System for outputting information on price related to human resource

Abstract

To provide a system which outputs information for determining whether or not the price is appropriate according to attributes of human resources.SOLUTION: A system for outputting information on a price related to human resource, is configured to: acquire, as an input, a human resource dataset including information related to a plurality of human resources, the human resource dataset being formed by associating information identifying a cluster to which each of the human resources belongs, the price of the human resource, and a first attribute value other than the price of the human resource; acquire, as an input, a price of a new human resource and a first attribute value; determine a cluster of the new human resource, based on the first attribute value of the new human resource; and output a result of determination regarding the price of the new human resource on the basis of the price of the new human resource and a distribution of prices of human resources belonging to the cluster determined for the new human resource.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a system for outputting information regarding prices relating to human resources. [Background technology]

[0002] In organizations such as companies, analysis is performed by clustering a large number of human resources. For example, Patent Document 1 describes a system that determines whether to hire a job seeker based on the clusters of human resources. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2004-21908 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, the conventional technology has a problem in that it is difficult to appropriately evaluate the price of a new employee to be recruited. In particular, it is difficult to evaluate a fair price according to the attributes of the employee.

[0005] The present invention has been made to solve such problems, and has an object to provide a system that outputs information for determining whether a price is appropriate depending on the attributes of the personnel. [Means for solving the problem]

[0006] In one example of the system for outputting information on human resource prices according to the present invention, The system receives as input a talent dataset including information relating to a plurality of talents, the talent dataset associating, for each talent, information identifying a cluster to which the talent belongs, a price of the talent, and a first attribute value other than the price of the talent; The system comprises: - The price and the first attribute value of the new talent are taken as input; - determining a cluster for the new talent based on a first attribute value of the new talent; - Outputting a judgment result regarding the price of the new talent based on the distribution of the prices of talent belonging to the determined cluster of the new talent and the price of the new talent.

[0007] In one example, the distribution situation is a distribution of the prices of talent belonging to each cluster, -upper limit, -lower limit, - at least one of the quartiles, - average value, It contains information representing at least one of the following:

[0008] In one example, the determination result includes a message indicating whether the price of the new talent falls within a predetermined range from the average or median price of talent in the determined cluster of the new talent. Effect of the Invention

[0009] The system according to the present invention can output information for determining whether a price is appropriate depending on the attributes of the human resources. [Brief description of the drawings]

[0010] [Figure 1] 1 is a diagram showing the configuration and processing contents of a human resources management system 100 according to a first embodiment of the present invention. [Diagram 2] 2 is an overview of the processing of the clustering device 10 of FIG. 1. [Diagram 3] An example of human resource evaluation information 53 in Figure 1. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings. [Embodiment 1] 1 is a diagram showing the configuration and processing contents of a human resources management system 100 according to a first embodiment of the present invention. The human resources management system 100 is a system that outputs information related to prices related to human resources. The human resources management system 100 includes a clustering device 10, a price validity evaluation device 20, a new human resources registration device 30, and a machine learning model construction device 40.

[0012] The clustering device 10, the price validity evaluation device 20, the new talent registration device 30, and the machine learning model construction device 40 each have a hardware configuration as a known computer, and include, for example, a calculation means and a storage means. The calculation means includes, for example, a processor, and the storage means includes, for example, a storage medium such as a semiconductor memory device and a magnetic disk device. A part or all of the storage medium may be a non-transitory storage medium.

[0013] Furthermore, the clustering device 10, the price validity evaluation device 20, the new talent registration device 30, and the machine learning model construction device 40 each include an input means and an output means. The input means includes input devices such as a keyboard and a mouse. The output means includes output devices such as a display and a printer. The input means and the output means may also include a communication device such as a network interface.

[0014] The storage means of the clustering device 10, the price validity evaluation device 20, the new talent registration device 30, and the machine learning model construction device 40 may store a program. The processor of each device may execute the program, causing each device to perform the functions described in this embodiment.

[0015] In the clustering device 10, the calculation means constitutes an information processing unit 11, and the output means constitutes an output unit 13. Also, the storage means constitutes a human resources information storage unit 14 and a human resources data set storage unit 15.

