JP6983145B2 - Data processing systems, methods, and programs - Google Patents

Description

The present invention relates to a technique for processing data relating to human behavior.

The decrease in the working population due to the declining birthrate and aging population has become a real problem. Organizations such as companies are required to continuously secure human resources and make effective use of those human resources in order to maintain and improve productivity and competitiveness.

It is expected that IT (Information Technology) technology will be utilized to realize this. And effective methods have been realized in some fields.

For example, IT technology is being promoted to reduce or replace some of the work of human resources (see Patent Document 1). For example, it can be expected that RPA (Robotic Process Automation) will reduce the work of human resources, and that RPA will replace the work of human resources.

Japanese Unexamined Patent Publication No. 2018-022257

Furthermore, it is desired to analyze the behavior of a person himself / herself by using IT technology in various fields and utilize the analysis result. For example, in an organization such as a company, it is desired to analyze the behavior of employees and utilize the analysis results so that the abilities of employees can be effectively utilized. Specifically, it is expected to realize measures such as curbing leave and turnover, highly productive work styles, and personnel allocation to improve productivity.

However, human behavior data contains a lot of noise data, and the diversity of the data is high, so that effective analysis results are often not obtained. For example, noise may be regarded as a feature in multiple regression analysis. In machine learning, noise may be over-learned. It is difficult to improve the accuracy of the prediction algorithm because multiple factors are intricately related to employee leave, turnover, and productivity.

An object of the present invention is to provide a technique for effectively analyzing human behavior.

The data processing system according to one embodiment of the present invention is a data processing system that processes data related to human behavior, and stores behavior data accumulating measurement values obtained by measuring the behavior of a predetermined behavior item by a person. Using the management unit, a coloring rule that defines the color corresponding to the measured value for each action item, and an arrangement rule that defines where in the image the area colored by the coloring rule is placed for each action item, An image output unit that generates an action image in which an area colored in a color corresponding to the measured value of each action item included in the action data for a predetermined period is arranged for each person, a person's action image, the nature of the person, and / Alternatively, it has an image analysis unit that learns correct answer information indicating a state and estimates the property and / or state of the person from the behavior image of the person based on the learning result.

It can support the effective utilization of the capacity of human resources.

It is a figure which shows the configuration example of the data processing system which concerns on Example 1. FIG. It is a figure which shows the configuration example of the work management apparatus. It is a figure which shows the configuration example of an image output device. It is a figure which shows the structural example of the image analyzer. It is a figure which shows the structural example of the working day table. It is a figure which shows the structural example of the coloring rule. It is a figure which shows the structural example of the arrangement rule. It is a figure which shows an example of the behavior image data. It is a schematic diagram which shows an example of a behavior image. It is a flowchart which shows an example of the image re-learning process. It is a figure which shows the display example of the display data. It is a figure which shows an example of the hardware composition of the apparatus which concerns on this disclosure. It is a figure for demonstrating information collection by a work management apparatus. It is a figure which shows an example of the behavior data. It is a figure which shows an example of the priority table which concerns on Example 2. FIG. It is a schematic diagram which shows an example of the behavior image which concerns on Example 2. FIG.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a diagram showing a configuration example of a data processing system according to the first embodiment.

The data processing system 100 is a system for processing data related to human behavior. As shown in FIG. 1, the data processing system 100 includes a work management device 101 which is an example of a data management unit, an image output device 201 which is an example of an image output unit, and an image analysis device 301 which is an example of an image analysis unit. And a terminal device 801 are provided. These devices 101, 201, 301, 801 are connected to a network 901 such as a wired LAN (Local Area Network) and / or a wireless LAN, and can transmit and receive data to and from each other. Note that these devices 101, 201, 301, and 801 may be computer devices shown in FIG. 12, respectively. Further, at least two of these devices 101, 201, and 301 may be configured as one computer device.

FIG. 2 is a diagram showing a configuration example of the work management device 101.

As shown in FIG. 2, the work management device 101 has a work day acquisition processing unit 102. The working day acquisition processing unit 102 acquires, for example, a working day table 103, which is an example of behavior data, from a predetermined DB (DataBase) or the like, and stores it in a storage device (see FIG. 12).

Behavior data is data that accumulates measured values that measure the behavior of a predetermined behavior item by a person. In the work day table 103, which is an example of behavior data, an example of an action item is "commuting time", and an example of a measured value corresponding to the action item "commuting time" is the time when an employee actually goes to work. The details of the working day table 103 will be described later (see FIG. 5).

FIG. 3 is a diagram showing a configuration example of the image output device 201.

