Human resource matching system and human resource matching method

Abstract

To provide a human resource matching system and a human resource matching method capable of accurately grasping personality of an individual, for example, and eliminating mismatch between the individual and an organization, for example.SOLUTION: A human resource matching system 100 which utilizes personality data acquired by answering a question with swipe operation includes: a storage unit 104 which stores one-side personality data which is a population on one side of personality data and other-side personality data which is a population on the other side of the personality data; and a matching generation unit 106 which matches the one-side personality data with the other-side personality data.SELECTED DRAWING: Figure 15

Description

[Technical Field]

[0001] The present invention relates to a talent matching system and a talent matching method that utilizes personality acquired by, for example, a respondent's swiping action. [Background technology]

[0002] In the so-called seller's market, the percentage of companies whose job offers are declined by new graduates is on the rise. Even when companies do hire new graduates, the number of new graduates who leave their jobs in the first year is on the rise every year. In this situation, while the costs of acquiring the talent that companies need in the new graduate market are on the rise every year, the number of new graduates who leave their jobs early after being hired is increasing, making it extremely difficult for companies to hire students, and new graduate recruitment has become a major management issue for companies.

[0003] The main reason why students leave their jobs early is because of mismatched interpersonal relationships within the company. The reason for this mismatch is the inability to accurately grasp the individual's personality.

[0004] In the student recruitment activities of companies such as those described above, there are high hopes for avoiding mismatches in human relationships in advance, as this will lead to continued satisfaction on both the company and the student side during the company's new graduate recruitment activities and after the recruitment, reduce early student turnover, and enable the company's management to proceed smoothly.

[0005] These issues are not limited to those between students and companies. They arise in all aspects of social life, including the relationship between working adults and corporate organizations, between students and schools, between various organizations, and even between individuals and their communities.

[0006] Furthermore, this is also an issue that lies in matching people together for the purpose of finding a marriage partner. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Publication No. 2019-168887 [Patent Document 2] Patent No. 6055794 Summary of the Invention [Problem to be solved by the invention]

[0008] Therefore, in order to solve the above problems, the present invention aims to provide a human resources matching system and a human resources matching method that can accurately grasp an individual's personality and eliminate mismatches between an individual and an organization, etc., in advance. [Means for solving the problem]

[0009] The first invention is a talent matching system that utilizes personality data acquired by answering questions with a swipe action, a storage unit that stores one-side personality data that is a population on one side of the personality data, and other-side personality data that is a population on the other side of the personality data; a matching generation unit that matches the one-party personality data with the other-party personality data; It is preferable that the personnel matching system utilizes personality data.

[0010] In this case, the system may have a one-side application that inputs the one-side personality data and a other-side application that inputs the other-side personality data, and the one-side personality data input from the one-side application and the other-side personality data input from the other-side application may be stored in a memory unit.

[0011] In this case, the matching generation unit may receive the one-side personality data and the other-side personality data as input, generate matching data, and output the matching data to the one-side application and the other-side application, respectively.

[0012] In this case, the matching generation unit may use the one-side personality data and the other-side personality data as input data, and output multiple matching models as recommended personality data using multiple possible matching cases as training data, for example, using a machine-learned predictive model.

[0013] In this case, the matching generation unit may search for and output content related to talent matching based on the recommended personality data.

[0014] The second invention is a personnel matching method that utilizes personality data acquired by answering questions with a swipe action, a first step of storing one-side personality data, which is a population on one side of the personality data, and another-side personality data, which is a population on the other side of the personality data; a second step of matching the one-party personality data with the other-party personality data; It is preferable that the method is a personnel matching method utilizing personality data, which has the above-mentioned features.

[0015] In this case, the method may include a third step of using a one-side application that inputs the one-side personality data and a other-side application that inputs the other-side personality data to store the one-side personality data input from the one-side application and the other-side personality data input from the other-side application.

[0016] In this case, the method may include a fourth step of receiving the one-side personality data and the other-side personality data as input and generating matching data, and a fifth step of outputting the matching data to the one-side application and the other-side application, respectively.

[0017] In this case, a sixth step may be included in which the one-side personality data and the other-side personality data are used as input data, multiple possible matching cases are used as training data, and multiple matching models are output as recommended personality data, for example, using a machine-learned predictive model.

[0018] In this case, the method may include a seventh step of searching for and outputting content relating to talent matching based on the recommended personality data.

[0019] A third invention is a human resources matching program that utilizes personality data acquired by answering questions with a swipe motion, a first step of storing one-side personality data representing one side of a population and another-side personality data representing another side of a population; a second step of matching the one-party personality data with the other-party personality data; It is preferable that the program is a talent matching program that utilizes personality data.

[0020] In this case, the method may include a third step of using a one-side application that inputs the one-side personality data and a other-side application that inputs the other-side personality data to store the one-side personality data input from the one-side application and the other-side personality data input from the other-side application.

[0021] In this case, the method may include a fourth step of receiving the one-side personality data and the other-side personality data as input and generating matching data, and a fifth step of outputting the matching data to the one-side application and the other-side application, respectively.

[0022] In this case, a sixth step may be included in which the one-side personality data and the other-side personality data are used as input data, multiple possible matching cases are used as training data, and multiple matching models are output as recommended personality data, for example, using a machine-learned predictive model.

[0023] In this case, the method may include a seventh step of searching for and outputting content relating to talent matching based on the recommended personality data.

[0024] In addition, as a combination of "one side" and "local side", variations such as "individual side" and "individual side", "individual side" and "organization side", "organization side" and "organization side" are valid as problem-solving means of the present invention. [Effects of the Invention]

[0025] According to the present invention, it is possible to accurately grasp each person's personality, and in turn, to eliminate mismatches between individuals and organizations in advance, for example. [Brief explanation of the drawings]

[0026] [Figure 1] 1 is a block diagram of a personality diagnostic system according to a first embodiment of the present invention. [Figure 2] FIG. 1 is a diagram illustrating steps constituting a personality diagnostic method according to a first embodiment of the present invention. [Figure 3] FIG. 2 is a diagram illustrating correction of a swipe time acquired by the personality diagnostic system according to the first embodiment of the present invention. [Figure 4]FIG. 1 is a diagram showing a scheme of personality diagnosis by swiping according to a first embodiment of the present invention. [Figure 5] 1 is a flowchart of a swipe diagnostic test according to a first embodiment of the present invention. [Figure 6] FIG. 10 is a diagram comparing the characteristics of personality data that can be acquired in a conventional personality test and the personality test of the first embodiment. [Figure 7] FIG. 1 is a diagram illustrating information (elements) for determining personality in the personality measurement method according to the first embodiment of the present invention. [Figure 8] FIG. 10 is a diagram showing the relationship between coordinate movement on a device and response strength in a swipe action. [Figure 9] FIG. 10 is a conceptual diagram illustrating an example of logic of a response filter. [Figure 10] FIG. 10 is an explanatory diagram showing an example of logic of a response filter. [Figure 11] FIG. 10 is a diagram showing an example of a filter strength detected when the logic has an independent learning function that allows the logic to independently learn the situation during an exam. [Figure 12] FIG. 10 is a diagram showing an example of averaged personality data. [Figure 13] A diagram showing various data stored in a database. [Figure 14] FIG. 10 is a diagram showing initial values ​​in parameter tuning of a response filter and an assumed tendency of successive responses. [Figure 15] FIG. 10 is a block diagram of a talent matching system according to a second embodiment of the present invention. [Figure 16] FIG. 10 is a diagram illustrating steps constituting a talent matching method according to a second embodiment of the present invention. [Figure 17] FIG. 10 is a diagram showing a scheme of a talent matching system according to a second embodiment of the present invention. [Figure 18] 10 is a flowchart of talent matching according to a second embodiment of the present invention. [Figure 19]FIG. 10 is a diagram showing an example of a machine learning scheme (recommended personality output) in the talent matching system according to the second embodiment of the present invention. [Figure 20] FIG. 10 is a diagram showing an example of a machine learning scheme (matching objective variable output) in the personnel matching system according to the second embodiment of the present invention. [Figure 21] FIG. 1 is a diagram illustrating the relationship between a general optimization model and an organizational optimization model for achieving an optimal match. [Figure 22] FIG. 10 illustrates an example of logic for defining a best match. [Figure 23] FIG. 1 is a conceptual diagram showing an example of an algorithm for matching individuals and organizations. DETAILED DESCRIPTION OF THE INVENTION

