Information processing program, information processing method, and information processing device

Abstract

To provide an information processing program, an information processing method and an information processing device that make personnel transfer work efficient.SOLUTION: The information processing program causes a computer to execute the following processing. One processing is processing to receive specification of a specific personnel among a plurality of personnel assigned to a specific assignment destination. One processing is processing to acquire a plurality of kinds of attribute information on effective personnel excluding the information on the specific personnel specified among the plurality of personnel. One processing is processing to estimate first importance of each of the plurality of kinds of attribute information through machine learning using training data including the plurality of kinds of attribute information of the effective personnel, and generate and output, based upon a plurality of kinds of attribute information on a plurality of personnel transfer candidates as candidates to be assigned to the specific assignment destination, information indicating the suitability of the plurality of personnel transfer candidates for the specific assignment destination by using the estimated first importance.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to an information processing program, an information processing method, and an information processing device. [Background technology]

[0002] In various organizations such as companies, personnel transfer operations occur periodically in order to revitalize and streamline the organization. In particular, public organizations and other organizations have personnel transfer events annually, with a certain number of employees being reassigned. The process for determining where to reassign personnel involves identifying those to be transferred, compiling self-reports, and interviewing their department heads, followed by the creation and adjustment of a transfer plan to change the employee's placement.

[0003] When preparing transfer proposals for personnel transfer work, it is important to comprehensively evaluate the suitability of each candidate for a vacant post from multiple perspectives. For example, perspectives include whether the transfer candidate's attributes are similar to the requirements of the vacant post, whether the post meets the transfer candidate's transfer wishes, and whether the transfer candidate is suitable for the characteristics of the employees in each department. The attributes of transfer candidates include personnel items such as age, performance, and overtime hours. Furthermore, the characteristics of the employees in each department include their transfer history, such as the types of departments they have been to and in what order.

[0004] In the task of creating transfer plans, which requires evaluation from multiple perspectives, when there are a large number of transfer candidates and a large number of target departments, there are many factors to consider, and the process of determining the combination of transfers is a heavy burden. Furthermore, the know-how of transfers is often treated as tacit knowledge, and it is not easy for new employees involved in creating transfer plans to utilize such know-how. For these reasons, creating transfer plans is one of the most demanding tasks for human resources personnel. Therefore, it is desirable to make improvements to the process of creating transfer plans that make it easier to process and increase efficiency.

[0005] In addition, as a technology related to replacing hardware in a system, a technology has been proposed in which, when analyzing the differences between the current system and the new system, parts of the application that have been modified are excluded from the analysis target. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2012-203580 Summary of the Invention [Problem to be solved by the invention]

[0007] However, when creating transfer proposals, it is extremely difficult to uniformly define multiple criteria for evaluating the suitability of transfer candidates for vacant positions across an organization, making complete automation impractical. This is because different criteria exist depending on the department, job type, etc. In other words, manual adjustment of parameters is unavoidable, and the ease of this adjustment is an important factor for streamlining the process. In the conventional process of creating transfer proposals, human resources personnel evaluate candidates based on multiple criteria for each department and job type, taking into account past performance in personnel transfers and their own experience, and then determine the placement combinations. This makes it difficult to streamline the process of creating transfer proposals.

[0008] Furthermore, while technology that analyzes the causes of differences between the old and new systems by excluding from the analysis the parts of the application that have been modified makes it easier to analyze the causes of the differences, it is difficult to facilitate evaluation from multiple perspectives.

[0009] The disclosed technology has been made in consideration of the above, and aims to provide an information processing program, an information processing method, and an information processing device that improve the efficiency of transfer plan creation work. [Means for solving the problem]

[0010] In one aspect of the information processing program, information processing method, and information processing device disclosed herein, a computer is caused to execute the following processes. One is a process of accepting designation of a specific person from among multiple personnel assigned to a specific assignment destination. Another is a process of acquiring multiple types of attribute information of available personnel excluding information on the designated specific person from among the multiple personnel. Another is a process of estimating a first importance for each of the multiple types of attribute information by machine learning using training data including the multiple types of attribute information of the available personnel, and generating and outputting information indicating the suitability of the multiple transfer candidates for the specific assignment destination using the estimated first importance based on the multiple types of attribute information of the multiple transfer candidates who are candidates for assignment to the specific assignment destination. [Effects of the Invention]