As shown in FIG. 3, the image output device 201 has a behavior image generation processing unit 202. The action image generation processing unit 202 generates an action image 203 for each person (employee) from the work day table 103 by using the image generation rule 206 including the coloring rule 204 and the arrangement rule 205.

The coloring rule 204 defines the color corresponding to the measured value for each action item. The arrangement rule 205 defines where in the image the area (square) colored by the coloring rule 204 is arranged for each action item. The details of the coloring rule 204 and the arrangement rule 205 will be described later (see FIG. 6).

The action image 203 is arranged with a color-colored area corresponding to the measured value of each action item included in the action data for a predetermined period. The details of the behavior image 203 will be described later (see FIGS. 8 and 9).

FIG. 4 is a diagram showing a configuration example of the image analyzer 301.

As shown in FIG. 4, the image analysis device 301 has an image input processing unit 302, an image learning processing unit 303, an image classification processing unit 304, and a result output processing unit 305.

The image input processing unit 302 receives the input of the action image 203 generated by the image output device 201.

The image learning processing unit 303 learns a large amount of behavior images 203, and generates an estimation algorithm for classifying the behavior images 203 into groups having common characteristics in terms of images. For example, the image learning processing unit 303 learns a large amount of behavior images 203 by deep learning and generates a model of a neural network corresponding to an estimation algorithm. In this way, deep learning in which noise is suppressed by imaging makes it possible to accurately analyze complex human behavior.

Further, the image learning processing unit 303 associates correct answer information with each group. For example, the manager associates the group with correct information indicating the nature and / or state (that is, the behavior pattern of the person) of the people corresponding to the behavior image 203 belonging to the same group. In this way, human behavior can be effectively analyzed by imaging the data of human behavior and processing the image to estimate the nature and / or state of the person.

The image classification processing unit 304 classifies the behavior image 203 of the subject into any group by using the estimation algorithm generated by the image learning processing unit 303. Then, the image classification processing unit 304 generates the correct answer information associated with the group as the classification result 306. That is, the image classification processing unit 304 estimates the behavior pattern of the subject by using the estimation algorithm. In this way, the behavior of the subject can be effectively analyzed by estimating the property and / or the state of the subject by using the estimation algorithm learned by imaging the data of the behavior of the person.

The result output processing unit 305 is an example of an information presentation unit, and displays data 307 from the classification result 306 generated by the image classification processing unit 304 and displays it on a display device (see FIG. 12). As a result, the nature and / or condition of the target person and the utilization measures for it can be presented, so that the effective utilization of the human ability can be supported.

Further, the result output processing unit 305 may display the behavior image 203 of the target person on the display device. As a result, the manager can intuitively grasp the nature and / or state of the subject by the behavior image 203.

FIG. 5 is a diagram showing a configuration example of the working day table 103.

As shown in FIG. 5, the work day table 103 has an employee ID, a date, a work time, and a leave time as data items. For example, the first line of FIG. 5 shows that "employee A" went to work at "08:25" at "2018/07/01" and left at "18:05".

FIG. 6 is a diagram showing a configuration example of the coloring rule 204.

The coloring rule 204 defines, for example, that for action items related to working hours, the areas of the same action item have the same hue, and the longer the working hours, the darker the color tone. By defining the coloring rule 204 in this way, dark areas appear in the work image of the worker who has a long working time, so that the length of the working time of the worker can be intuitively grasped. ..

The upper table of FIG. 6 is an example of defining the correspondence relationship between the action item, the color scheme type, and the hue.

The first row of the table in the upper part of FIG. 6 stipulates that the color scheme type "shade 1" and the hue "blue" are used for the action item "starting time". Here, as shown in the first row of the table in the lower part of FIG. 6, “shade 1” is a shade “dark” in the range “minimum value less than 10%” and a range “10% or more and less than 60%”. It is stipulated that the shade "slightly dark" is used for "", the shade "slightly light" is used for the range "60% or more and less than 90%", and the shade "light" is used for the range "90% or more to the maximum value". That is, according to this coloring rule 204, in the action image 203, the square of the action item “starting time” is colored in a darker color as the time is earlier.

The second row of the table in the upper part of FIG. 6 stipulates that the color scheme type "shade 2" and the hue "green" are used for the action item "end time". Here, as shown in the second row of the lower table of FIG. 6, “shade 2” is a shade “light” in the range “minimum value less than 10%” and a range “10% or more and less than 60%”. It is stipulated that the shade "slightly light" is used for "", the shade "slightly dark" is used for the range "60% or more and less than 90%", and the shade "dark" is used for the range "90% or more to the maximum value". That is, according to this coloring rule 204, in the action image 203, the square of the action item “working time” is colored in a darker color as the time is later.