[0016] In the price validity evaluation device 20, the calculation means constitutes an information processing unit 21, the input means constitutes an input unit 22, and the output means constitutes an output unit 23. In addition, the storage means constitutes a human resources data set storage unit 25, a parameter storage unit 26, and a human resources evaluation information storage unit 27.

[0017] In the new talent registration device 30, the output means constitutes an output unit 33. Also, the storage means constitutes a new talent information storage unit .

[0018] In the machine learning model construction device 40 , the calculation means constitutes an information processing unit 41 , the input means constitutes an input unit 42 , and the storage means constitutes a human resources dataset storage unit 44 and a machine learning model storage unit 45 .

[0019] FIG. 2 shows an overview of the processing of the clustering device 10. The operation of the clustering device 10 will be described below with reference to FIG. 1 and FIG. 2. The clustering device 10 is configured to be able to communicate with an external environment. The external environment is, for example, another computer or a storage medium. The clustering device 10 copies and acquires human resource information 51 from the external environment, and stores it in the human resource information storage unit 14.

[0020] The human resource information 51 includes, for example, information on multiple human resources currently employed or previously employed in an organization using the human resource management system 100. For each of the multiple human resources, the human resource information 51 associates the price of the human resource with attribute values ​​other than the price of the human resource (including a first attribute value described below). In this embodiment, the number of attribute values ​​is one or more, and the contents and formats can be appropriately defined by those skilled in the art, but in this embodiment, they include years of experience, languages ​​used, and field of work. The price may be expressed in any format, for example, hourly rate.

[0021] The information processing unit 11 of the clustering device 10 acquires the talent information from the talent information storage unit 14, clusters the talent based on the attribute values, and generates a plurality of talent clusters. The specific clustering process can be appropriately designed by a person skilled in the art based on known techniques, and for example, the k-means method can be used. Although a total of five clusters, cluster 0 to cluster 4, are illustrated in FIG. 2, the number of clusters to be generated is not limited to this, and may be designed arbitrarily, or may be automatically determined by the clustering device 10.

[0022] As a result of the clustering process, a talent dataset is generated. The talent dataset associates, for each talent, information identifying the cluster to which the talent belongs in addition to what is included in the talent information 51. That is, the talent dataset storage unit 15 includes cluster information and talent information, and the talent dataset associates, for each talent, information identifying the cluster to which the talent belongs, the price of the talent, and the attribute value of the talent. The information processing unit 11 stores the generated talent dataset in the talent dataset storage unit 15.

[0023] Next, we will explain the operation of the new talent registration device 30. The new talent registration device 30 is configured to be able to communicate with the external environment, similar to the clustering device 10. The new talent registration device 30 copies and acquires new talent information 52 from the external environment, and stores it in the new talent information storage unit 34.

[0024] New human resource information 52 includes information on one or more new human resources. A new human resource is, for example, a human resource for which the organization using the human resource management system 100 should consider whether to newly employ the human resource, or a human resource for which the organization should consider whether to continue employing the human resource. New human resource information 52 can be in the same format as human resource information 51, and for each of a plurality of human resources, the price of the human resource and an attribute value other than the price of the human resource (including a first attribute value described later) are associated with the human resource.

[0025] Next, a description will be given of the operation of the machine learning model construction device 40. Like the clustering device 10, the machine learning model construction device 40 is configured to be able to communicate with the external environment.

[0026] The input unit 42 acquires a talent dataset as an input from the output unit 13 of the clustering device 10, and stores it in the talent dataset storage unit 44. The machine learning model storage unit 45 stores the machine learning model. The information processing unit 41 acquires a talent dataset (including talent information) from the talent dataset storage unit 44. Furthermore, the information processing unit 41 performs machine learning processing on the machine learning model based on this information, and generates a trained model.

[0027] Machine learning is performed by inputting the attribute values ​​of each human resource and inferring the cluster of the human resource as the output. The above-mentioned human resource data set is data that associates the attribute values ​​of each human resource with the cluster to which the human resource belongs, and therefore can be used as training data. The specific model format of the machine learning model can be appropriately designed by those skilled in the art, but can be, for example, SVM (support vector machine).

[0028] In learning, it is not necessary to use all of the attribute values ​​of the personnel; it is sufficient to use at least one type of attribute value. The attribute values ​​(one or more types of attribute values) used in learning are called "first attribute values." The first attribute values ​​do not include price. Also, in this embodiment, the first attribute values ​​do not include the years of experience of each personnel, but as a variant, the first attribute values ​​may include the years of experience.