[0011] In one aspect, the present invention can improve the efficiency of the transfer plan creation work. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram of a transfer candidate recommending device according to an embodiment. [Figure 2] FIG. 2 is a diagram showing an outline of the transfer candidate recommendation process. [Figure 3] FIG. 3 is a diagram for explaining the scoring method used in the examples. [Figure 4] FIG. 4 is a diagram showing an example of the transfer candidate recommendation screen. [Figure 5] FIG. 5 is a flowchart of a transfer candidate recommendation process performed by the transfer candidate recommendation device according to the embodiment. [Figure 6] FIG. 6 is a hardware configuration diagram of the transfer candidate recommending device. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the information processing program, the information processing method, and the information processing device disclosed in the present application will be described in detail with reference to the accompanying drawings. Note that the information processing program, the information processing method, and the information processing device disclosed in the present application are not limited to the following embodiments. [Example]

[0014] Fig. 1 is a block diagram of a transfer candidate recommendation device according to an embodiment. The transfer candidate recommendation device 1, which is an information processing device, is a device for assisting a user in creating a transfer plan by providing the user with an evaluation result of the suitability of a transfer candidate for an assignment based on personnel data input from a terminal device 2. Fig. 2 is a diagram showing an outline of a transfer candidate recommendation process. Here, an outline of the transfer candidate recommendation process performed by the transfer candidate recommendation device 1 will be described using Fig. 2.

[0015] The transfer candidate recommendation device 1 acquires personnel data 20. The personnel data 20 includes information on the assigned post, and the name and attributes of each transfer candidate. The attributes of the transfer candidate include personnel items such as age, performance, and overtime hours. Upon receiving the input of the personnel data 20, the transfer candidate recommendation device 1 executes the following recommendation process (step S1).

[0016] The transfer candidate recommendation device 1 searches for the initial values ​​of each parameter used in the transfer candidate recommendation process from past personnel transfer information (step S11). The past personnel transfer information includes information on employees who were assigned to their assigned posts in the previous personnel transfer, excluding employees who were assigned as a result of compromise. The parameters used in the transfer candidate recommendation process include the requirements for the assigned post, information used in each of the multiple scoring methods to be used, the importance of each scoring method, and the importance of each piece of information used in each scoring method. Next, the transfer candidate recommendation device 1 performs unsupervised learning using the information on transfer candidates included in the personnel data 20, and calculates a score for each transfer candidate using a recommendation model for each scoring method (step S12).

[0017] Thereafter, the transfer candidate recommendation device 1 provides the calculated scores for each transfer candidate and notifies the user P of the information on the transfer candidates recommended for the assigned post (step S2). The user P can change the scoring methods and the importance of each piece of information used in each scoring method from the initial value. When the importance is changed, the transfer candidate recommendation device 1 recalculates the score using the changed importance and provides the recalculated score to the user P.

[0018] User P refers to the scores provided by the transfer candidate recommendation process executed by the transfer candidate recommendation device 1 and decides which staff member to assign to the destination post from among the transfer candidates. User P then inputs the assignment results and, if any staff member whose assignment was decided upon was decided upon through compromise, information indicating that the assignment is a compromise assignment to the transfer candidate recommendation device 1. The transfer candidate recommendation device 1 stores the assignment results and the information indicating that the assignment is a compromise assignment input by user P. The transfer candidate recommendation device 1 then uses the stored assignment results and information indicating that the assignment is a compromise assignment in the next initial value search process (step S11).

[0019] Next, we will explain the details of the transfer candidate recommendation device 1. As shown in Figure 1, the transfer candidate recommendation device 1 has a display unit 11, an input unit 12, an input / output control unit 13, a memory unit 14, a score calculation unit 15, and an initial value calculation unit 16.

[0020] 3 is a diagram illustrating the scoring methods used in the embodiment. In this embodiment, three scoring methods are used to calculate the degree of suitability of each candidate for the post to which they will be assigned: a scoring method 31 using candidate equivalence, a scoring method 32 using the presence or absence of the candidate's preference, and a scoring method 33 using transfer history. The importance of each of the scoring methods 31 to 33 varies depending on the post to which the candidate will be assigned.

[0021] Scoring method 31 is a method that uses a recommendation model to calculate a score based on the proximity to the assigned position when personnel items are taken into consideration comprehensively. For example, scoring method 31 calculates a score representing the suitability of each transfer candidate based on the difference between the average attributes of the transferees who transferred from the assigned position and the attributes of each transfer candidate. Here, scoring method 31 uses multiple personnel items as attribute information used in calculating the score. For example, personnel items used in scoring method 31 include age, performance, and overtime hours. Figure 3 shows a case in which scoring method 31 calculates a score using three evaluation items. Since the abilities required for each assigned position differ, the importance of these evaluation items also differs depending on the assigned position.

