of each document does not need to be reflected when generating the document work content of user A this week, in order to quickly extract the document work content of user A this week, the document data included in the first data subset can be subjected to keyword extraction according to a preset keyword screening strategy, and the task data corresponding to each document data is obtained to form the first work log data subset. The keyword screening strategy is used to extract the Chinese document name of the document data to form the document task data, or to extract several core keywords of the document content to form the document task data.

[0122] In one embodiment, the first extraction module 140 is specifically used for:

[0123] Obtaining the i-th document data in the first data subset, and obtaining the document modification time of the i-th document data; wherein the initial value of i is 1, and the value range of i is [1, N1], and N1 represents the total number of document data included in the first data subset;

[0124] If it is determined that the document name of the i-th document data is a Chinese name, the document name of the i-th document data is obtained as the keyword of the i-th document data according to the keyword screening strategy, and the i-th task data is composed of the keyword of the i-th document data and the document modification time of the i-th document data;

[0125] If it is determined that the document name of the i-th document data is a digital name, the body text of the i-th document data is obtained, and the i-th core keyword combination corresponding to the body text of the i-th document data is obtained according to the keyword screening strategy, and the i-th task data is composed of the i-th core keyword combination and the document modification time of the i-th document data;

[0126] Update the value of i by adding 1 to the value of i;

[0127] If it is determined that i does not exceed N1, returning to the step of obtaining the i-th document data in the first data subset and obtaining the document modification time of the i-th document data;

[0128] If it is determined that i exceeds N1, task data No. 1 to task data No. N1 are obtained to form a first work log data subset.

[0129] In the embodiment of the present application, the complete acquisition process of task data No. 1 is taken as an example to illustrate. Specifically, the first document data No. 1 in the first data subset is first acquired, and the document modification time of the document data No. 1 is acquired (for example, continuing to refer to the above example, the document modification time should be between 24:00 on XX3 / XX4 / 20XX and 24:00 on XX1 / XX2 / 20XX).

[0130] Then, it is determined whether the document name of the No. 1 document data is a Chinese name (such as XXX project specification) or a numerical name (such as 12). If it is determined that the document name of the No. 1 document data is a Chinese name, it means that user A has standardized the name of the document based on the preset task specification, so the document name of the No. 1 document data can be directly used as the keyword of the No. 1 document data based on the keyword screening strategy, without extracting the content of the specific document of the No. 1 document data to extract the keyword, and directly obtaining the Chinese name of the No. 1 document data can be more efficient and accurate.

[0131] If it is determined that the document name of the No. 1 document data is a digital name, it means that user A did not name the document based on the preset task specification. If the document name of the No. 1 document data is directly used as the keyword of the No. 1 document data, the task content cannot be intuitively reflected. At this time, the No. 1 core keyword combination corresponding to the body text of the No. 1 document data is obtained according to the keyword screening strategy, and the No. 1 core keyword combination and the document modification time of the No. 1 document data form the No. 1 task data. Wherein, when obtaining the No. 1 core keyword combination corresponding to the body text of the No. 1 document data according to the keyword screening strategy, the body text of the No. 1 document data can be sequentially subjected to word segmentation processing, keyword extraction processing based on the TF-IDF model (i.e., word frequency-inverse text frequency index model), and the top 3 keywords with TF-IDF values ​​are selected from the keyword set according to the preset keyword screening number (such as set to 3) to form the No. 1 core keyword combination corresponding to the body text of the No. 1 document data, so that the No. 1 core keyword combination and the document modification time of the No. 1 document data form the No. 1 task data. Or when obtaining the No. 1 core keyword combination corresponding to the body text of the No. 1 document data according to the keyword screening strategy, the text summary (generally within 20 words) of the body text of the No. 1 document data can be directly extracted based on the LDA model (i.e., implicit Dirichlet distribution model, which is also a topic model) as the No. 1 core keyword combination, and finally the No. 1 core keyword combination and the document modification time of the No. 1 document data form the No. 1 task data. By analogy, continue to obtain the No. 2 task data of the No. 2 document data in the first data subset, until the No. N1 task data of the No. N1 document data is obtained, thereby forming the first work log data subset.