[0029] Next, the operation of the price validity evaluation device 20 will be described. The input unit 22 acquires the talent data set as an input from the output unit 13 of the clustering device 10, and acquires new talent information as an input from the output unit 33 of the new talent registration device 30, and stores these in the talent data set storage unit 25. Note that at this point, the talent data set includes the talent price, the first attribute value, and cluster information, but the new talent information includes only the new talent price and the first attribute value, and does not include cluster information.

[0030] The information processing unit 21 determines a cluster for each new talent based on the first attribute value of the new talent, using the trained model generated by the machine learning model construction device 40. When determining the cluster, as shown in FIG. 1, a talent dataset stored in the talent dataset storage unit 25 may also be used (this may be omitted).

[0031] After the cluster of each new talent is determined, the information processing unit 21 performs a judgment on the price of each new talent based on the distribution of the prices of talents belonging to the determined cluster of the new talent and the price of the new talent, in light of the judgment criteria stored in advance in the parameter storage unit 26. The information processing unit 21 generates talent evaluation information 53 including the judgment result, and stores it in the talent evaluation information storage unit 27.

[0032] The output unit 23 acquires this personnel evaluation information 53 from the personnel evaluation information storage unit 27 and outputs it. This allows the user of the personnel management system 100 to view the personnel evaluation information 53.

[0033] FIG. 3 shows an example of the talent evaluation information 53. In this example, the judgment results for 10 new talents are shown. The judgment criteria in this example are for determining which of the following ranges A to E the price of the new talent falls into. The talent evaluation information 53 includes information indicating which range each new talent falls into. - Range A: The price of the new talent is below the lower limit (minimum value) of the price distribution of the cluster to which the new talent belongs. - Range B: The price of the new talent is in the bottom 25% range (above the minimum and below the first quartile) of the price distribution of the cluster to which the new talent belongs. - Range C: The price of the new talent is in the middle range (above the first quartile and below the third quartile) of the price distribution of the cluster to which the new talent belongs. - Range D: The price of the new talent is in the top 25% range (above the third quartile and below the maximum value) of the price distribution of the cluster to which the new talent belongs. - Range E: The price of the new talent exceeds the upper limit (maximum value) of the price distribution of the cluster to which the new talent belongs.

[0034] In the example of FIG. 3, the price of new talent No. 1 ("application price" or "application unit price" in FIG. 3) is 8,048 yen, and this new talent No. 1 is classified into cluster No. 0. In this example, the price of 8,048 yen falls into range D (within the top 25%) in the price distribution of cluster No. 0. The talent evaluation information 53 includes a message for new talent No. 1, which corresponds to range D, saying "The application unit price is in the top 25% range of the cluster predicted by the AI ​​(over 7,758 yen), so confirm the validity." In particular, in this embodiment, this message includes the price of "7,758 yen" which corresponds to the third quartile.

[0035] Similarly, for other new talent, messages corresponding to each range are included (note that the example in FIG. 3 does not show new talent that falls into ranges A and E). Here, by including in the message the price corresponding to the minimum value for range A, the price corresponding to the first quartile for range B, the price corresponding to the third quartile for range D, and the price corresponding to the maximum value for range E, the user of the talent management system 100 can easily know how far the price of that new talent is from the median value of the cluster.

[0036] In addition, the judgment is not based only on the price of the new talent, but on a comparison with the distribution situation of the new talent in a cluster of similar attributes. Therefore, the employer can judge whether the price of the new talent is appropriate based on a relative base price according to the attributes, not an absolute base price.

[0037] In this embodiment, if the price falls within range C, no message is displayed to alert the user, and in the example of FIG. 3, a message stating "There is no particular problem" is displayed. On the other hand, if the price falls within any of ranges A, B, D, or E, a message is displayed to alert the user. In the example of FIG. 3, a message stating "Confirm validity" is displayed. This allows the user of the human resource management system 100 to reliably verify validity without overlooking new human resources with prices far from the median.

[0038] The message may be highlighted. In the example of FIG. 3, it is highlighted by underlining, but other forms of highlighting (e.g., background coloring) are also possible.