[0027] [First embodiment] First, a personality measurement system, a personality diagnostic system, a personality measurement method, a personality diagnostic method, and a personality diagnostic program according to a first embodiment of the present invention will be described with reference to the drawings. The personality measurement system and the personality diagnostic system may be installed or downloaded into an existing housing (e.g., hardware such as a terminal or device, e.g., a mobile phone, a smartphone, or a personal computer) like software such as application software, or may be implemented by the hardware such as a terminal or device (e.g., a personal computer, a server, etc.).

[0028] In detail, the personality measurement system, personality diagnostic system, personality measurement method, personality diagnostic method, and personality diagnostic program according to the first embodiment of the present invention are devices, processes, or programs for diagnosing a respondent's personality, assuming that the respondent performs a swiping motion (also simply referred to as "swiping") on the screen of a terminal or device such as a smartphone.

[0029] Specifically, for example, the swipe time required for a swipe action is the time it takes for a respondent to touch a touch panel screen with their finger, slide their finger across the screen, and then lift their finger from the screen. While the respondent reads and understands the question, they do not perform a swipe action to select an answer, but instead understand the question and think about the answer before performing a swipe action to select an answer. Once the swipe action is completed, the respondent's answer is determined and the next question is displayed on the screen. Then, the respondent begins to read and understand the next question, and then performs a swipe action to select an answer to the next question. Therefore, the swipe time includes the play time while the respondent reads and understands the question.

[0030] [1-1. Personality Diagnostic System] As shown in FIG. 1, the personality diagnostic system 10 includes, for example, a time setting unit 12, a memory unit 14, a measurement unit 16, a judgment unit 18, a discrimination unit 20, a generation unit 22, and a filter setting unit 24.

[0031] Here, a system configured with at least the measurement unit 16 and the judgment unit 18 of the personality diagnostic system 10 is referred to as a personality measurement system 11. The personality measurement system 11 is a device that can accurately measure the personality of a respondent based on the respondent's swiping action. The personality measurement system 11 may also additionally include a judgment unit 20.

[0032] For ease of explanation, FIG. 1 illustrates a configuration in which each unit of the personality diagnostic system 10 is provided on a single device, but this configuration is not limited to this. For example, a configuration in which the units are distributed across multiple different devices may be adopted. For example, a single device or device may be used solely for inputting swipe actions, with the units being provided on a server or cloud located outside the device or device. Alternatively, a single device or device may be equipped with only the measurement unit 16, with the other units being provided on a server or cloud located outside the device or device. Furthermore, some of the units of the personality diagnostic system 10 may be configured with components such as a CPU, and other units may be provided by performing calculations using software programs or the like.

[0033] Here, the time setting unit 12, the measurement unit 16, the determination unit 18, the discrimination unit 20, the generation unit 22, and the filter setting unit 24 are configured by, for example, a central processing unit (CPU) or a control unit (controller).

[0034] The storage unit 14 is configured with a RAM (Random Access Memory), a ROM (Read Only Memory), storage, etc. However, it is not limited to these configurations.

[0035] Furthermore, if the personality diagnostic system 10 is configured as software or a program such as application software, it may mean, for example, steps or commands for driving the time setting unit 12, memory unit 14, measurement unit 16, judgment unit 18, discrimination unit 20, generation unit 22 and filter setting unit 24, but is not limited to this.

[0036] A question is output based on question data previously stored in the storage unit, and the time setting unit 12 sets a standard answer time for the respondent to answer the question. The standard answer time set by the time setting unit 12 is stored in the storage unit 14. The timing at which the time setting unit 12 sets the standard answer time can be freely changed, and it is also possible to set many different patterns depending on the respondent's attributes (for example, age, gender, educational background, humanities or science, desired content, survey results, etc.).

[0037] The time setting unit 12 is a device or process for registering the personality data of respondents in a database.

[0038] The storage unit 14 stores various data and information required for personality diagnosis. For example, it is possible to store the standard answer time set by the time setting unit 12 in association with the question content at that time. The storage unit 14 also stores personal information (address, name, age, etc.) and attribute data of the respondent.

[0039] The storage unit 14 is a device or process for registering personality data of respondents in a database.

[0040] The measurement unit 16 measures the swipe time, for example, the time required for a respondent to swipe on a touch panel screen. Specifically, the respondent answers questions displayed on the screen of a device or apparatus by swiping on the screen, and the measurement unit 16 measures the time required for the swipe. A swipe refers to an action in which the respondent slides their finger across the touch panel screen while keeping it in contact with the screen.

[0041] The time required for a swipe action refers to the time it takes for a respondent to touch their finger to the touch panel screen, slide their finger across the screen, and then lift their finger from the screen. If multiple questions are prepared, when a swipe action is performed on the first question, the answer to the first question is selected and the content of the second question is displayed on the screen. When a swipe action is performed on the second question, the answer to the second question is selected and the content of the third question is displayed on the screen. When a swipe action is performed on the third question, the answer to the third question is selected and the content of the fourth question is displayed on the screen. This cycle is repeated the number of times set in advance for the number of questions.

[0042] It should be noted that, for example, the detection technology of Japanese Patent No. 6055794 (see Patent Document 1 as prior art described in the specification) can be used or applied as a method for measuring the swipe time by the measurement unit 16. Various data or information measured by the measurement unit 16 is stored in the storage unit 14 as appropriate.

[0043] The measurement unit 16 is a device or process for registering the personality data of respondents in a database.

[0044] The determination unit 18 determines the answer strength based on the measurement results by the measurement unit 16. For example, the memory unit 14 stores answer strength determination data or an answer strength determination table for determining the answer strength. The determination unit 18 may determine the answer strength based on the measurement results by the measurement unit 18, for example, using the answer strength determination data or the answer strength determination table. The answer strength determination data or the answer strength determination table by the determination unit 18 may be saved in the memory unit 14.

[0045] Regarding the determination of answer strength, if the answer time is short, it can be assumed that the respondent is answering based on their beliefs and confidence, and therefore the answer strength is determined to be high. On the other hand, if the answer time is long, it can be assumed that the respondent is answering while hesitant, and therefore the answer strength is determined to be low. If a standard answer time is set for each question by the time setting unit 12, the answer strength may be determined based on whether the standard answer time is shorter or longer than the standard answer time.

[0046] Even if the answer strength is not based on the setting of the standard answer time, it is also possible to determine the answer strength based on, for example, the standard deviation from the answerer's general answer time.