[0022] Scoring method 32 is a method that uses a recommendation model to obtain a score that represents the suitability of a transfer candidate based on the candidate's transfer preferences. For example, if the desired position in the self-reported form matches the position, a certain number of points are added to the score according to the order of the preferences. Specifically, if the candidate has three transfer preferences, the score is added according to the order of the preferences so that the first preference is the highest and the third preference is the lowest.

[0023] Scoring method 33 is a method that uses a recommendation model to calculate a score based on the positions held in the past. For example, it is possible that experience in other specific positions can be utilized in the position to which an employee is assigned. Therefore, a score for the position is calculated based on the employee's past affiliation history.

[0024] Returning to FIG. 1, the explanation of each component will be continued. The display unit 11 is a monitor or the like. The input unit 12 is a device for inputting information, such as a keyboard or a mouse. A user P refers to the display unit 11 and inputs information to the transfer candidate recommendation device 1 using the input unit 12.

[0025] The input / output control unit 13 receives input of personnel data 20 from the terminal device 2. Then, the input / output control unit 13 acquires information on the assignment post from the personnel data 20 and outputs it to the initial value calculation unit 16. The information on the assignment post includes information identifying the post and information on the personnel items of the person who has transferred out of that post. Thereafter, the input / output control unit 13 receives input of various initial value information from the initial value calculation unit 16, including information on the initial values ​​of each personnel item that is a post requirement according to the assignment post, the initial values ​​of the importance of each scoring method 31 to 33, and the initial value of the importance of each personnel item used in the scoring method 31.

[0026] Furthermore, the input / output control unit 13 acquires information on each transfer candidate, including information on the personnel items of each transfer candidate, information on the transfer preferences of each transfer candidate, and information on the transfer history of each transfer candidate, from the personnel data 20. Then, the input / output control unit 13 outputs the acquired information on each transfer candidate to the score calculation unit 15. Thereafter, the input / output control unit 13 receives input of the score of each transfer candidate from the score calculation unit 15.

[0027] 4 is a diagram showing an example of a transfer candidate recommendation screen. The input / output control unit 13 generates the transfer candidate recommendation screen 100 shown in FIG. 4 using the acquired information of various initial values ​​and the scores of each transfer candidate, and displays it on the display unit 11.

[0028] The transfer candidate recommendation screen 100 has arranged thereon a post requirement table 101 showing the requirements for the post at the assigned location and an aptitude evaluation table 102 showing the results of the aptitude evaluation of each transfer candidate. Furthermore, the transfer candidate recommendation screen 100 has arranged thereon a mixer 103 for adjusting the importance of each scoring method 31 to 33, a mixer 104 for adjusting the importance of each personnel item used in the scoring method 31, and a save button 105.

[0029] The input / output control unit 13 provides columns for planned transfer year 111, post department name 112, and personnel items 113 to 115 in the post requirement table 101. The input / output control unit 13 registers information about the post to which the employee will be assigned in the post department name 112. The unit of the post to which the employee will be assigned may be a section, a department, or a more detailed unit, as shown in FIG. 4. The input / output control unit 13 also registers initial values ​​of personnel items that are requirements for the post to which the employee will be assigned in the personnel items 113 to 115.

[0030] Input / output control unit 13 also adjusts the positions of knobs 131 to 133 of mixer 103 so as to indicate the initial values ​​of the importance of each scoring method 31 to 33. Furthermore, input / output control unit 13 adjusts the positions of knobs 141 to 143 of mixer 104 so as to indicate the initial values ​​of the importance of each personnel item used in scoring method 31.

[0031] The input / output control unit 13 also provides columns for ranking 121, identification information 122, personnel items 123 to 125, score 126, placement decision status 127, and compromise placement information 128 in the suitability evaluation table 102. The identification information 122 is information for identifying transfer candidates, and in this embodiment, names are registered. The input / output control unit 13 registers the names of each transfer candidate in the identification information 122 in order of score according to the ranking 121. The input / output control unit 13 also registers the values ​​of the personnel items of each transfer candidate that correspond to the personnel items that are requirements for the post in the personnel items 123 to 125. Furthermore, the input / output control unit 13 registers the score of each transfer candidate calculated by the score calculation unit 15 in the score 126.