[0039] The talent evaluation information 53 may include information other than the judgment result. In the example of Fig. 3, the talent evaluation information 53 includes the name of the cluster. The name of the cluster can be appropriately determined and input by the user of the talent management system 100. In this embodiment, the talent evaluation information 53 includes the price, years of experience, and other attribute values ​​(work field group, etc.) for each new talent.

[0040] In the above-mentioned embodiment 1, the specific process of the judgment can be changed as appropriate. For example, the information representing the distribution state used for the judgment is not limited to that used in embodiment 1. In embodiment 1, the information representing the distribution state includes the upper limit, lower limit, first quartile, and third quartile in the distribution of the prices of the human resources belonging to each cluster, but it is not necessary to use all of these. In a modified example, the distribution state is, in the distribution of the prices of the human resources belonging to each cluster, -upper limit, -lower limit, - at least one of the quartiles, - average value, It is sufficient to include information that represents at least one of the above. It is also possible to use information other than these.

[0041] The format of the judgment result can be changed arbitrarily. The judgment result of the talent evaluation information 53 may include a message indicating whether the price of the new talent is close to the average price in the cluster, and / or may include a message indicating whether the price of the new talent is close to the median price in the cluster. The judgment result of the talent evaluation information 53 may also include a message indicating whether the price of the new talent is within a predetermined range from the average price in the cluster, and / or may include a message indicating whether the price of the new talent is within a predetermined range from the median price in the cluster. Such a designation can be realized, for example, via the judgment criteria in the parameter storage unit 26.

[0042] Those skilled in the art can arbitrarily add, change or delete components within the scope of the present invention in the above-mentioned embodiment 1. For example, in the human resource management system 100, any of the clustering device 10, the price validity evaluation device 20, the new human resource registration device 30, and the machine learning model construction device 40 may be divided into multiple computers, and two or more of the clustering device 10, the price validity evaluation device 20, the new human resource registration device 30, and the machine learning model construction device 40 may be configured by a single computer.

[0043] The clustering device 10 may be omitted. A known clustering device may be used instead of the clustering device 10. Even if a clustering device is not used, it is sufficient that a human resources data set in an appropriate format is prepared. When a clustering device is not used, the input unit 22 of the price validity evaluation device 20 may directly acquire a data set in an appropriate format.

[0044] The new talent registration device 30 may be omitted. In that case, the input unit 22 of the price validity evaluation device 20 may directly acquire the new talent information 52 from the external environment.

[0045] The machine learning model construction device 40 may be omitted. A known machine learning model construction device may be used instead of the machine learning model construction device 40. Even if a machine learning model construction device is not used, an existing trained model may be used in some cases. [Explanation of symbols]

[0046] 10…Clustering device 11...Information processing section 13...Output section 14…Human Resources Information Storage Section 15…Human Resources Data Set Storage Section 20...Price validity evaluation device 21...Information processing section 22...Input section 23...Output section 25…Human Resources Data Set Storage Section 26...Parameter storage section 27…Human Resources Evaluation Information Storage Unit 30…New Personnel Registration Device 33...Output section 34…New personnel information storage section 40...Machine learning model building device 41...Information processing section 44…Human Resources Data Set Storage Section 45…Machine learning model memory section 51…Human Resources 52…New personnel information 53. Personnel evaluation information 100…Human Resources Management System

Claims

1. A system for outputting information regarding prices related to human resources, The system receives as input a talent dataset including information relating to a plurality of talents, the talent dataset associating, for each talent, information identifying a cluster to which the talent belongs, a price of the talent, and a first attribute value other than the price of the talent; The system comprises: - taking the price and the first attribute value of the new talent as input; - determining a cluster for the new talent based on a first attribute value of the new talent; - outputting a judgment result regarding the price of the new talent based on the distribution of the prices of talent belonging to the determined cluster of the new talent and the price of the new talent; system.

2. The distribution situation is the distribution of the prices of talent belonging to each cluster, -upper limit, -lower limit, - at least one of the quartiles, - average value, The system of claim 1 , further comprising information representative of at least one of:

3. The system of claim 1 , wherein the determination result includes a message indicating whether the price of the new talent falls within a predetermined range from the average or median price of talent in the determined cluster of the new talent.

Images