[0132] The second extraction module 150 is used to obtain a second data subset of the target screening data set, which is attendance type data, and merge the attendance data included in the second data subset in ascending order of date to obtain a second work log data subset.

[0133] In the embodiment of the present application, the attendance type data of user A stored in the general cloud server is obtained by the punch-in terminal and uploaded to the current data storage area corresponding to user A in the cloud server, and the time of the punch-in data corresponding to the obtained second data subset is between 24:00 on March XX4, 20XX and 24:00 on January XX2, 20XX. At this time, the attendance data included in the second data subset can be merged in ascending order of date to form a second work log data subset composed of attendance data. It can be seen that based on this method, the target attendance data can be quickly extracted.

[0134] In one embodiment, the second extraction module 150 is specifically used for:

[0135] Obtain the j-th attendance data subset in the second data subset; wherein the initial value of j is 1, and the value range of j is [1, N2], and N2 represents the total number of days in the first time period;

[0136] Based on a preset daily working time period, obtaining the j-day attendance mark corresponding to the j-day attendance data subset; wherein the j-day attendance mark is one of a normal attendance mark and an abnormal attendance mark;

[0137] Combine the j-th day's on-duty attendance time point and the j-th day's off-duty attendance time point of the j-th day's attendance data subset with the j-th day's attendance identifier to obtain the j-th attendance data;

[0138] Update the value of j by adding 1 to it.

[0139] If it is determined that j does not exceed N2, return to the step of obtaining the j-th attendance data subset in the second data subset;

[0140] If it is determined that j exceeds N2, the attendance data No. 1 to No. N2 are obtained to form a second work log data subset.

[0141] In the embodiment of the present application, the complete acquisition process of the first attendance data is taken as an example to illustrate. Specifically, the first-day attendance data subset in the second data subset (which includes the first-day attendance time point and the first-day off-duty attendance time point) is first acquired, and then the first-day attendance identifier corresponding to the first-day attendance data subset is acquired based on the preset daily working time period. The preset daily working time period is 9:00-18:00. When obtaining the attendance mark of the first day, it is specifically determined whether the attendance time point of the first day is before 9 o'clock and whether the attendance time point of the first day is after 18:00. If it is determined that the attendance time point of the first day is before 9 o'clock and the attendance time point of the first day is after 18:00, the attendance mark of the first day corresponding to the attendance data subset of the first day is determined to be a normal attendance mark; if it is determined that the attendance time point of the first day is after 9 o'clock or the attendance time point of the first day is before 18:00, the attendance mark of the first day corresponding to the attendance data subset of the first day is determined to be an abnormal attendance mark. Since the attendance time point of the 1st day, the attendance time point of the 1st day and the attendance mark of the 1st day corresponding to the attendance data subset of the 1st day are known at this time, the attendance time point of the 1st day, the attendance time point of the 1st day and the attendance mark of the 1st day corresponding to the attendance data subset of the 1st day can be combined to obtain the attendance data No. 1. By analogy, continue to obtain the attendance data No. 2 of the attendance data subset of the 2nd day in the second data subset, until the attendance data No. N2 of the attendance data subset of the N2th day is obtained, thereby forming the second work log data subset. It can be seen that based on this method of obtaining attendance data day by day, it is convenient to display the attendance data more clearly later.

[0142] The third extraction module 160 is used to obtain a third data subset of picture type data in the target screening data set, and obtain an image recognition result set corresponding to the third data subset according to a pre-trained image recognition model to form a third work log data subset.

[0143] In the embodiment of the present application, the picture type data of user A stored in the cloud server generally refers to the work photos collected by the user through the first type of user terminal when the user is out visiting customers (taking photos with customers or taking photos of the front desk of the customer company, etc.) or visiting some products that can be authorized to be photographed, and uploaded to the current data storage area corresponding to user A in the cloud server, and the shooting time of the picture type data corresponding to the third data subset obtained is between 24:00 on March XX4, 20XX and 24:00 on January XX2, 20XX. At this time, the picture recognition results of each picture included in the third data subset can be extracted based on a pre-trained image recognition model (such as a convolutional neural network, a target detection network, and other image recognition models) to form a third work log data subset. It can be seen that based on this method, the specific work items of the work pictures can be quickly extracted.