In the fourth row of the upper table in FIG. 6, the color scheme type "color coding 1" and the hue "1: blue 2: green 3: yellow 4: red" are used for the action item "health management time". Determine that. Here, as shown in the third row of the lower table in FIG. 6, "color coding 1" is the hue "color 1" in the range "0 or less" and the hue "color 1" in the range "0 or more and less than 15". It is defined that "color 2" is used, the hue "color 3" is used for the range "15 or more and less than 45", and the hue "color 4" is used for the range "45 or more".

FIG. 7 is a diagram showing a configuration example of the arrangement rule 205.

As shown in FIG. 7, the arrangement rule 205 is information in which daily action items are assigned to each of a plurality of cells (areas) divided vertically and horizontally. For example, as shown in FIG. 7, the arrangement rule 205 assigns daily action items to a plurality of cells so that dates are arranged in either the vertical direction or the horizontal direction and action items are arranged in the other direction. To determine. By defining the arrangement rule 205 in this way, the measured values for each date of the same item are arranged vertically or horizontally in a row, so that the similarity of behavior is likely to appear in the feature amount of the image.

Further, as shown in FIGS. 6 and 7, the action image 203 can be easily generated by coloring the mass with the color corresponding to the measured value of the action item according to the coloring rule 204 and the arrangement rule 205.

FIG. 8 is a diagram showing an example of behavior image data 213.

As shown in FIG. 8, the action image generation processing unit 202 generates the action image data 213 by arranging the dates in the vertical direction and arranging the action items in the horizontal direction according to the arrangement rule 205. Then, as shown in FIG. 8, the action image generation processing unit 202 associates the value of the action item in the action image data 213 with the hue and the shade according to the coloring rule 204. For example, the behavior image generation processing unit 202 associates the hue “blue” and the shade “light” with the start time “08:56” of “2018/04/02” according to the coloring rule 204 shown in FIG.

FIG. 9 is a diagram showing an example of the behavior image 203.

The behavior image generation processing unit 202 generates the behavior image 203 as shown in FIG. 9 based on the behavior image data shown in FIG. For example, the mass P of the action image 203 shown in FIG. 9 corresponds to the start time “08:56” of the action image data “2018/04/02” shown in FIG. Therefore, the action image generation processing unit 202 colors the mass P of the action image 203 with light blue. The behavior image generation processing unit 202 also colors the corresponding squares in the behavior image 203 for the closing time and the health management time. As a result, the start time, the end time, and the health management time can be reflected in the behavior image 203.

The image learning processing unit 303 generates an estimation algorithm for classifying into groups having common features in the image based on the behavior image 203 colored in this way.

For example, when the coloring rule 204 of FIG. 6 is obeyed, the behavior image 203 of an employee whose work time and leaving time are on time every sunrise becomes a light color as a whole. Employees who perform such work are considered to be suitable for management work, for example. Therefore, the correct answer information that it is suitable for management work is associated with the group A which has a light color as a whole as a common feature in the image. In this case, the image classification processing unit 304 may determine that the target person corresponding to the behavior image 203 classified into the A group may be suitable for the management work.

For example, when the coloring rule 204 of FIG. 6 is obeyed, the behavior image 203 of an employee whose work time is earlier than the scheduled time every sunrise and / or whose leaving time is later than the scheduled time is darkened as a whole. Employees who work in this way are considered to have a high need to follow up on their health. Therefore, the correct answer information that it is necessary to follow the health aspect is associated with the group B which has a dark color as a whole as a common feature in the image. In this case, the image classification processing unit 304 may determine that it is necessary to follow the health aspect of the subject corresponding to the behavior image 203 classified into the B group.

In addition, the image classification processing unit 304 is healthy when the recent behavior image 203 is classified into the B group even though the normal behavior image 203 of the subject is classified into the A group. It may be determined that it is necessary to follow the surface.

Further, the image classification processing unit 304 associates the correct answer information that the stress tolerance is high with respect to the B group, and stress tolerance is applied to the subject corresponding to the behavior image 203 classified into the B group in the image classification processing unit 304. It may be determined that there is a high possibility.

For example, when the coloring rule 204 of FIG. 6 is obeyed, the behavior image 203 of an employee who works flexibly has a mottled pattern (disjoint) color as a whole. Employees who perform such work are considered to be more suitable for spot support of specific projects than routine work, for example. Therefore, the correct answer information that it is suitable for spot correspondence is associated with the C group which has a mottled pattern as a whole as a common feature in the image. In this case, the image classification processing unit 304 may determine that the target person corresponding to the behavior image 203 classified into the C group may be suitable for spot correspondence.