[0047] Here, it is preferable that the determination unit 18 determines the answer strength based on the remaining time remaining after subtracting a predetermined time from the swipe time. This allows the respondent to exclude the idle time from when they start reading the question until they understand it, and by measuring the true time required to reach an answer, the answer strength can be determined more accurately. For example, the swipe time is the time it takes for a respondent to touch a touch panel screen with their finger, slide their finger across the screen, and then lift their finger from the screen. However, while the respondent is reading and understanding the question, they do not perform a swipe motion to select an answer. Furthermore, after understanding the question and thinking about an answer, the respondent begins a swipe motion to select an answer. When the respondent begins a swipe motion and completes the swipe motion, the respondent's answer is determined and the next question is displayed on the screen. Then, the respondent begins to read and understand the next question, and then performs a swipe motion to select an answer for the next question. Therefore, although the swipe time includes the idle time the respondent spends reading and understanding the question, this idle time does not accurately represent the answer strength. Therefore, we decided to exclude the play time from the swipe time.

[0048] To express it graphically, for example, as shown in Figure 2, if the "swipe time" is "T," which is the time it takes for a respondent to touch a touch panel screen with their finger, slide their finger across the screen, and then lift their finger from the screen, the initial time "S" is the predetermined time (play time) required to read and understand the content of the question. Therefore, by correcting the play time "S" and calculating the remaining time as the swipe time "TS," a precise swipe time can be determined to accurately judge the strength of the response.

[0049] In addition to the two patterns of "strong" and "weak," the strength of the response can also be expressed as "medium," and it is also possible to express it in multiple patterns using many levels, such as a 5-point scale, a 10-point scale, etc.

[0050] The determination unit 18 is a device or process for registering personality data of respondents in a database.

[0051] The determination unit 20 determines the direction of the swipe motion by the respondent as the answer direction. The answer direction can also be called the swipe direction. The direction of the swipe motion means the swipe direction, and techniques for detecting the swipe direction itself have been known for some time. For example, the techniques of Japanese Patent Publication No. 2023-095529, Japanese Patent Publication No. 2023-172808, Japanese Patent Publication No. 2023-052046, Japanese Patent Publication No. 2022-158827, and Japanese Patent Publication No. 2020-039857 can also be applied.

[0052] Here, the answer direction, which refers to the direction of a swipe motion by a respondent, means the direction in which the respondent touches and moves their finger on the screen of the device, but the swipe direction may be predetermined for each answer to each question. For example, if the question asks for "yes" or "no," selecting "yes" can involve swiping to the right on the screen, and selecting "no" can involve swiping to the left on the screen. Also, for example, if the question asks for a number, such as "number 1" or "number 2," selecting "number 1" can involve swiping to the right on the screen, and selecting "number 2" can involve swiping to the left on the screen.

[0053] Although two response directions have been exemplified, the present invention is not limited to this. For example, it is possible to obtain a total of three or four response directions by adding at least one direction, either upward or downward, in addition to the left and right directions. Furthermore, by swiping diagonally on the screen, it is possible to obtain even more response directions.

[0054] The determination unit 20 is a device or process for registering personality data of respondents in a database.

[0055] The generation unit 22 generates personality data of the respondent from the answer strength determined by the determination unit 18 and the answer direction determined by the discrimination unit 20. By composing the personality data from "answer strength x answer direction," it is possible to obtain personality data (big data) composed of patterns for many respondents. The personality data generated by the generation unit 22 may be linked to the respondent and stored in the storage unit at a predetermined timing.

[0056] The generating unit 22 is a device or process for registering the personality data of respondents in a database.

[0057] The filter setting unit 24 sets a threshold value for the answer filter for determining that the answer strength is invalid when the swipe time becomes an abnormal answer time, thereby removing answers with reduced reliability from the personality data.

[0058] Here, the abnormal response time is a time set by, for example, the time setting unit 12. The time setting unit 12 sets the standard response time, but can also set an abnormal response time at the same time. The abnormal response time means a time that is clearly too short or a time that is clearly too long. It is preferable to set an optimal abnormal response time for each question content.

[0059] The reason why a time that is clearly too short is considered an abnormal answer time is because, when it is clearly too short, there is a risk that the respondent is answering for fun or has answered without carefully considering the content of the question. On the other hand, the reason why a time that is clearly too long is considered an abnormal answer time is because, when it is clearly too long, there is a risk that the respondent has forgotten to answer or has chosen an answer based on criteria that differ from their true feelings. By removing these times as abnormal answer times from the personality data, abnormal noise can be removed, thereby increasing the reliability of the personality data.

[0060] If the swipe time is an abnormal answer time, the filter setting unit 24 can exclude the question content and answer content from the personality data by setting a threshold value for the answer filter for determining that the answer strength is invalid. Note that the filter setting unit 24 can also set an abnormal flag on the data of the question content and answer content for which the swipe time is an abnormal answer time, to indicate that the value is abnormal.

[0061] The filter setting unit 24 is a device or process for registering personality data of respondents in a database.

[0062] [1-2. Personality diagnostic method and personality diagnostic program] 3, the personality diagnostic method includes, for example, a first step S100, a second step S110, a third step S120, a fourth step S130, a fifth step S140, a sixth step S150, a seventh step S160, an eighth step S170, a ninth step S180, and a tenth step S190. The steps may also be referred to as "personality diagnostic steps" or simply as "diagnostic steps."

[0063] The personality diagnostic method is not limited to a configuration that includes all steps, but may include at least one step. Furthermore, the order of steps is not limited to the order shown in Figure 3, and the order of steps may be changed as appropriate as long as the order allows the results of the previous step to be used.

[0064] The personality diagnostic method can also be called a personality diagnostic program, in which case steps S100 to S190 of the personality diagnostic method become steps of the personality diagnostic program.

[0065] In a first step S100, the measurement unit 14 measures the time required for a swipe motion as a swipe time.

[0066] In the first step S100, the measurement unit 16 mainly measures the time required for the respondent to swipe on a touch panel screen as the swipe time. Specifically, the respondent answers questions displayed on the screen of a device or apparatus by swiping on the screen, and the time required for the swipe is measured. A swipe refers to the action of the respondent sliding their finger across the touch panel screen while touching it. Details are as described in the personality diagnostic system 10, and will not be repeated here.

[0067] In the second step S110, the determination unit 18 determines the strength of the response based on the measurement result obtained in the first step S100.

[0068] In the second step S110, the determination unit 18 mainly determines the answer strength based on the measurement result by the measurement unit 16 in the first step S100, for example. Specifically, answer strength determination data or an answer strength determination table for determining the answer strength is stored in the storage unit 14, and the determination unit 18 determines the answer strength based on the measurement result by the measurement unit 16 using the answer strength determination data or the answer strength determination table. Note that the details are the same as those described in the personality diagnostic system 10, and therefore will not be repeated here.

[0069] In the third step S120, the time setting unit 12 sets a standard answer time for each question.

[0070] In the third step S120, the time setting unit 12 mainly outputs a question based on question data previously stored in the storage unit 14, and sets a standard answer time for the respondent to answer the question. The standard answer time set by the time setting unit 12 is stored in the storage unit 14. Details are the same as those described in the personality diagnostic system 10, and will not be repeated here.

[0071] In a fourth step S130, the standard answer time and the question content are stored in the storage unit 14 in association with each other.

[0072] In the fourth step S130, the memory unit 14 is the main device, and various data and information required for the personality diagnosis are stored in the memory unit 14. For example, it is possible to store the standard answer time set by the time setting unit 12 in association with the question content at that time. The memory unit 14 also stores personal information (address, name, age, etc.) and data related to attributes of the respondent. Details are the same as those described in the personality diagnosis system 10, and will not be repeated here.

[0073] In a fifth step S140, the determination unit 20 determines the direction of the swipe motion as the answer direction.