[0032] The input / output control unit 13 performs the above registration and generates the initial screen of the transfer candidate recommendation screen 100. Then, the input / output control unit 13 causes the display unit 11 to display the generated initial screen of the transfer candidate recommendation screen 100.

[0033] Thereafter, when one or more of the knobs 131 to 133 of the mixer 103 and the knobs 141 to 143 of the mixer 104 are moved from the input unit 12 to input a change in importance, the input / output control unit 13 outputs information on each changed importance to the score calculation unit 15. Then, the input / output control unit 13 requests the score calculation unit 15 to recalculate the scores. Then, the input / output control unit 13 receives input of the recalculated scores of each transfer candidate from the score calculation unit 15. Then, the input / output control unit 13 moves the knobs 131 to 133 and 141 to 143 on the transfer candidate recommendation screen 100 to the specified positions, and registers the newly calculated scores of each transfer candidate in the score 126 field. At this time, if a change in the score causes a change in the ranking, the input / output control unit 13 corrects the ranking of the transfer candidates in accordance with the new scores. Then, the input / output control unit 13 causes the display unit 11 to display a newly generated transfer candidate recommendation screen 100 in accordance with the input importance.

[0034] Furthermore, when information indicating the decision to assign a specific transfer candidate is input from the input unit 12, the input / output control unit 13 registers information indicating the selection in the assignment decision status 127 field corresponding to the selected specific transfer candidate. Furthermore, when information indicating a compromise assignment for one of the selected transfer candidates is input from the input unit 12, the input / output control unit 13 sets a flag indicating a compromise assignment in the compromise assignment information 128 field corresponding to that transfer candidate. The input of information indicating a compromise assignment for this assignment post to the input / output control unit 13 is an example of "accepting the designation of a specific person among multiple people assigned to a specific assignment destination." Then, the input / output control unit 13 displays a newly generated transfer candidate recommendation screen 100 on the display unit 11 in accordance with the input information indicating the decision to assign and the information indicating the compromise assignment.

[0035] Furthermore, when the save button 105 is pressed using the input unit 12 after an input has been made in the placement determination status 127 field, the input / output control unit 13 stores in the memory unit 14 the information on the importance and importance of each scoring method 31 to 33 represented by the mixers 103 and 104 at that time, and the information on the importance of each personnel item used in the scoring method 31. Furthermore, the input / output control unit 13 selects transfer candidates to whom information indicating selection has been added in the placement determination status 127 field, but to whom a flag indicating compromise placement has not been added in the compromise placement information 128 field, and stores in the memory unit 14 the information shown in the personnel items 123 to 125 of each selected transfer candidate.

[0036] The storage unit 14 stores and accumulates the results of personnel transfers carried out each year, i.e., which transfer candidates were assigned to which posts, and the attributes of each assigned transfer candidate, including personnel items. The storage unit 14 also stores and accumulates the importance of each of the scoring methods 31 to 33 set for each year and the importance of each personnel item used in the scoring method 31. The storage unit 14 also stores the types of personnel items that are post requirements to be evaluated by the scoring method 31 for each assigned post.

[0037] The initial value calculation unit 16 receives input of information about the assignment post for which a transfer candidate is to be recommended from the input / output control unit 13. Next, the initial value calculation unit 16 acquires information about the personnel items of each out-transferee who has transferred out of that post from the information about the assignment post. Furthermore, the initial value calculation unit 16 acquires the types of personnel items that are the post requirements for that assignment post from the storage unit 14. Then, the initial value calculation unit 16 calculates the average value of each out-transferee who has transferred out of that post for each type of personnel item, and sets this as the initial value of the personnel item that is the post requirement for that assignment post.

[0038] Additionally, the initial value calculation unit 16 calculates the initial value of the importance of each of the scoring methods 31 to 33 according to the assignment post to which the transfer candidate is to be recommended. Below, a method for calculating the initial value of the importance of each of the scoring methods 31 to 33 by the initial value calculation unit 16 will be explained. Hereinafter, the assignment post to which the transfer candidate is to be recommended will be referred to as the "transfer target post."

[0039] The initial value calculation unit 16 groups each department based on past data such as the values ​​of personnel items of employees in each department. For example, the initial value calculation unit 16 acquires this information from a database of the personnel department (not shown). Next, the initial value calculation unit 16 acquires information on personnel items, transfer requests, and transfer history of employees who were actually assigned to each group in the past. Then, the initial value calculation unit 16 calculates a combination of the importance of each scoring method 31 to 33 for each group that will result in the most accurate score, i.e., so that the scores of employees who were actually assigned will be as high as possible. This calculated importance is called the scoring importance based on past data.