[0144] In one embodiment, the third extraction module 160 is specifically used for:

[0145] Obtaining the kth image in the third data subset; wherein the initial value of k is 1, and the value range of k is [1, N3], and N3 represents the total number of images in the third data subset;

[0146] Acquire the image recognition type of the kth image based on the image recognition model; wherein the image recognition type is one of a location picture type, a person picture type, a document picture type, or an object picture type;

[0147] If it is determined that the image recognition type of the k-th image is a location picture type, obtaining the shooting location information and shooting time of the k-th image, and forming the k-th picture information by the shooting location information and shooting time of the k-th image;

[0148] If it is determined that the image recognition type of the k-th image is a person picture type, obtaining the person unique number and shooting time of the k-th image, and forming the k-th picture information by the person unique number and shooting time of the k-th image;

[0149] If it is determined that the image recognition type of the k-th image is a document image type, obtaining the text recognition result and shooting time of the k-th image, and forming the k-th image information by the text recognition result and shooting time of the k-th image;

[0150] If it is determined that the image recognition type of the k-th image is an object picture type, obtaining the object recognition result and shooting time of the k-th image, and forming the k-th picture information by the object recognition result and shooting time of the k-th image;

[0151] Update the value of k by adding 1 to the value of k;

[0152] If it is determined that k does not exceed N3, returning to the step of obtaining the kth image in the third data subset;

[0153] If it is determined that j exceeds N3, the picture information No. 1 to the picture information No. N3 are obtained to form a third work log data subset.

[0154] In the embodiment of the present application, the complete acquisition process of the first image is taken as an example for explanation. Specifically, the first image in the third data subset is first acquired, and then the image type and specific image recognition result of the first image are acquired based on the image recognition model, so as to realize the extraction of work content from the work image. Since the image recognition type of the first image is one of the location image type, the person image type, the document image type or the thing image type, if it is determined that the image recognition type of the first image is the location image type (generally, if it is detected that there are no human bodies, documents, products and other categories in the image, it can be determined as a location image type), the shooting location information and shooting time of the first image are obtained, and the shooting location information and shooting time of the first image constitute the first picture information; wherein, when the first type of user end with a camera generally takes a photo to obtain the taken photo, the image attribute data (such as including shooting time, shooting location and other information) is known, and the shooting location information and shooting time can be quickly obtained by reading the image attribute data.

[0155] If it is determined that the image recognition type of the first image is a person picture type (generally, if a human body is detected in the picture, and there are no location scenes, documents, or products, it can be determined to be a person picture type), the person unique number and shooting time of the first image are obtained, and the first picture information is composed of the person unique number and shooting time of the first image. When generating the person unique number of the first image, the person unique number can be composed of the date corresponding to the shooting time plus the shooting sequence number of the current photo (for example, the first image is the first photo taken today), so that the person photo No. 01 on March XX5, 20XX is used as the person unique number, and then combined with the shooting time of March XX5, 20XX HH1:MM1:SS1 to obtain the first picture information.

[0156] If it is determined that the person picture type of the first image is a document picture type (generally, if it is detected that there is a document in the picture, but no human body, real location scene, or product exists, it can be determined to be a document picture type), the text recognition result of the first image is obtained based on the image recognition model (such as the convolutional recurrent neural network model, also known as the CRNN model), and the shooting time is obtained from the picture attribute data. The text recognition result of the first image and the shooting time constitute the first picture information.

[0157] If the image recognition type of the first image is determined to be a physical image type (generally, if a physical product is detected in the image, and there are no human bodies, documents, or real scene locations, it can be determined to be a physical image type), the object recognition result of the first image is obtained based on the image recognition model (such as the target detection network included therein, more specifically, the single-lens multi-box detector model, i.e., the SSD model) (e.g., the object recognition result is product 1), and the shooting time is obtained from the image attribute data. The object recognition result of the first image and the shooting time form the first image information. It can be seen that based on the specific identification of the work photo type and the specific photo content, it is convenient to more clearly display other non-text editing work items in the future.