[0074] In the fifth step S140, the determination unit 20 mainly determines the direction of the swipe action by the respondent as the answer direction. The answer direction can also be called the swipe direction. The direction of the swipe action means the swipe direction. Details are as described in the personality diagnostic system 10, so they will be omitted here.

[0075] In a sixth step S150, the generation unit 22 generates personality data from the response strength and response direction.

[0076] In the sixth step S150, the generation unit 22 takes the lead in generating personality data of the respondent from the answer strength determined by the determination unit 18 and the answer direction determined by the discrimination unit 20. By composing the personality data from answer strength x answer direction, it is possible to obtain personality data (big data) composed of patterns for many respondents. Details are as described in the personality diagnostic system 10, and therefore will be omitted here.

[0077] In a seventh step S160, the personality data is linked to the respondent and stored in the storage unit 14. Here, the storage unit 14 is the main body, and various data and a large amount of information required for the personality diagnosis are stored in the storage unit 14.

[0078] An eighth step S170 determines the strength of the response based on the remaining time obtained by subtracting a predetermined time from the swipe time.

[0079] The eighth step S170 is mainly performed by the determination unit 18 and reinforces the second step S110. As shown in FIG. 2, the determination unit 18 preferably determines the strength of the answer based on the remaining time obtained by subtracting a predetermined time (play time) from the swipe time. This makes it possible to exclude the time taken by the respondent to start reading and understand the question, and by measuring the true time taken to reach the answer, the strength of the answer can be determined more accurately. Note that details are the same as those described in the personality diagnostic system 10, and will not be repeated here.

[0080] In a ninth step S180, the time setting unit 12 sets an abnormal response time for each question.

[0081] In the ninth step S180, for example, the time is set by the time setting unit 12. The time setting unit 12 sets the standard answer time, but can also set an abnormal answer time at the same time. An abnormal answer time means a time that is clearly too short or a time that is clearly too long. It is preferable to set an optimal abnormal answer time for each question content. Details are as described in the personality diagnostic system 10, so they will not be described here.

[0082] A tenth step S190 sets a threshold value of the response filter for determining that the response strength is invalid if the swipe time becomes an abnormal response time.

[0083] The tenth step S190 is mainly performed by the filter setting unit 24. If the swipe time is an abnormal answer time, the filter setting unit 24 sets a threshold value for the answer filter for determining that the answer strength is invalid, thereby excluding the question content and answer content from the personality data. The filter setting unit 24 can also set an abnormal flag for the data file of the question content and answer content for which the swipe time is an abnormal answer time, indicating that the value is an abnormal value. Note that details are as described in the personality diagnostic system 10, and therefore will not be described here.

[0084] In the tenth step S190, the filter setting unit 24 may set an abnormal answer time in advance and remove question content and answer content for which the swipe time is determined to be an abnormal answer time. The function of the filter setting unit 24 may also be implemented as an answer filter function.

[0085] Among the steps of the personality diagnostic method and personality diagnostic program, a method consisting of at least the first step S100 and the second step S110 is called a personality measurement method, and a program for executing the method is called a personality measurement program. Furthermore, the personality measurement method and the personality measurement program may further include an additional fifth step S140 and a sixth step S150.

[0086] According to the first embodiment of the present invention, reliable personality data of respondents can be obtained, which makes it possible to resolve mismatches between individuals and organizations in advance, for example.

[0087] [First Example] Next, examples of the personality diagnostic system, the personality diagnostic method, and the personality diagnostic program according to the first embodiment of the present invention will be described with reference to the drawings.

[0088] (Definition of personality data) Personality data refers to information specific to a user that is obtained from application software, such as thoughts, interests, concerns, and decision-making speed.

[0089] (Characteristics of personality data) Personality data is not static and changes over time, and the range of change can be interpreted as part of the personality data.

[0090] (Personality diagnostic scheme) As shown in Figure 4, personality test questions are displayed on the screen of a device (also called a terminal) 30 such as a smartphone, and the user (also called a respondent or test-taker) swipes on the screen to select an answer. By swiping, data including the answer direction (swipe direction) and answer strength (time required to swipe) is stored in a database 34 via an answer filter 32.

[0091] The reply filter 32 is an example of the filter setting unit 24 of the embodiment. The database 34 is an example of the storage unit 14 of the embodiment.

[0092] A valid answer time range is set for each question, and the answer results within the valid time range and the time required to answer are recorded as personality data. At this time, the time required for the user's swipe action (swipe time) is used to determine the respondent's personality.

[0093] (Personality diagnostic test procedure) As shown in FIG. 5, the user starts taking a personality test on, for example, application software installed on the device (S200).

[0094] The user answers the questions in the personality diagnostic test by, for example, swiping left or right on a card on which the questions are written (S210). The card is displayed on the screen of the device, for example.

[0095] Here, if the user swipes to the right on the screen, the answer is that the right option on the card is more important than the left option. In this case, the direction in which the user swipes is the answer direction, and the time it takes to swipe is the answer strength, and this combination is the answer result for that question. For each question, an answer filter 32 is applied that determines that an answer that takes an unnaturally short or long time to swipe is an invalid answer (S220). The threshold of the answer filter 32 can be changed depending on the difficulty level and the amount of text of each question.

[0096] The personality data that has penetrated the answer filter 32 is stored in the database 34 in association with the user (examinee) (S230).

[0097] The personality data is collected for a predetermined number of axes and used as the personality data of the individual (S240). The personality data is used, for example, in the talent matching of the second embodiment.

[0098] Conventional personality diagnostic tests collect users' continuously changing personality data in a multiple-choice format, which results in missing information when the data is collected, reducing the reliability of the user's personality data.

[0099] To address these issues, as shown in Figure 6, by acquiring the time it takes to make a decision, which is negatively correlated with the strength of the user's will, as raw data without clustering, it is possible to build a data infrastructure that includes information about the user's personality and prevents any gaps.

[0100] As shown in Figure 7, in order to understand the situation at the time of the user's answer in detail, not only the swiping situation such as the swipe speed, but also information about the surrounding environment of the device (terminal) 30 such as time information and acceleration information may be obtained and used.

[0101] As shown in Figure 8, as raw data, the coordinate movement on the device 30 as a result of the swipe may be used as the answer direction, and the total time from the display of the swipe target to the completion of the swipe may be obtained as the answer strength, and these may be collected into a single object and passed to the answer filter 32.

[0102] As shown in FIGS. 9 and 10, the logic of the answer filter 32 may have question-specific recognition time and answer difficulty as main variables.

[0103] The answer filter 32 may use, as a secondary variable, not only the above-mentioned question-specific information but also the average answer time required for each user to answer.

[0104] As shown in Figure 11, the average response time required for each user is not a simple average, but a basic response time that can vary depending on the surrounding information and time of day when taking the test. The basic response time can also be achieved by providing an independent learning function to the logic.

[0105] As shown in Figure 12, it is possible to apply the personality axis included in the questions of each question item, but excessive averaging can lead to the risk of homogenizing the personality data. For this reason, it is preferable to be careful when using it within the range of the answer filter 32.

[0106] 13, data that did not detect the answer filter 32 may be held as archived data without being physically deleted, in order to be used in the learning process of the answer filter 32. For this reason, the database 34 may store archived answer data DT2 in addition to the personality data DT1.

[0107] As shown in FIG. 14, if there are answer results for other tests that have the same personality data structure as the target to be acquired, they may be used as initial values ​​for tuning the parameters of the answer filter 32.