[0040] Next, the initial value calculation unit 16 acquires the initial value of the importance of each of the scoring methods 31 to 33 for the transfer target post after manual adjustment in the previous personnel transfer from the storage unit 14. The acquired initial value of the importance is called the initial value of the previous scoring importance.

[0041] Next, the initial value calculation unit 16 acquires information on effectively assigned staff who were reassigned in personnel changes in the previous year and to whom no compromise placement information has been added. Next, the initial value calculation unit 16 performs unsupervised machine learning for each group of departments to calculate a combination of the importance of each scoring method 31 to 33 so that the accuracy of the score is the highest, i.e., so that the scores of effectively assigned staff are as high as possible. This calculated combination is called the scoring importance calculated excluding compromise placement.

[0042] Thereafter, the initial value calculation unit 16 calculates the average of the scoring importance calculated excluding the scoring importance based on past data, the previous initial value of the scoring importance, and the compromise arrangement.The initial value calculation unit 16 then sets the calculated value as the current initial value of the importance of each scoring method 31 to 33 for the transfer target post.The importance of each scoring method 31 to 33 for this transfer target post is an example of the "second importance."

[0043] Furthermore, the initial value calculation unit 16 calculates an initial value of importance for each of the personnel items that are the post requirements used in the scoring method 31 for the transfer target post. Hereinafter, the personnel items that are the post requirements used in the scoring method 31 for the transfer target post will be referred to as "personnel items for the transfer target post." Below, a method for calculating the initial value of importance for each of the personnel items for the transfer target post by the initial value calculation unit 16 will be described.

[0044] The initial value calculation unit 16 acquires the values ​​of personnel items of out-transferees at the time of transfer in each group obtained by grouping each department, and calculates the average value of each personnel item. The initial value calculation unit 16 also acquires the values ​​of personnel items of in-transferees in each group from the storage unit 14, and calculates the average value of each personnel item. Next, the initial value calculation unit 16 calculates the difference between the average value of out-transferees and the average value of in-transferees for each personnel item for each group. Here, the closer the difference between the average value of out-transferees and the average value of in-transferees is to 0, the more importance is placed on equivalent exchange in the previous transfer.

[0045] Therefore, the initial value calculation unit 16 checks the standard deviation of the differences for each personnel item. Next, for personnel items with a large standard deviation of the differences, the initial value calculation unit 16 assumes that the degree of equivalent exchange is low and lowers the importance from the predetermined importance value. Also, for personnel items with a small standard deviation of the differences, the initial value calculation unit 16 assumes that the degree of equivalent exchange is high and raises the importance from the predetermined importance value. Thereafter, the initial value calculation unit 16 normalizes the importance of each personnel item to align the scales of the personnel items. In this way, the initial value calculation unit 16 determines the importance of each personnel item for each group. Thereafter, the initial value calculation unit 16 obtains the importance of each personnel item of the transfer target post from the initial values ​​of the importance of each personnel item of the group to which the transfer target post belongs. The obtained importance is called the personnel item importance based on past data.

[0046] Furthermore, the initial value calculation unit 16 acquires from the storage unit 14 the final importance value of each personnel item after manual adjustment at the time of the previous transfer for the transfer target post. This acquired initial importance value is called the previous initial importance value of personnel item. This final importance value of each personnel item after manual adjustment at the time of the previous transfer is an example of the "previous change result of the first importance."

[0047] Additionally, for each group of departments, the initial value calculation unit 16 acquires information on effectively assigned staff, which are staff who were reassigned in a personnel transfer in the previous year and to whom compromised placement information is not attached. This acquisition of information by the initial value calculation unit 16 is an example of "acquiring multiple types of attribute information on effective staff, excluding information on specified specific staff among multiple staff." In this case, staff who were reassigned in a personnel transfer in the previous year and to whom compromised placement information is not attached are an example of "effective staff." Next, the initial value calculation unit 16 calculates the difference between the average value of out-transferees and the average value of effectively assigned staff for each personnel item of the transfer target post. Next, the initial value calculation unit 16 calculates the standard deviation of the difference and performs unsupervised machine learning to adjust the importance so that the larger the standard deviation of the difference, the lower the importance and the smaller the standard deviation, the higher the importance. After performing this adjustment, the initial value calculation unit 16 calculates a combination of importance for each personnel item of the transfer target post for each group so that the accuracy of the importance is maximized, i.e., so that the scores of effectively assigned staff are as high as possible. This calculated combination is called the personnel item importance calculated excluding compromise placement.