[0158] The log report generating module 170 is used to combine the first work log data subset, the second work log data subset and the third work log data subset to obtain work log report data.

[0159] In an embodiment of the present application, after the first work log data subset, the second work log data subset and the third work log data subset of the user (such as user A in the above example) are obtained in the cloud server, the first work log data subset, the second work log data subset and the third work log data subset are automatically combined to obtain work log report data. It can be seen that it is obtained after analyzing and summarizing various types of data based on the user, without the need for the user to collect work materials and then organize them, which improves data processing efficiency.

[0160] In one embodiment, the log report generating module 170 is specifically used to:

[0161] Acquire image information of the document image type in the first work log data subset and the third work log data subset to form a work content item report data set;

[0162] Acquire the picture information of the location picture type in the second work log data subset and the third work log data subset to form an attendance item report data set;

[0163] Acquire picture information of a person picture type in the third work log data subset, and acquire picture information of a thing picture type in the third work log data subset to form an auxiliary work content item report data set;

[0164] The work content item report data set, the attendance item report data set and the auxiliary work content item report data set are combined to obtain work log report data.

[0165] In an embodiment of the present application, since the work pictures included in the third work log data subset may be identified as pictures of customer visits, pictures of real objects, pictures of documents (such as minutes of a meeting) or pictures of real-life locations of the user, after further analyzing the work pictures included in the third work log data subset, the picture information of the location picture type in the third work log data subset may be further combined with the second work log data subset as an auxiliary attendance data to obtain an attendance item report data set; the picture information of the person picture type and the picture information of the thing picture type in the third work log data subset are combined to form an auxiliary work content item report data set; the picture information of the document picture type in the third work log data subset is combined with the first work log data subset to form a work content item report data set. Finally, the work content item report data set, the attendance item report data set and the auxiliary work content item report data set are combined, more specifically, filled into the corresponding filling areas in the preset work log report template (there are work content item report data set filling areas, attendance item report data set filling areas and auxiliary work content item report data set filling areas in the work log report template) to obtain work log report data. It can be seen that based on this method of automatically filling and generating report data, there is no need for manual sorting, and not only are there more data dimensions, but the efficiency is also higher.

[0166] The device automatically combines the first work log data subset, the second work log data subset and the third work log data subset to obtain work log report data without manual sorting, thereby improving the efficiency of work log data acquisition and improving data accuracy.

[0167] This application also provides a processing device, see Figure 3 , Figure 3 A schematic diagram of the structure of the processing device of the present application is shown. Specifically, the processing device provided by the present application includes a processor, and the processor is used to implement the following when executing the computer program stored in the memory. Figure 1 The steps in the corresponding embodiment; or, the processor is used to implement the following when executing the computer program stored in the memory. Figure 3 The functions of each module in the corresponding embodiment.

[0168] Exemplarily, the computer program may be divided into one or more modules / units, one or more modules / units are stored in a memory and executed by a processor to complete the present application. One or more modules / units may be a series of computer program instruction segments capable of completing a specific function, and the instruction segments are used to describe the execution process of the computer program in a computer device.

[0169] The processing device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the illustration is merely an example of a processing device and does not constitute a limitation on the processing device, and may include more or fewer components than shown in the illustration, or a combination of certain components, or different components. For example, the processing device may also include an input / output device, a network access device, a bus, etc., and the processor, the memory, the input / output device, the network access device, etc. are connected via a bus.

[0170] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the processing device, and uses various interfaces and lines to connect various parts of the entire processing device.

[0171] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the computer device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the processing device (such as audio data, video data, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0172] The display screen is used to display characters of at least one character type output by the input and output unit.

[0173] Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the above-described device, processing equipment and its corresponding modules can refer to the following. Figure 1 The description in the corresponding embodiments will not be repeated here in detail.

[0174] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0175] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the present application as follows: Figure 1 For the steps in the corresponding embodiments, the specific operations can refer to Figure 1 The description in the corresponding embodiments will not be repeated here.