[0108] [Second embodiment] Next, the second embodiment of the present invention is a talent matching system, talent matching method and talent matching program that use personality data obtained by the personality measurement system, personality diagnostic system, personality measurement method, personality diagnostic method and personality diagnostic program of the first embodiment, and will be described with reference to the drawings.

[0109] [Background to the second embodiment] First, the background to the talent matching system, talent matching method, and talent matching program that utilize personality data according to the second embodiment will be described. In the so-called seller's market, the percentage of companies whose job offers are declined by new graduates is on the rise. Even when companies do hire new graduates, the number of new graduates who leave their jobs in the first year is on the rise every year. In this situation, while the costs of acquiring the talent that companies need in the new graduate market are on the rise every year, the number of new graduates who leave their jobs early after being hired is increasing, making it extremely difficult for companies to hire students, and new graduate recruitment has become a major management issue for companies. The main reason why students leave their jobs early is due to mismatched interpersonal relationships within the company, and more specifically, the inability to accurately grasp individual personalities. Therefore, there are high hopes for avoiding mismatches in human relationships beforehand in a company's student recruitment activities, as this will lead to continued satisfaction on both the company and the student side during the company's new graduate recruitment activities and after the recruitment, reduce early student turnover, and enable the company's management to proceed smoothly. In this embodiment, the present invention was arrived at based on the knowledge that this problem can be widely applied not only to the relationship between "students" and "companies" but also to the relationship building between "individuals" and "organizations," "individuals" and "communities," "individuals" and "individuals," "organizations" and "organizations," etc. These correspondences can be applied directly to the "one side" and "other side" of the present invention.

[0110] The talent matching system, talent matching method, and talent matching program utilizing personality data according to the second embodiment of the present invention utilize the personality data obtained in the first embodiment and the first example. In the second embodiment, optimal matching between one side's personality and the other side's personality is realized, but for ease of explanation, an example will be given in which one side is the "individual side" and the other side is the "organization side." However, this is not limited to optimal matching between an "individual side personality" and an "organization side personality," and can be similarly applied to optimal matching between an "individual side personality" and an "individual side personality," as well as optimal matching between an "organization side personality" and an "organization side personality."

[0111] A talent matching system, a talent matching method, and a talent matching program that utilize personality data according to a second embodiment of the present invention will be described with reference to the drawings. The talent matching system may be installed or downloaded into an existing housing (e.g., hardware such as a terminal or device, e.g., a mobile phone, a smartphone, or a personal computer) like software such as application software, or may be implemented by the hardware such as a terminal or device (e.g., a personal computer, a server, etc.) itself.

[0112] In detail, the talent matching system, talent matching method, and talent matching program according to the second embodiment of the present invention are devices, processes, or programs for proposing optimal matches between individuals and organizations by utilizing personality data obtained when a respondent performs a swiping action (also simply referred to as "swiping") on the screen of a terminal or device such as a smartphone.

[0113] [2-1. Talent Matching System] 15, the talent matching system 100 includes a system main body 102 having, for example, a storage unit 104 and a matching generation unit 106. The system main body 102 is communicatively connected to an individual side device 110 on which individual side application software 108 (appropriately referred to as "individual side application 108") is installed, and an organization side device 114 on which organization side application software 112 (appropriately referred to as "organization side application 112") is installed.

[0114] The storage unit 104 corresponds to the storage unit 14 in the first embodiment, and may be a single storage unit that performs both functions. The matching generation unit 106 corresponds to the generation unit 22 in the first embodiment, and may be a single generation unit that performs both functions.

[0115] Furthermore, the talent matching system 100 or the system main body 102 may have the same configuration as the personality diagnostic system 10, or may be a single system having the functions of both.

[0116] For ease of explanation, FIG. 15 illustrates a configuration in which each unit of the talent matching system 100 is provided on a single device, but this is not limiting. For example, a configuration in which each unit is distributed across multiple different devices or devices may be adopted. For example, a single device or device may be equipped with only the match generation unit 106, with the remaining units being located on a server or cloud external to the device or device. Alternatively, a single device or device may be equipped with only the memory unit 104, with the remaining units being located on a server or cloud external to the device or device. Furthermore, some of the units of the talent matching system 100 may be configured with components such as a CPU, and the remaining units may be secured by performing arithmetic processing using software programs or the like.

[0117] Here, the storage unit 104 may be configured with a RAM (Random Access Memory), a ROM (Read Only Memory), a storage, or the like.

[0118] The matching generation unit 106 may be configured, for example, as a central processing unit (CPU) or a control unit (controller, not shown), but is not limited to these configurations.

[0119] The storage unit 104 stores the personality data of individuals, organizations, etc. acquired by answering questions with a swipe action in the first embodiment. Furthermore, the storage unit 104 stores individual-side personality data of the personality data in which the population is an individual, and organization-side personality data of the personality data in which the population is an organization.

[0120] In detail, the personal side personality data obtained by an individual performing a swiping motion is input from the personal side application 108 installed on the personal side device 110. The personal side personality data is data consisting of the swipe direction and swipe time of the individual's swipe motion. Similarly, the organization side personality data obtained by an organization (such as an individual constituting the organization, a person in charge or a representative of the organization) performing a swipe motion is input from the organization side application 112 installed on the organization side device 114. The organization side personality data is data consisting of the swipe direction and swipe time of the organization's swipe motion. The individual side personality data and organization side personality data obtained as described above are stored in the storage unit 104.

[0121] The personality data of the organization is not necessarily limited to data acquired by swiping. For example, the personality data may be data collected from a survey conducted by the organization, attributes, policies, mission, objectives, philosophy, and identification of desired personnel.

[0122] The match generation unit 106 matches the individual personality data with the organization personality data, in other words, the match generation unit 106 may provide an optimal match between the individual personality data and the organization personality data.

[0123] Specifically, the matching generation unit 106 receives the individual side personality data input from the individual side device 110 and the organization side personality data input from the organization side device 114, and generates matching data at a predetermined timing. The matching data is output to the individual side device 110 via the individual side application 108, and is output to the organization side device 114 via the organization side application 112.

[0124] Here, the match generation unit 106 may use the individual-side personality data and the organization-side personality data as input data, and may output multiple matching models as recommended personality data using, for example, a machine-learned prediction model with multiple possible matching cases as training data. Specifically, the recommended personality data may be output to the individual-side application 108 and / or the organization-side application 112, and therefore may be acquired via the individual-side device 110 or the organization-side device 114.

[0125] Furthermore, the match generation unit 106 may search for and output content related to talent matching based on the recommended personality data.

[0126] [2-2. Talent Matching Methods and Programs] 16, the talent matching method includes, for example, a first step S300, a second step S310, a third step S320, a fourth step S330, a fifth step S340, a sixth step S350, and a seventh step S360. The steps may also be referred to as "talent matching steps" or simply "matching steps."

[0127] The talent matching method is not limited to a configuration that covers all steps, but may include at least one step. Also, the order of steps is not limited to the order shown in Figure 16, and the order of steps may be changed as appropriate as long as the order allows the results of the previous steps to be used.

[0128] The talent matching method may also be referred to as a talent matching program, in which case steps S300 to S360 of the talent matching method become steps of the talent matching program.

[0129] In a first step S300, the individual-side personality data, which is of the personality data and has an individual as the population, and the organization-side personality data, which is of the personality data and has an organization as the population, are stored in the storage unit 104.

[0130] In the second step S310, the individual's personality data is matched with the organization's personality data. The matching between the individual's personality data and the organization's personality data may be performed by a person in charge, or the matching generation unit 106 may have a matching function.