[0048] Thereafter, the initial value calculation unit 16 calculates the personnel item importance based on past data, the previous initial value of the personnel item importance, and the average of the personnel item importance calculated excluding compromise placement.The initial value calculation unit 16 then sets the calculated value as the current initial value of the importance of each personnel item for the transfer target post.The importance of each personnel item for the transfer target post is an example of the "first importance."

[0049] Thereafter, the initial value calculation unit 16 outputs information on the initial values ​​of each personnel item of the transfer target post, the current initial values ​​of the importance of each scoring method 31 to 33 for the transfer target post, and the current initial values ​​of the importance of each personnel item of the transfer target post to the input / output control unit 13. This initial value calculation unit 16 is an example of an "importance calculation unit."

[0050] The score calculation unit 15 acquires information on each transfer candidate, including information on the personnel items of each transfer candidate, information on the transfer wishes of each transfer candidate, and information on the transfer history of each transfer candidate, from the input / output control unit 13. Furthermore, the score calculation unit 15 acquires information on the initial values ​​of each personnel item of the transfer target post, the current initial values ​​of the importance of each scoring method 31 to 33 for the transfer target post, and the current initial values ​​of the importance of each personnel item of the transfer target post, from the input / output control unit 13. Then, the score calculation unit 15 calculates a score, which is information indicating the suitability of each transfer candidate for the transfer target post, using the information on each transfer candidate, and the initial values ​​of each personnel item and each importance value for the transfer target post.

[0051] For example, in the case of scoring method 31, the score calculation unit 15 expresses the average of the out-transferees and the values ​​of each transfer candidate for the personnel items of the transfer target post in a vector space. Then, the score calculation unit 15 expresses the distance between the average vector of the out-transferees and the vector of each transfer candidate as a score. Here, the score calculation unit 15 expresses the score as a number between 0 and 1.

[0052] Furthermore, in the case of scoring method 32, if there is a transfer target post among the first to third choice of transfer desires for each transfer candidate, the score calculation unit 15 assigns a score according to the order of the transfer desires. For example, the score calculation unit 15 assigns a score of 0 if there is no transfer target post among the transfer desires, a score of 0.3 if it is the third choice, a score of 0.6 if it is the second choice, and a score of 1 if it is the first choice.

[0053] In addition, in the case of scoring method 33, the score calculation unit 15 acquires the past affiliation history of each employee assigned to the transfer target post. Then, the score calculation unit 15 expresses the difference between the past affiliation history of each employee assigned to the transfer target post and the transfer history of each transfer candidate as a score. In this case, the score calculation unit 15 calculates the score so that the smaller the difference, the higher the score, and the larger the difference, the lower the score.

[0054] Thereafter, the score calculation unit 15 calculates the score of each transfer candidate as the sum of the scores calculated using each of the scoring methods 31 to 33. Then, the score calculation unit 15 outputs the score of each transfer candidate to the input / output control unit 13.

[0055] Thereafter, if the importance level is changed by user P, score calculation unit 15 receives input of the changed importance level from input / output control unit 13. In this case, score calculation unit 15 recalculates the score of each transfer candidate using information about each transfer candidate, as well as the initial values ​​of each personnel item for the transfer target post and the new importance level values. This recalculation is an example of "regenerating information indicating the suitability of multiple transfer candidates for a specific assignment using the changed first importance level when the first importance level has been changed." Thereafter, score calculation unit 15 outputs the recalculated score of each transfer candidate to input / output control unit 13.

[0056] 5 is a flowchart of a transfer candidate recommendation process performed by the transfer candidate recommendation device 1 according to the embodiment. Next, the flow of the transfer candidate recommendation process performed by the transfer candidate recommendation device 1 according to the embodiment will be described with reference to FIG.

[0057] The input / output control unit 13 receives the personnel data 20 from the terminal device 2 (step S101). Then, the input / output control unit 13 acquires information on the assigned post included in the personnel data 20 and outputs it to the initial value calculation unit 16.