FIG. 10 is a flowchart showing an example of the image re-learning process.

As described above, the image learning processing unit 303 generates an estimation algorithm from a large number of action images 203. However, when the classification result 306 by the estimation algorithm is significantly different from the administrator's assumption, or when the behavior data having a new tendency increases, it becomes necessary to adjust the estimation algorithm. In such a case, the image re-learning process shown in FIG. 10 may be executed. Next, the image learning process shown in FIG. 10 will be described.

First, the working day acquisition processing unit 102 acquires the working day table 103 in the working day acquisition processing (S101).

Next, the behavior image generation processing unit 202 generates the behavior image 203 from the working day table 103 in the behavior image generation processing (S102).

Next, the image learning processing unit 303 generates an estimation algorithm based on a large amount of action images 203 in the image learning process (S103).

Next, the image classification processing unit 304 classifies the behavior image 203 into any group by using the estimation algorithm in the image classification processing (S104).

The manager determines whether or not the tendency of the classification result 306 of S104 is correct (S105).

When the administrator determines that the tendency of the classification result 306 is correct (S105: YES), the flow ends this process.

When the manager determines that the tendency of the classification result 306 is not correct (S105: NO), the manager then decides what to adjust (S106).

When the administrator decides to adjust the image generation rule 206 (S106: image generation rule), the flow returns to S102 and performs the action image generation process using the action generation rule different from the previous one. As a result, an action image 203 different from the previous one is generated.

If the administrator decides to adjust the training data (S106: training data), the flow returns to S103 and generates a different group and estimation algorithm from the previous time. As a result, the behavior image 203 can be classified into a group different from the previous one.

FIG. 11 is a diagram showing a display example of the display data 307.

The result output processing unit 305 generates display data 307 based on the classification result 306, and displays the display data 307 on the display device as illustrated in FIG.

For example, regarding health care, using an estimation algorithm that classifies into three groups, "care required", "caution required", and "no problem", the behavior image of each employee in a certain department in March 2018. As a result of classifying 203, it is assumed that the behavior image 203 of one employee is classified into the group of "need care" and the behavior image 203 of two people is classified into the group of "need attention". In this case, as illustrated in FIG. 11, the result output processing unit 305 indicates that one employee is "needs care" and two employees are "need attention" as the March prediction result. indicate.

As a result, the boss of the department, the personnel department, etc. can take care of the employee before the health of the employee deteriorates.

FIG. 12 is a diagram showing an example of the hardware configuration of the apparatus according to the present disclosure.

As shown in FIG. 12, the work management device 101, the image output device 201, the image analysis device 301, and the terminal device 801 according to the present disclosure include the internal bus 1001, the processor 1002 connected to the internal bus 1001, and the main. It includes a memory 1003, a storage device 1004, a communication device 1005, an input device 1006, and a display device 1007.

The processor 1002 is, for example, a CPU (Central Processing Unit) or the like. The main memory 1003 is, for example, a DRAM (Dynamic Random Access Memory) or the like. The storage device 1004 is, for example, an SSD (Solid State Drive) or an HDD (Hard Disk Drive). The communication device 1005 is, for example, an Ethernet (registered trademark) board, a wireless LAN board, or the like. The input device 1006 is, for example, a keyboard, a mouse, or the like. The display device 1007 is, for example, a liquid crystal display or the like.

The functions of the work management device 101, the image output device 201, and the image analysis device 301 described above may be realized by the processor 1002 executing a computer program in the storage device 1004 and / or the main memory 1003.

In this embodiment, a simplified example is shown focusing on the start time, the end time, and the health management time as action items. However, the present invention is not limited to this, and various other items related to actions may be used as action items.

FIG. 13 is a diagram for explaining information collection by the work management device 101. FIG. 14 is a diagram showing an example of behavior data.

As shown in FIG. 13, the work management device 101 acquires various information related to behavior data from various DBs via various information integration devices. Then, the work management device 101 generates behavior data in which the measured values corresponding to the various behavior items on each date are entered for each employee from the acquired various information, and stores the behavior data in the storage device, as shown in FIG. Store. As shown in FIG. 14, the behavioral data may include measurements related to various work hours such as vacation classification, midnight time and business trip time, in addition to the above-mentioned start time, end time, and health management time. ..

In the first embodiment, in the action image 203, an example of allocating an area of the same size to all action items is shown, but the present invention is not limited thereto. In the second embodiment, a priority is given to the action item, and an example in which the area of each action item in the action image 203 is set to the size according to the priority is shown.