[0131] In the third step S320, the personal side personality data inputted, for example, from the personal side application 108 of the personal side device 110 and the organization side personality data inputted, for example, from the organization side application 112 of the organization side device 114 may be stored in the memory unit 104.

[0132] In a fourth step S330, the matching generation unit 106 receives the individual-side personality data and the organization-side personality data and generates matching data.

[0133] In a fifth step S340, the matching data is output to the personal side application 108 and the organization side application 112. This allows the matching data to be acquired via the personal side device 110 and the organization side device 114. Note that the matching data may be output to only one of the personal side application 108 and the organization side application 112.

[0134] In a sixth step S350, the matching generation unit 106 uses the individual-side personality data and the organization-side personality data as input data, and uses multiple possible matching cases as training data, for example, a machine-learned prediction model, to generate and output multiple matching models as recommended personality data. This allows matching data to be acquired via the individual-side device and the organization-side device. Note that the recommended personality data may be output to only one of the individual-side application 108 and the organization-side application 112.

[0135] Here, the recommended personality data may be output together with the matching data or in place of the matching data.

[0136] In a seventh step S360, the match generation unit 106 searches for and outputs content related to talent matching based on the recommended personality data.

[0137] According to the second embodiment of the present invention, an optimal matching model between an individual and an organization can be provided by utilizing reliable personality data of respondents. At the same time, since the matching generation unit 106 has its own AI function, the matching generation unit 106 can receive input of individual-side personality data and organization-side personality data, and generate and output recommended personality data using, for example, a machine-learned prediction model with multiple conceivable matching cases as training data. As a result, mismatches between, for example, individuals and organizations can be reliably resolved. Furthermore, since the matching generation unit 106 searches for and outputs content related to talent matching based on the recommended personality data, individuals or organizations can refer to and share optimal content.

[0138] [First Example] Next, examples of a talent matching system, a talent matching method, and a talent matching program according to a second embodiment of the present invention will be described with reference to the drawings.

[0139] (Overview of talent matching) In the field of human resources, this technology improves engagement and well-being for users (both individuals and organizations) by using machine learning, for example, to match content to users based on either or both the individual's personality and the organization's personality.

[0140] (Human Resource Matching Procedures) (1) Collect user personality data through a personal application (reference numeral 116 in FIG. 17) or an organizational SaaS (reference numeral 122 in FIG. 17). (2) Personality data (reference numerals 118 and 124 in FIG. 17) of individuals or organizations is stored in a database (reference numeral 120 in FIG. 17). (3) The personality data (reference numerals 118 and 124 in FIG. 17) stored in the database (reference numeral 120 in FIG. 17) is input into a matching generator (reference numeral 126 in FIG. 17), and the matching generator 126 outputs a matching of career and personnel-related content (reference numeral 128 in FIG. 17). (4) The matched content is delivered to the user using a personal application (reference numeral 116 in FIG. 17) or an organizational SaaS (reference numeral 122 in FIG. 17).

[0141] (Human Resources Matching Scheme) The logic of this embodiment is partially or entirely equipped with artificial intelligence. It inputs individual personality data and organization personality data, outputs personalities to be matched, and performs content matching based on the recommended personality data.

[0142] 17 , for example, from personal software 116, personality information 118 of an individual user is stored in database 120. Similarly, for example, from organizational software 122, personality information 124 of an organization is stored in database 120. In matching generator 126, content matching 128 is performed based on personality information 118 of the individual user and personality information 124 of the organization. The content matching result is output to personal software 116 and organizational software 122.

[0143] Note that the "individual software 116" in Fig. 17 corresponds to the "individual-side application 108" shown in Fig. 15. The "organization software 122" in Fig. 17 corresponds to the "organization-side application 112" shown in Fig. 15. The "matching generator 126" in Fig. 17 corresponds to the "matching generation unit 106" shown in Fig. 15.

[0144] (Human Resource Matching Procedures) Learning of artificial intelligence is performed as shown in Fig. 18. Personality data of individuals or organizations is prepared as input, and assumed matching examples are prepared as training data, and a model is trained based on multiple general learning methods (S400). Here, the input training data varies depending on the use case of the matching generator 126. The matching generator 126 may prepare multiple models depending on the use case and perform different learning. The matching generator 126 receives data that is the same in format as the data used for learning in S400, and outputs matching targets in the form of personality data (S410). The matching generator 126 searches for and outputs content based on the personality data (S420). Based on user feedback (FB), data to be used for learning in S400 is extracted, and by repeating S400 to S420, the accuracy of the model can be improved. Although supervised learning is given as an example here, other machine learning logics such as unsupervised learning or deep learning may also be applied. Furthermore, in each embodiment and example, not only machine learning but also various other models such as k-means and factor analysis may be used.

[0145] Figure 19 is a schematic diagram of machine learning using recommended personalities. As shown in Figure 19, individual user personality information 118 and organization personality information 124 are input, and a recommended personality 130 is determined. Based on the recommended personality, a matching generator 126 performs content matching 128. Then, feedback (FB) 132 is obtained from the user, and the various personality information 118, 124 is further updated.

[0146] Figure 20 is a schematic diagram of machine learning that uses evaluation scores. As a new output pattern, a personnel matching model that takes both personality information 118 and 124 to be compared as input and uses a matching objective variable (evaluation score) 134 for matching as output, rather than the similarity search for recommended personality 130 shown in Figure 19, is also required in lightweight use cases.

[0147] (Definition of best match) Next, the logic for defining the best match is described. Consider the match of an organization with individuals and sub-organizations within it for various purposes and contexts. The main factors that determine whether a match is a "good match" are as follows: (1) Training data showing that users reacted to a match as "good" (2) Typical organizational characteristics include, for example, a sales organization or a development organization, and are linked to the purpose and role of the organization. (3) Characteristics specific to an organization include, for example, the unique corporate culture of the organization, which are linked to the current state of the organization (see Figure 21).

[0148] The base of a model for "a certain organization under a certain purpose" is something that combines both general organizational characteristics and organizational-specific characteristics, and machine learning can be introduced based on user reactions to adjust the details of this base.

[0149] As shown in Figure 22, when the weighting functions applied to each of the n axes are analyzed at the point where a "good match" can be determined as a result of the adjustments, the weighting function parameters for those n items become the "individuality" of that organization within the entire organization.

[0150] (Logic for matching individuals (1-1 personality match)) Matching a candidate's personality with other people's personality data under certain conditions, or selecting the best match, for example, to match individuals with the right mentor or supervisor.

[0151] If there is insufficient training data for two personalities, we set an axis that emphasizes convergence or divergence that is appropriate for some context, rather than the correlation on the n-axis. For example, if we assume that it is desirable for extroversion in a sales organization to converge in the extroverted direction, then a match in this context would be strongly evaluated as a good match if it is in the extroverted direction for extroversion.

[0152] (Individual and organization matching algorithm (1-n matching algorithm)) As shown in Figure 23, this applies to cases within an organization where an individual is matched with a sub-organization within the organization, as well as cases outside the organization where an individual is matched with an organization. For example, this applies to the assignment of a new employee.

[0153] (1) Depending on the intended use, determine what constitutes a "good match" (see the definition of the best match). Determine the base model. (2) Network data is constructed based on the base model determined above for the set of personality data of the organization. The network model constructed varies depending on the type of base model selected. (3) When adding personal data as a new node to the constructed network structure, the system scans the individual or organization to be added, aiming to evaluate it as a better match. (4) The match result can be defined as the best match under the objective. If objective answer data such as on-site evaluations are available, machine learning can be used in the logic. (5) Personality data is time-series data, and therefore changes over time for individuals or organizations.