[0058] The initial value calculation unit 16 acquires information on personnel items of the employee enrolled in the transfer target post from the information on the assigned post, calculates an average value, and sets it as an initial value. The initial value calculation unit 16 also calculates the scoring importance based on past data, the initial value of the previous scoring importance, and the scoring importance calculated excluding compromise placement, calculates an average, and sets it as the initial value of the importance of each scoring method 31 to 33 for the transfer target post. The initial value calculation unit 16 also calculates the personnel item importance based on past data, the initial value of the previous personnel item importance, and the personnel item importance calculated excluding compromise placement, calculates an average, and sets it as the initial value of the importance of each personnel item for the transfer target post (step S102). The initial value calculation unit 16 then outputs each calculated initial value to the input / output control unit 13.

[0059] Next, the input / output control unit 13 outputs information on each transfer candidate, including information on the personnel items of each transfer candidate acquired from the personnel data 20, information on the transfer wishes of each transfer candidate, and information on the transfer history of each transfer candidate, to the score calculation unit 15. The input / output control unit 13 also outputs each initial value calculated by the initial value calculation unit 16 to the score calculation unit 15. For each transfer candidate, the score calculation unit 15 calculates a score for the scoring method 31 using the importance of each personnel item of the transfer target post. For each transfer candidate, the score calculation unit 15 also calculates scores for each scoring method 32 and 33 using the transfer wishes and transfer history. Then, the score calculation unit 15 calculates a score for each transfer candidate using the scores and importance of each scoring method 31 to 33 (step S103). Thereafter, the score calculation unit 15 outputs the calculated score for each transfer candidate to the input / output control unit 13.

[0060] The input / output control unit 13 registers the initial values ​​of the personnel items of the transfer target position, the importance of each scoring method 31 to 33, the importance of the personnel items of the transfer target position, and the score of the transfer candidate, generates a transfer candidate recommendation screen 100, and displays it on the display unit 11 (step S104).

[0061] Thereafter, the input / output control unit 13 determines whether or not each importance level has been changed by input from the input unit 12 (step S105). If the importance level has been changed (step S105: Yes), the transfer candidate recommendation process returns to step S103.

[0062] On the other hand, if the importance has not been changed (step S105: No), the input / output control unit 13 acquires information on the selection result and compromise placement of the transfer candidate who has been decided to be assigned to the assigned post from the input unit 12 (step S106). Then, the input / output control unit 13 adds the selection result and the compromise placement information to the transfer candidate recommendation screen 100 and displays it on the display unit 11.

[0063] Thereafter, the input / output control unit 13 determines whether or not a save instruction has been received based on whether or not the save button 105 on the transfer candidate recommendation screen 100 has been selected using the input unit 12 (step S107). If a save instruction has not been received (step S107: No), the input / output control unit 13 returns to step S105.

[0064] On the other hand, if a save instruction is received (step S107: Yes), the input / output control unit 13 stores and saves the result of the adjustment of importance for this transfer, the selection result of the candidate who has been decided to be assigned to the assignment post, and the compromise placement information in the storage unit 14 (step S108). The transfer candidate recommendation device 1 executes the above process for each assignment post every time a personnel transfer operation occurs and a transfer plan creation task is performed.

[0065] As described above, the transfer candidate recommendation device according to this embodiment performs unsupervised learning using the past history of each assignment position, the results of previous adjustments, and information indicating a compromise assignment. As a result, the transfer candidate recommendation device according to this embodiment calculates the importance of each scoring method used in the appropriateness evaluation and the initial importance of each personnel item used in each scoring method, and provides these to the user. Furthermore, the transfer candidate recommendation device calculates a score for each transfer candidate according to the adjusted importance and provides the user with a ranking according to the score. This allows users performing transfer proposal creation tasks to easily understand the initial importance values ​​based on different judgment criteria due to differences in departments, job types, etc. Furthermore, by excluding information on transfer candidates who were assigned as a compromise due to the lack of suitable candidates, different judgment criteria due to differences in departments, job types, etc. can be more appropriately reflected in the initial importance values. In other words, users performing transfer proposal creation tasks can easily determine transfer candidates to assign based on the scores of each transfer candidate according to the evaluation required for the assignment position, thereby improving the efficiency of transfer proposal creation tasks.

[0066] (Hardware configuration) 6 is a hardware configuration diagram of a transfer candidate recommending device 1. Next, an example of the hardware configuration of the transfer candidate recommending device 1 according to the embodiment will be described with reference to FIG.

[0067] 6, the transfer candidate recommendation device 1 includes a CPU (Central Processing Unit) 91, a memory 92, a hard disk 93, a network interface 94, a display device 95, and an input device 96. The CPU 91 is connected to the memory 92, the hard disk 93, the network interface 94, the display device 95, and the input device 96 via a bus.