[0176] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0177] Due to the instructions stored in the computer-readable storage medium, the present application can be executed. Figure 1 The steps in the corresponding embodiments, therefore, can realize the present application as follows Figure 1 The beneficial effects that can be achieved in the corresponding embodiments are detailed in the previous description and will not be repeated here.

[0178] The above is a detailed introduction to the artificial intelligence-based work log generation method, device and storage medium provided by the present application. The principles and implementation methods of the present application are explained by using specific examples in the embodiments of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for generating a work log based on artificial intelligence, characterized in that: The method comprises: When it is determined that the time interval between the current system time and the last work log generation time is equal to the preset first time period, a work log data screening time interval is generated according to the current system time and the last work log generation time, and user identity information is obtained; Obtain a user work log required data type set according to the user position type in the user identity information; wherein the user work log required data type set includes at least document type data, attendance type data and picture type data; Generate a current screening condition according to the user work log required data type set and the work log data screening time interval, and obtain a target screening data set corresponding to the user identity information according to the current screening condition; Acquire a first data subset of document type data in the target screening data set, perform keyword extraction on each document data included in the first data subset according to a preset keyword screening strategy, and obtain task data corresponding to each document data to form a first work log data subset; Acquire a second data subset of the target screening data set, which is attendance type data, and merge the attendance data included in the second data subset in ascending order of date to obtain a second work log data subset; Acquire a third data subset of picture type data in the target screening data set, and acquire an image recognition result set corresponding to the third data subset according to a pre-trained image recognition model to form a third work log data subset; The first work log data subset, the second work log data subset and the third work log data subset are combined to obtain work log report data.

2. The method according to claim 1, characterized in that The step of generating a work log data screening time interval according to the current system time and the last work log generation time, and obtaining user identity information, includes: The last work log generation time is used as the starting time point, the current system time is used as the ending time point, and the time period between the starting time point and the ending time point is used as the work log data screening time interval; Obtain user login information corresponding to the current data storage area, and obtain user identity information according to the user information set included in the user login information; wherein the user information set includes at least user account data and user identity information; the user identity information includes at least user name, user gender, user age, user work number, user position type and position responsibility remarks information.

3. The method according to claim 1, characterized in that The step of obtaining a user work log required data type set according to the user position type in the user identity information includes: When it is determined that the user position type is a first position type, obtaining a user work log required data type set corresponding to the first position type; wherein the first position type is a business position type; When it is determined that the user position type is the second position type, a user work log requirement data type set corresponding to the second position type is obtained; wherein the second position type is a technical support position type.

4. The method according to claim 1, characterized in that The keyword extraction is performed on each document data included in the first data subset according to a preset keyword screening strategy to obtain task data corresponding to each document data to form a first work log data subset, including: Obtaining the i-th document data in the first data subset, and obtaining the document modification time of the i-th document data; wherein the initial value of i is 1, and the value range of i is [1, N1], and N1 represents the total number of document data included in the first data subset; If it is determined that the document name of the i-th document data is a Chinese name, the document name of the i-th document data is obtained as the keyword of the i-th document data according to the keyword screening strategy, and the i-th task data is composed of the keyword of the i-th document data and the document modification time of the i-th document data; If it is determined that the document name of the i-th document data is a digital name, the body text of the i-th document data is obtained, and the i-th core keyword combination corresponding to the body text of the i-th document data is obtained according to the keyword screening strategy, and the i-th task data is composed of the i-th core keyword combination and the document modification time of the i-th document data; Update the value of i by adding 1 to the value of i; If it is determined that i does not exceed N1, returning to the step of obtaining the i-th document data in the first data subset and obtaining the document modification time of the i-th document data; If it is determined that i exceeds N1, task data No. 1 to task data No. N1 are obtained to form a first work log data subset.

