JP7477988B2 - Condition determination device and program - Google Patents

Description

The present invention relates to a condition determination device and a program.

An increasing number of companies are using systems to keep track of their employees' physical and mental health on a daily basis.

For example, Patent Document 1 describes a method in which employees are asked to select one of three buttons - "sunny," "cloudy," or "rainy" - when they leave work, and the employee's health condition is determined based on the button selected.

JP 2019-164737 A

However, with conventional methods, it was not possible to take into account the tendencies of individual employees' responses (tendencies in which buttons they select) when making a judgment. For example, if someone who usually selects "rain" selects "rain," and if someone who usually selects "sunny" suddenly selects "rain," the latter is considered to be a more serious situation, but there was no mechanism for making such a judgment.

The present invention was made in consideration of the above-mentioned circumstances, and one of its objectives is to provide a system that can determine appropriate conditions by taking into account the tendencies of an individual's responses.

A condition determination device according to one embodiment of the present invention is a condition determination device that determines the condition of a user, and includes an answer content acquisition unit that acquires the content of the user's answers to one or more questions, a condition value calculation unit that calculates the actual condition value of the user based on the content of the answers, a condition value estimation unit that estimates the range of the user's current condition value based on time series changes in the user's previously calculated condition value and time series changes in the condition values of multiple users including one or more other users, and a determination unit that determines the user's condition based on whether the actual condition value is within the estimated range.

A program according to one embodiment of the present invention causes a computer for determining a user's condition to function as an answer content acquisition unit that acquires the user's answers to one or more questions, a condition value calculation unit that calculates the user's actual condition value based on the answers, a condition value estimation unit that estimates the range of the user's current condition value based on time series changes in the user's previously calculated condition value and time series changes in the condition values of multiple users including one or more other users, and a determination unit that determines the user's condition based on whether the actual condition value is within the estimated range.

The present invention provides a system that can determine appropriate conditions by taking into account the tendencies of an individual's responses.

1 is a diagram showing a schematic configuration of a condition determination system 1 according to an embodiment of the present invention. 4 is a flowchart of a condition value calculation process in the condition determination system 1 according to the embodiment of the present invention. 3 is a flowchart of a condition determination process in the condition determination system 1 according to the embodiment of the present invention. FIG. 2 is a diagram illustrating a questionnaire response screen displayed on a user terminal 30 in the condition determination system 1 according to the embodiment of the present invention. FIG. 2 is a diagram illustrating a distribution of estimated values of a condition value in the condition determination system 1 according to the embodiment of the present invention. 1 is a diagram illustrating an example of a judgment result screen displayed on a management terminal 20 in a condition judgment system 1 according to an embodiment of the present invention. FIG. 3A to 3C are diagrams illustrating examples of time-series changes and estimation ranges of condition values in the condition determination system 1 according to the embodiment of the present invention.

The following describes in detail an embodiment of the present invention with reference to the drawings. Note that identical elements are given the same reference numerals and duplicated descriptions are omitted.

Figure 1 is a diagram showing the configuration of a condition determination system 1 according to this embodiment. As shown in Figure 1, the condition determination system 1 includes a management server (condition determination device) 10, a management terminal 20, and a user terminal 30. The management server 10, the management terminal 20, and the user terminal 30 are connected via a communication network N.

The communication network N may be, for example, the Internet, a LAN, a telephone line, an in-house network, a mobile communication network, Bluetooth (registered trademark), WiFi (Wireless Fidelity), other communication lines, or a combination thereof, and may be either wired or wireless. In addition, multiple user terminals 30 may be included.

The management server 10 is a server that manages the condition information of employees of a company or the like. The condition information is a numerical value (condition value) calculated based on the responses to a questionnaire that employees periodically answer.