[0154] Some aspects, such as interests, change quickly, while others, such as those strongly influenced by innate traits, change slowly. If an individual experiences a paradigm shift in interests due to external factors, it is possible to predict the favorable timing of organizational change through time-series regression analysis. It is also possible to give personality data a self-learning function. Therefore, if sufficient training data can be obtained, the system can actively recommend matches at the appropriate time.

[0155] [Example of generating a personality image] Next, an example of generating a personality image in personality matching obtained by the personality measurement system, personality diagnostic system, personality measurement method, personality diagnostic method, and personality diagnostic program according to each embodiment of the present invention will be described.

[0156] A. Example of generation in relationship analysis By analyzing the relationships between each personality item and specific values ​​(KPIs) of the organization using analytical methods such as factor analysis, correlation analysis, and network analysis, we identify specific personality items that affect KPIs. We then combine these personality items to create a personality profile. Examples of KPIs are as follows: (A) Evaluation (a) Grades (c) Performance (D) Engagement (E) Stress (F) Other survey results

[0157] B. Example of generation by trend analysis The personality tendencies of a specific group of individuals are analyzed to identify personality items that are extremely strong or extremely weak. These personality items are then combined to generate a personality profile. Examples of groups are: (A) By position, such as executive officers, managers, and management (a) By KPI, such as top performers and high performers (c) By region or area, such as head office or branch office (d) By department, such as Sales Department 1, Administration Department, etc. (E) By job type, such as new sales, accounting, customer support, etc. (F) By time of joining and years of service for new employees, senior employees, etc. (g) By employment status, such as full-time employee, contract employee, part-time employee, or casual employee

[0158] C. Examples of generation by personal analysis The personality of a specific individual is analyzed to identify personality items that tend to be extremely strong or extremely weak. These personality items are then combined to generate a personality profile. Alternatively, all personality items are utilized. Examples of eligible individuals include: (A) Management (a) Corporate representative (b) Directors (c) Officer (d) Officials (e) Organizational representative (f) Branch Representative (g) Branch Representative (a) By KPI (a) Excellent grades (b) High performers (c) By role (a) Leader (b) New employees (c) Mentor (d) Mentee

[0159] [Examples of using personality matching] Next, examples of the use of personality matching obtained by the personality measurement system, personality diagnostic system, personality measurement method, personality diagnostic method, and personality diagnostic program according to each embodiment of the present invention will be described.

[0160] A. Recruitment Matching (A) Method We extract candidates from the pool who closely match the ideal personality profile of the organization, and then select them by matching them. We also provide recruitment services, temporary staffing, and placement services. (a) Effects This will prevent mismatches in recruitment, leading to improved employee retention and reduced early employee turnover.

[0161] B. Placement Matching (A) Method Employees from other organizations who closely match the organization's ideal personality profile are extracted and matched. (a) Effects Optimizing the overall allocation of personnel leads to improved productivity.

[0162] C. Personal Matching (A) Method Individuals with similar personalities are extracted and matched. (a) Effects Mismatches will be prevented by utilizing mentor and mentee selection and counseling.

[0163] D. Content Matching (A) Method Compare the personality of a specific individual with the organization's ideal personality and extract any discrepant personality items. Match with content that will change that specific personality. Enable the provision of personalized development methods and training content. (a) Effects By reducing the deviation from the ideal personality and bringing it closer to the ideal personality, we aim to improve the retention rate and productivity of existing personnel.

[0164] [Analysis results for area managers and their use in personnel selection] Next, we will explain the analysis results of area general managers and their use in personnel selection using the personality measurement system, personality diagnostic system, personality measurement method, personality diagnostic method, and personality diagnostic program according to each embodiment of the present invention.

[0165] (Current issues) Although area managers are selected based on their experience and intuition as store managers, the temperament and skills required of area managers are completely different, and they are not able to perform as expected.

[0166] (Purpose of the initiative) We clarified the relationship between "personality" and "evaluation" and considered whether this could be used to select candidates suitable for area general managers.

[0167] (Acquired data) (1) Personality survey This is a diagnosis that was independently developed by blankpad Inc., the applicant of this application, under the supervision of a professor of personality psychology at Waseda University, and aims to provide a multi-layered and multifaceted understanding not only at the individual level but also at the organizational level. (2) Evaluation The evaluation was on a five-point scale (C, B-, B, B+, A). (3) Stress survey This questionnaire was created based on the Simple Occupational Stress Questionnaire, which is based on recommendations from the Ministry of Health, Labor and Welfare, and is intended to assess stress levels. It is a multi-dimensional questionnaire that can simultaneously measure not only stress reactions but also work-related stress factors and modifying factors. In addition, psychological stress reactions can measure not only negative reactions but also positive reactions that contribute to improving performance.

[0168] (Analysis method) (1) Trend analysis We analyzed the common characteristics of all area general managers in terms of personality and stress. (2) Correlation analysis We analyzed the strength of the relationships between personality and evaluation, personality and stress, and evaluation and stress.

[0169] (Analysis results - personality trends) The current overall trend among area general managers is a practical and adaptable leadership style that excels in managing daily operations and building relationships. (Personality items)-(Score)-(Interpretation) (Strong Efficiency Consciousness) - (3.86 / 5) - (Prioritizes time efficiency and practicality over aesthetic sensibility) (Strength of respectful expressions) - (3.82 / 5) - (Shows respect and has a respectful attitude towards others) (Strength of Intuition) - (3.59 / 5) - (Emphasis on sensations and real-life experiences, judgments based on concrete facts) (Curiosity) - (3.55 / 5) - (Curious about new ideas and experiences and open to change) (comment) The score ranges from 0 to 5 and indicates how strong the product is. A score of 3.5 or above indicates a very strong tendency and is selected.

[0170] (Analysis results - correlation between personality and evaluation) "Broad perspective," "strong sense of deadlines," and "strong sense of responsibility" are three things that are strongly correlated with high evaluations of area general managers. (Personality items)-(Correlation coefficient with evaluation)-(Interpretation) (Broad Perspective) - (0.62) - (Places emphasis on seeing things from a broader perspective rather than a narrow one) (Strong sense of deadlines) - (0.62) - (Places importance on deadlines and achieving goals, and places importance on time management) (Strong sense of responsibility) - (0.59) - (Strong sense of responsibility and fulfills one's role well) (comment) Taking into account the total number of test takers, a correlation coefficient of 0.5 or higher is set as the standard value, and those exceeding this value in both evaluation and stress are selected as important personality items.

[0171] (Analysis results - current status of correlated personality items) Looking at the ratings for each personality category, area managers with high ratings (B+ or A) have higher scores than area managers with low ratings (C or B-). (Personality items)-(Average score by rating)-(Difference (A)-(B)) (Strength of responsibility) - (4.20 for a rating of B+ or A, 1.73 for a rating of C or B-) - (Area managers with high ratings have a +2.47) (Strength of deadline awareness) - (4.20 for a rating of B+ or A, 2.27 for a rating of C or B-) - (Area managers with high ratings have a +1.93) (Breadth of Vision) - (3.60 for a rating of B+ or A, 2.77 for a rating of C or B-) - (Area managers with high ratings have a +0.83 rating)

[0172] Next, the expected future uses of the personality measurement system, personality diagnostic system, personality measurement method, personality diagnostic method, and personality diagnostic program according to each embodiment of the present invention will be described.