5. The method according to claim 1, characterized in that The attendance data included in the second data subset are merged in ascending order of date to obtain a second work log data subset, including: Obtain the j-th attendance data subset in the second data subset; wherein the initial value of j is 1, and the value range of j is [1, N2], and N2 represents the total number of days in the first time period; Based on a preset daily working time period, obtaining the j-day attendance mark corresponding to the j-day attendance data subset; wherein the j-day attendance mark is one of a normal attendance mark and an abnormal attendance mark; Combine the j-th day's on-duty attendance time point and the j-th day's off-duty attendance time point of the j-th day's attendance data subset with the j-th day's attendance identifier to obtain the j-th attendance data; Update the value of j by adding 1 to it. If it is determined that j does not exceed N2, return to the step of obtaining the j-th attendance data subset in the second data subset; If it is determined that j exceeds N2, the attendance data No. 1 to No. N2 are obtained to form a second work log data subset.

6. The method according to claim 1, characterized in that The acquiring, according to the pre-trained image recognition model, an image recognition result set corresponding to the third data subset to form a third work log data subset includes: Obtaining the kth image in the third data subset; wherein the initial value of k is 1, and the value range of k is [1, N3], and N3 represents the total number of images in the third data subset; Acquire the image recognition type of the kth image based on the image recognition model; wherein the image recognition type is one of a location picture type, a person picture type, a document picture type, or an object picture type; If it is determined that the image recognition type of the k-th image is a location picture type, obtaining the shooting location information and shooting time of the k-th image, and forming the k-th picture information by the shooting location information and shooting time of the k-th image; If it is determined that the image recognition type of the k-th image is a person picture type, obtaining the person unique number and shooting time of the k-th image, and forming the k-th picture information by the person unique number and shooting time of the k-th image; If it is determined that the image recognition type of the k-th image is a document image type, obtaining the text recognition result and shooting time of the k-th image, and forming the k-th image information by the text recognition result and shooting time of the k-th image; If it is determined that the image recognition type of the k-th image is an object picture type, obtaining the object recognition result and shooting time of the k-th image, and forming the k-th picture information by the object recognition result and shooting time of the k-th image; Update the value of k by adding 1 to the value of k; If it is determined that k does not exceed N3, returning to the step of obtaining the kth image in the third data subset; If it is determined that j exceeds N3, the first picture information to the N3 picture information are obtained to form a third work log data subset.

7. The method according to claim 6, characterized in that The combining the first work log data subset, the second work log data subset and the third work log data subset to obtain work log report data includes: Acquire image information of the document image type in the first work log data subset and the third work log data subset to form a work content item report data set; Acquire the picture information of the location picture type in the second work log data subset and the third work log data subset to form an attendance item report data set; Acquire picture information of a person picture type in the third work log data subset, and acquire picture information of a thing picture type in the third work log data subset to form an auxiliary work content item report data set; The work content item report data set, the attendance item report data set and the auxiliary work content item report data set are combined to obtain work log report data.

8. A work log generation device based on artificial intelligence, characterized in that: include: A first acquisition module is used to generate a work log data screening time interval according to the current system time and the last work log generation time when it is determined that the time interval between the current system time and the last work log generation time is equal to a preset first time period, and to acquire user identity information; A second acquisition module is used to acquire a user work log required data type set according to the user position type in the user identity information; wherein the user work log required data type set includes at least document type data, attendance type data and picture type data; A data screening module, used to generate a current screening condition according to the user work log required data type set and the work log data screening time interval, and obtain a target screening data set corresponding to the user identity information according to the current screening condition; A first extraction module is used to obtain a first data subset of document type data in the target screening data set, perform keyword extraction on each document data included in the first data subset according to a preset keyword screening strategy, and obtain task data corresponding to each document data to form a first work log data subset; A second extraction module is used to obtain a second data subset of attendance type data in the target screening data set, and merge the attendance data included in the second data subset in ascending order of date to obtain a second work log data subset; A third extraction module is used to obtain a third data subset of picture type data in the target screening data set, and obtain an image recognition result set corresponding to the third data subset according to a pre-trained image recognition model to form a third work log data subset; The log report generating module is used to combine the first work log data subset, the second work log data subset and the third work log data subset to obtain work log report data.

9. A processing device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and the processor executes the method according to any one of claims 1 to 7 when calling the computer program in the memory.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the method according to any one of claims 1 to 7.

Images