FIG. 15 is a diagram showing an example of a priority table. FIG. 16 is a schematic diagram showing an example of the behavior image 403 according to the second embodiment.

The priority table 401 is a table for setting priorities for action items. In the priority table 401 illustrated in FIG. 15, the highest priority "1" is set for the start time and the end time, and the lowest priority "3" is set for the health management time. In this case, the action image generation processing unit 202 has a relatively high priority of the start time and the end time in the generation of the action image 403, as shown in FIG. 16, and therefore the size of the mass corresponding to the start time and the end time. And reduce the size of the mass corresponding to the health care time because the priority of the health care time is relatively low.

As a result, the size of the square of the action item with higher priority becomes larger, so that the influence of the square with higher priority becomes larger when analyzing the characteristics of the action image 403. That is, by allocating a high-priority action item to a large area, it is possible to generate an action image 403 that enables analysis in consideration of the priority. Further, by setting the priority for each action item by the priority table 401, the extracted feature group can be adjusted.

Each of the above-described examples is an example for the purpose of explaining the present invention, and the scope of the present invention is not limited to those examples. One of ordinary skill in the art can practice the invention in various other embodiments without departing from the scope of the invention.

100 ... Data processing system 101 ... Work management device 102 ... Work day acquisition processing unit 201 ... Image output device 202 ... Behavior image generation processing unit 301 ... Image analysis device 302 ... Image input processing unit 303 ... Image learning processing unit 304 ... Image Classification processing unit 305 ... Result output processing unit

Claims (11)

A data processing system that processes data related to human behavior.
A data management unit that stores behavior data that accumulates measured values that measure the behavior of a given action item by a person, and
For each person, using a coloring rule that defines the color corresponding to the measured value for each action item and an arrangement rule that defines where in the image the area colored by the coloring rule is placed for each action item. An image output unit that generates an action image in which areas colored in colors corresponding to the measured values of each action item included in the action data for a predetermined period are arranged, and an image output unit.
An image analysis unit that learns a person's behavior image and correct answer information indicating the person's property and / or state, and estimates the person's property and / or state from the person's behavior image based on the learning result.
Data processing system with. The image analysis unit is for estimating the nature and / or state of the person based on the behavior image of the person by learning the behavior image of the person and the correct answer information indicating the nature and / or state of the person. An estimation algorithm is generated, and the nature and / or state of the subject is estimated from the behavior image of the subject based on the estimation algorithm.
The data processing system according to claim 1. The arrangement rule is information in which daily action items are assigned to each of a plurality of cells divided vertically and horizontally.
The image output unit generates an action image in which each cell is colored with a color corresponding to the measured value of the action item assigned to the cell based on the data of a person for a predetermined period.
The data processing system according to claim 1. The coloring rule stipulates that for action items related to working hours, the areas of the same action item have the same hue, and the longer the working hours, the darker the color tone.
The data processing system according to claim 1. The arrangement rule assigns daily action items to the plurality of cells so that the dates are arranged in either the vertical direction or the horizontal direction and the action items are arranged in the other direction.
The data processing system according to claim 3. Action items have priority,
The image output unit sets the area of the action item to the size according to the priority.
The data processing system according to claim 1. It also has an information presentation unit that displays information about the nature and / or condition of the subject.
The data processing system according to claim 1. The information presentation unit further displays the behavior image of the subject.
The data processing system according to claim 7. The learning is deep learning,
The data processing system according to claim 1. It is a data processing method for a computer device to process data related to human behavior.
Stores behavior data that accumulates measured values that measure the behavior of a given action item by a person,
For each person, using a coloring rule that defines the color corresponding to the measured value for each action item and an arrangement rule that defines where in the image the area colored by the coloring rule is placed for each action item. Generates an action image in which areas colored in colors corresponding to the measured values of each action item included in the action data for a predetermined period are arranged.
Learning a person's behavior image and correct information indicating the person's nature and / or state, and estimating the person's nature and / or state from the person's behavior image based on the learning result.
The data processing method that the computer does. A data processing program for processing data related to human behavior.
Stores behavior data that accumulates measured values that measure the behavior of a given action item by a person,
For each person, using a coloring rule that defines the color corresponding to the measured value for each action item and an arrangement rule that defines where in the image the area colored by the coloring rule is placed for each action item. Generates an action image in which areas colored in colors corresponding to the measured values of each action item included in the action data for a predetermined period are arranged.
Learning a person's behavior image and correct information indicating the person's nature and / or state, and estimating the person's nature and / or state from the person's behavior image based on the learning result.
A data processing program that lets a computer do things.

Images