The management server 10 is a general-purpose computer and may be composed of a single computer or multiple computers distributed over a communication network N. The management server 10 comprises a control device (answer content acquisition unit, condition value calculation unit, condition value estimation unit, judgment unit) 11 and an external storage device 12. The control device 11 comprises, as hardware, a CPU, memories such as ROM and RAM, an input interface, an output interface, a communication interface, and buses connecting these. The control device 11 realizes various functions by the CPU executing programs stored in the ROM etc.

The external storage device 12 is a hard disk drive or the like. The external storage device 12 implements a condition information management table T1.

The condition information management table T1 stores the responses of each employee to the questionnaire and the history of the condition values calculated based on the responses. The questionnaire is conducted periodically (e.g., monthly), and time series data of the condition values for each employee for a certain number of times in the past (several months) is stored.

The management terminal 20 is a terminal that allows company managers to view the responses to questionnaires registered by employees and the condition values of each employee calculated based on the responses. The management terminal 20 is a terminal that can communicate with the management server 10 via the communication network N, such as a tablet terminal, smartphone, notebook PC (personal computer), or PC.

The user terminal 30 is a terminal through which a user (employee) accesses the management server 10 and registers responses to a questionnaire. The user terminal 30 is a terminal capable of communicating with the management server 10 via the communication network N, such as a smartphone, tablet terminal, notebook PC, or mobile phone.

Next, the flow of condition determination by the condition determination system 1 will be explained using the flowcharts in Figures 2 and 3.

Figure 2 is a flowchart of the process of calculating a condition value based on the responses of employee A to a questionnaire. First, employee A accesses a dedicated website for the in-house condition determination system 1 from the user terminal 30 (step S101).

A survey response screen as shown in FIG. 4 is displayed on the display of the user terminal 30 (step S102). As shown in FIG. 4, the survey response screen displays the content of the question (three items in the example of FIG. 4), buttons for selecting an answer (e.g., a five-point selection), an area for inputting free comments, etc.

When employee A completes answering the question using the input means (e.g., touch panel) of the user terminal 30 (step S103: YES), the answer is sent to the management server 10 (step S104).

The management server 10 stores the received answer content in the condition information management table T1. The management server 10 also calculates the current condition value based on the received answer content and stores it in the condition information management table T1 (step S105). For example, the answer content for each question item may be expressed as a score from 1 to 5 points (5 points for very good and 1 point for very bad), and the total points for each question may be used as the condition value.

Figure 3 is a flowchart of the process of determining the condition of employee A from the actual condition value of employee A. The management server 10 creates a prediction model for the condition value using historical data (time series data of the condition value) of the condition values of K employees stored in the information management table T1 (step S201).

Specifically, a prediction model is created that predicts the possible range of the condition value for the N+1th time from the trend (time series change) of the condition values for the past N times for K people (for example, condition values for the 12 months up to the previous month). The prediction model can be created using existing statistical methods or machine learning methods.

Furthermore, the management server 10 uses the created prediction model to estimate the possible range of the condition value for the current time (N+1 time) based on the trend of employee A's condition values for the past N times (step S202). The possible range of the condition value can be represented, for example, by a probability distribution as shown in FIG. 5.

The management server 10 judges the condition of employee A based on whether or not the actual condition value of employee A (the condition value calculated in step S105 of FIG. 2) is within the range of the estimated condition value (step S203).

For example, as shown in Figure 5, in the probability distribution of condition values, the condition value (x) that is included in 95% of the total is set as the estimated range (the "standard" range in the figure), and it is determined whether the actual condition value is included in the standard range.

When the actual condition value is not within the standard range, the management server 10 judges the condition of employee A according to the difference between the actual condition value and the standard range. For example, when the actual condition value is lower than the standard (the "bad" range in FIG. 5), it is judged as "poor condition." On the other hand, when the actual condition value is higher than the standard (the "good" range in FIG. 5), it is judged as "good condition."