[0173] [Expected future use] By incorporating into the selection process whether the area manager matches the personality traits required to receive high evaluations (= good performance), it becomes possible to select the right person. As a specific usage flow, when selecting an area general manager, the following flow is assumed, similar to that of a regular aptitude test. Diagnostic test: Area general manager candidates will take a personality test similar to this one. Check the match level in the report: Based on the selected axis, check the match level and the matching and non-matching points in the report. Confirmation during the selection process: Use this information when actually selecting from area manager candidates. The selection criteria are as follows: No. 4-5 were added because they showed a correlation with evaluation. No. 1-3 were given a stronger weight because they showed a correlation with both evaluation and stress. No.1 As a personality trait, strong sense of deadlines The interpretation is that they place importance on deadlines and achieving goals, and place importance on time management. No.2 As a personality trait, a strong sense of responsibility The interpretation is that they have a strong sense of responsibility and fulfill their roles well. No.3 As a personality item, breadth of perspective The interpretation is that it places emphasis on overseas activities and intercultural exchange. No.4 As a personality trait, high logical thinking ability The interpretation is that they are interested in natural science and technology, and place importance on logic and experiments. No.5 High creativity as a personality trait The interpretation is that it places importance on creative ideas and expression, and values ​​new ideas.

[0174] (Analysis results - correlation between stress and evaluation) Three traits that have a strong correlation with evaluation - "strong sense of responsibility," "strong sense of deadlines," and "broad perspective" - ​​also have a high correlation with low stress. (Personality items)-(Correlation coefficient with stress)-(Interpretation) (Strong sense of responsibility) - (0.63) - (Strong sense of responsibility and fulfills one's role well) (Strong sense of deadlines) - (0.62) - (Places importance on deadlines and achieving goals, and places importance on time management) (Broad Perspective) - (0.51) - (Places emphasis on seeing things from a broader perspective rather than a narrow one) (comment) Taking into account the total number of test takers, a correlation coefficient of 0.5 or higher is set as the standard value, and those that exceed this value in both evaluation and stress are selected as important personality items.

[0175] (Analysis results - correlation between personality and evaluation) In addition, "high logical thinking ability" and "high creativity" are two other factors that are highly correlated with high evaluations of area general managers. (Personality items)-(Correlation coefficient with evaluation)-(Interpretation) (High logical thinking ability) - (0.53) - (Interested in natural science and technology, and values ​​logic and experiments) (High creativity) - (0.52) - (Puts emphasis on creative ideas and expression, and values ​​new ideas) (comment) Taking into account the total number of test takers, items with a correlation coefficient of over 0.5 are deemed to have a strong correlation and are selected as personality items.

[0176] (Analysis results - stress trends) They have strengths such as good interpersonal relationships, a strong support system, and high job satisfaction. These factors are thought to support satisfaction and productivity even under high-stress conditions. On the other hand, excessive workload and the resulting stress reactions are major issues. (List of good items) - (Compared to national average) (Support from superiors)-(30%↑) (Stress due to work environment) - (26%↑) (Work satisfaction) - (21%↑) (Support from colleagues)-(20%↑)

[0177] (List of items requiring improvement) - (Compared to national average) (Perceived physical burden) - (32%↓) (Psychological job strain (quality)) - (26%↓) (Psychological work burden (amount)) - (24%↓) (comment) All points that were 20% above the national average were extracted and listed, divided into good points and points requiring improvement.

[0178] (Analysis results - Relationship between stress and evaluation) Overall, people with lower stress levels tended to give higher ratings. In particular, people who are able to maintain better mental and physical conditions (physical and mental responses caused by stress) tend to receive higher ratings. (Item list)-(Correlation coefficient) (Total stress score)-(0.62) (Physiological and mental responses caused by stress) - (0.60) (Factors considered to be causes of stress) - (0.54) (Other factors affecting stress) - (0.42) (comment) Taking into account the total number of test takers, a correlation coefficient of 0.5 or higher is set as the standard value, and those that exceed this value in both evaluation and stress are selected as important temperaments.

[0179] [List of stress items] (Physiological and mental reactions caused by stress) Lively Irritability Fatigue Anxiety Feeling depressed Physical complaints (Factors that may cause stress) Psychological workload (amount) Psychological job strain (quality) Perceived physical strain Stress in interpersonal relationships at work Stress from the work environment Degree of control over work Sense of skill utilization ·Job suitability Job satisfaction (Other factors that affect stress) Support from your superiors Support from colleagues Support from family and friends Satisfaction

[0180] The above-described embodiments are merely examples of the present invention, and any design modifications that can be made by a person skilled in the art are included within the scope of the present invention. [Explanation of symbols]

[0181] 10 Personality Diagnostic System 11 Personality Measurement System 12 Time setting section 14 Storage section 16 Measurement section 18 Judgment section 20 Discrimination part 22 Generation part 24 Filter setting section 30 devices 32 Answer filter (filter setting section) 34 Database (storage section) 100 Talent Matching System 102 System main body 104 Storage section 106 Matching Generation Unit 108 Personal Application Software 110 Personal Devices 112 Organizational Application Software 114 Organizational Device 116 Personal Software 118 Personal User Information 120 Database (storage section) 122 Organizational Software 124 Organizational Personality Information 126 Matching Generator 128 Content Matching 130 Recommended Personalities 132 User Feedback (FB) 134 Matching objective variable (evaluation score)

Claims

1. A human resources matching system that utilizes personality data acquired by answering questions with a swipe action, a storage unit that stores one-side personality data that is a population on one side of the personality data and other-side personality data that is a population on the other side; a matching generation unit that matches the one-party personality data with the other-party personality data; A talent matching system that utilizes personality data.

2. a one-side application for inputting the one-side personality data; an other-side application for inputting the other-side personality data; and 2. A human resources matching system utilizing personality data as described in claim 1, wherein the one-side personality data input from the one-side application and the other-side personality data input from the other-side application are stored in a memory unit.

3. 3. The human resources matching system utilizing personality data as described in claim 2, wherein the matching generation unit receives input of the one-side personality data and the other-side personality data, generates matching data, and outputs it to the one-side application and the other-side application, respectively.

4. The human resources matching system utilizing personality data as described in claim 3, wherein the matching generation unit uses the one-side personality data and the other-side personality data as input data, and outputs multiple matching models as recommended personality data using a predictive model trained using multiple conceivable matching cases as training data.

5. The talent matching system utilizing personality data according to claim 4 , wherein the matching generation unit searches for and outputs content relating to talent matching based on the recommended personality data.

6. A human resources matching method utilizing personality data acquired by answering questions with a swipe action, a first step of storing one-side personality data, which is a population on one side of the personality data, and another-side personality data, which is a population on a local side of the personality data; a second step of matching the one-party personality data with the other-party personality data; A personnel matching method that utilizes personality data.

7. a one-side application for inputting the one-side personality data; an other-side application for inputting the other-side personality data; Using 7. The personnel matching method utilizing personality data according to claim 6, further comprising a third step of storing the one-side personality data input from the one-side application and the other-side personality data input from the other-side application.

8. a fourth step of receiving the one-party personality data and the other-party personality data to generate matching data; a fifth step of outputting the matching data to the one-side application and the other-side application, respectively; The personnel matching method utilizing personality data according to claim 7, comprising:

9. The human resource matching method utilizing personality data as described in claim 8, further comprising a sixth step of using the one-side personality data and the other-side personality data as input data, and using a predictive model trained using multiple conceivable matching cases as training data, to output multiple matching models as recommended personality data.

10. 10. The talent matching method utilizing personality data according to claim 9, further comprising a seventh step of searching for and outputting content relating to talent matching based on the recommended personality data.

Images