[0068] The display device 95 is a monitor or the like, and realizes the function of the display unit 11 illustrated in Fig. 1. The input device 96 is a keyboard, a mouse, or the like, and realizes the function of the input unit 12 illustrated in Fig. 1.

[0069] The network interface 94 is an interface for communication between the transfer candidate recommendation device 1 and an external device. For example, the network interface 94 relays communication between the CPU 91 and the terminal device 2.

[0070] The hard disk 93 is an auxiliary storage device. The hard disk 93 realizes, for example, the function of the storage unit 14 illustrated in Fig. 1. The hard disk 93 also stores various programs including programs that realize the functions of the input / output control unit 13, the score calculation unit 15, and the initial value calculation unit 16 illustrated in Fig. 1.

[0071] The memory 92 is a main storage device and may be, for example, a dynamic random access memory (DRAM).

[0072] The CPU 91 reads various programs from the hard disk 93, expands them into the memory 92, and executes them. As a result, the CPU 91 realizes the functions of the input / output control unit 13, the score calculation unit 15, and the initial value calculation unit 16 illustrated in FIG. [Explanation of symbols]

[0073] 1. Transfer candidate recommendation device 2. Terminal Device 11 Display section 12 Input section 13 Input / Output Control Unit 14 Storage section 15 Score calculation section 16 Initial value calculation section

Claims

1. Accepting designation of a specific person among multiple people assigned to a specific assignment location; acquiring multiple types of attribute information of valid personnel excluding the information of the specified specific personnel from among the plurality of personnel; Using training data including the multiple types of attribute information of the available personnel, a first importance level for each of the multiple types of attribute information is estimated by machine learning, and based on the multiple types of attribute information of multiple transfer candidates who are candidates for assignment to the specific assignment destination, information indicating the suitability of the multiple transfer candidates for the specific assignment destination is generated and output using the estimated first importance level. An information processing program that causes a computer to execute a process.

2. The information processing program according to claim 1, further comprising: calculating a difference between an average of each of the multiple types of attribute information of the effective personnel and an average of each of the multiple types of attribute information of the employees transferring out from the specific assignment; and estimating a first importance for each of the multiple types of attribute information based on a standard deviation of the difference.

3. The information processing program described in claim 1 or 2, characterized in that each of the first importance levels is changeable, and when the first importance level is changed, information indicating the suitability of the multiple transfer candidates for the specific assignment destination is regenerated and provided using the changed first importance level.

4. The information processing program according to claim 3, wherein the first importance is estimated based on a result of a previous change to the first importance in addition to the result of the machine learning using the training data.

5. An information processing program as described in any one of claims 1 to 4, characterized in that the first importance is estimated based on the results of machine learning using the training data as well as the results of machine learning using the multiple types of attribute information of past transferees and transferees to the specific assignment destination.

6. An information processing program as described in any one of claims 1 to 5, characterized in that when there are multiple methods for generating information indicating suitability for the specific assignment destination, and information indicating suitability for one of the specific assignment destinations is generated using second importance assigned to each of the multiple generation methods through machine learning using training data including related information including the multiple types of attribute information of the available personnel, the second importance is estimated so that the suitability of the available personnel is high among subjects including the available personnel, and the estimated second importance is used to generate and output information indicating the suitability of the multiple transfer candidates for the specific assignment destination.

7. Accepting designation of a specific person among multiple people assigned to a specific assignment location; acquiring multiple types of attribute information of valid personnel excluding the information of the specified specific personnel from among the plurality of personnel; Using training data including the multiple types of attribute information of the available personnel, a first importance level for each of the multiple types of attribute information is estimated by machine learning, and based on the multiple types of attribute information of multiple transfer candidates who are candidates for assignment to the specific assignment destination, information indicating the suitability of the multiple transfer candidates for the specific assignment destination is generated and output using the estimated first importance level.

1. An information processing method comprising:

8. an input / output control unit that receives a designation of a specific person among a plurality of people assigned to a specific assignment location; an importance calculation unit that acquires multiple types of attribute information of available personnel excluding information on the specified specific personnel from among the multiple personnel, and estimates a first importance for each of the multiple types of attribute information by machine learning using training data including the multiple types of attribute information of the available personnel; a score calculation unit that generates and outputs information indicating suitability of the plurality of transfer candidates for the specific assignment destination using the first importance estimated by the importance calculation unit based on the plurality of types of attribute information of the plurality of transfer candidates who are candidates for assignment to the specific assignment destination; An information processing device comprising:

Images