The judgment results may be notified to the management terminal 20 for viewing by the manager. FIG. 6 is a diagram illustrating an example of a judgment result screen displayed on the display of the management terminal 20. As shown in FIG. 6, a list of the responses of each employee to the latest questionnaire is displayed, and employees who are judged to be in "poor form" or "good form" can be confirmed. The manager can take the necessary measures as a priority for employees who are judged to be in "poor form." In addition, by clearly indicating employees who are judged to be in "good form," it is possible to discover employees who are in good form.

A concrete example will be used to explain the time series change in condition values and the estimation of the range. Graphs (1) and (2) in FIG. 7 respectively show the change in condition values for each month of the past year and the latest actual measured value R for two employees. The estimated range P shows the range of condition values estimated using a prediction model. As is clear from the graph, the estimated range P for (2), which has a large variance, is wider than that for (1), which has a small variance in condition values. Also, both actual measured values R this time are lower than the previous ones, but only (1) is lower than the estimated range P, and (2) is within the estimated range P, so only (1) is judged to be "ill". Thus, although the actual measured value R itself is lower for (1) at "9" and (2) at "7", (2) is lower, (1) is judged to be "ill" in relation to the estimated range P.

As described above, according to this embodiment, the employee's condition is determined based on whether the actual condition value calculated based on the employee's responses to a questionnaire falls within a range estimated based on time-series changes in the condition values of multiple users. This makes it possible to appropriately determine whether an employee is in "bad" or "good" condition, regardless of the actual condition value itself.

In addition, the condition is judged as "good" or "bad" depending on the difference from the estimated range (high or low). This makes it possible to discover not only employees who are in bad shape and need immediate attention, but also employees who are doing well.

In addition, condition values are estimated using a predictive model created using statistical and machine learning techniques, allowing for highly accurate estimations that appropriately reflect large amounts of historical data.

In the above embodiment, the answers to each question are converted into numbers, and the sum of these numbers is used as the condition value, but the method of calculating the condition value is not limited to this, and any method that can reflect the answers to the questions may be used.

The present invention is not limited to the above-described embodiment, and can be implemented in various other forms without departing from the spirit of the present invention. Therefore, the above-described embodiment is merely illustrative in all respects and should not be interpreted in a restrictive manner. For example, the above-described processing steps can be arbitrarily changed in order or executed in parallel as long as no inconsistencies occur in the processing content.

Reference Signs List 1... condition determination system 10... management server 11... control device 12... external storage device 20... management terminal 30... user terminal T1... condition information management table N... communication network

Claims (3)

A condition determination device for determining a user's condition, comprising:
an answer content acquisition unit that acquires answers from the user to one or more questions;
a condition value calculation unit that calculates an actual condition value of the user based on the answer content;
a condition value estimation unit that estimates a range of the current condition value of the user based on a time series change in a condition value previously calculated for the user and a time series change in condition values of a plurality of users including one or more other users;
a determination unit that determines a condition of the user based on whether the actual condition value is within the estimated range,
The one or more questions include at least a question about job satisfaction and a question about interpersonal relationships in the workplace;
The condition value estimation unit uses data on time-series changes in the condition values of the multiple users and a prediction model created by a machine learning technique to estimate the range so that the range is wider for users whose condition values have more variability in time-series changes than for users whose condition values have less variability in time-series changes. The determination unit is
The condition determining device according to claim 1 , wherein the condition is determined based on a difference from the range. A computer that judges the user's condition,
an answer content acquisition unit that acquires answers of the user to one or more questions;
a condition value calculation unit that calculates an actual condition value of the user based on the answer content;
a condition value estimation unit that estimates a range of the current condition value of the user based on a time series change in a condition value previously calculated for the user and a time series change in condition values of a plurality of users including one or more other users;
a determination unit that determines a condition of the user based on whether the actual condition value is within the estimated range ;
The one or more questions include at least a question about job satisfaction and a question about interpersonal relationships in the workplace;
The condition value estimation unit uses data on time-series changes in the condition values of the multiple users and a prediction model created by a machine learning technique to estimate the range so that the range is wider for users whose condition value has more variability in time-series changes than for users whose condition value has less variability in time-series changes.

Images