Recruitment support device, recruitment support method, and recruitment support program

Abstract

In order to provide suitable personnel support, a recruitment support device (1) is provided with: a reception unit (11) that receives a request regarding the place of assignment of target personnel; an estimation unit (12) that estimates candidates for the place of assignment of the target personnel that match the request using a trained model that has learned the relationship between at least one of the skill and work history of each of a plurality of persons and the place of assignment of each of the plurality of persons; and an output unit (13) that outputs information indicating the candidates for the place of assignment of the target personnel estimated by the estimation unit. (FIG. 1)

Description

[Technical field]

[0001] The present invention relates to a recruitment support device that supports personnel affairs. [Background technology]

[0002] Conventionally, technology for supporting personnel affairs using information processing devices has been researched. For example, the following Patent Document 1 discloses an information processing device that outputs photographic data of employees and generates a reference model for evaluating people by using subjects selected from the output photographic data as models.

[0003] According to the technology of Patent Document 1, for example, it is possible to perform high performer analysis to analyze workers with good performance, low performer analysis to analyze workers with poor performance, absentee analysis to analyze workers who are on leave, retiree analysis, etc. If this is used to evaluate employment candidates, it becomes possible to hire people who are likely to become high-performing workers. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2021-39777 Summary of the Invention [Problem to be solved by the invention]

[0005] However, even if a person has excellent abilities, there are many cases where they do not get along with the department, work content, superiors, colleagues, etc. of the company where they were hired, and they take a leave of absence or resign without being able to demonstrate their abilities. In addition, the technology of Patent Document 1 can also perform analysis of employees who take leave of absence and those who resign, but taking a leave of absence and resigning are also greatly influenced by compatibility with the company where they are hired, as described above, in addition to personal qualities. For this reason, those who are evaluated as unlikely to take a leave of absence or resign in the analysis of employees who take leave of absence and those who resign in Patent Document 1 are not necessarily able to continue working. This is not limited to human resource recruitment, but also applies to transfers and team formation within an organization.

[0006] One aspect of the present invention has been made in consideration of the above-mentioned problems, and one example of its objective is to provide technology that can provide personnel support in an optimal manner by taking into account various information regarding the recipient of the personnel. [Means for solving the problem]

[0007] A recruitment support device according to one aspect of the present invention comprises a receiving means for receiving a request regarding the assignment of a target personnel, an estimation means for estimating potential assignment destinations for the target personnel that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of individuals and the affiliations of each of the plurality of individuals, and an output means for outputting information indicating the potential assignment destinations for the target personnel estimated by the estimation means.

[0008] A recruitment support method according to one aspect of the present invention includes at least one processor receiving a request regarding a department for assigning a target personnel, estimating potential departments for the target personnel that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of individuals and the departments to which each of the plurality of individuals belongs, and outputting information indicating the estimated potential departments for the target personnel.

[0009] A recruitment support program according to one aspect of the present invention causes a computer to function as: a receiving means for receiving a request regarding the assignment of a target personnel; an estimation means for estimating potential assignment destinations for the target personnel that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of individuals and the affiliations of each of the plurality of individuals; and an output means for outputting information indicating the potential assignment destinations for the target personnel estimated by the estimation means. Effect of the Invention

[0010] According to one aspect of the present invention, personnel support can be carried out appropriately by taking into consideration various information related to the recipient of the human resources. [Brief description of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a configuration of a recruitment support device according to a first exemplary embodiment of the present invention. [Diagram 2] 1 is a flowchart showing a flow of a recruitment support method according to a first exemplary embodiment of the present invention. [Diagram 3] FIG. 1 is a diagram illustrating feature learning in graph-based relationship learning. [Figure 4] FIG. 11 is a block diagram showing a configuration of a recruitment support device according to a second exemplary embodiment of the present invention. [Diagram 5] FIG. 11 is a schematic diagram showing an example of a method for generating a target human resources graph by a graph generator according to a second exemplary embodiment of the present invention. [Figure 6] FIG. 11 is a schematic diagram illustrating an example of a prediction method performed by a link predictor according to a second exemplary embodiment of the present invention. [Figure 7] 13 is a schematic diagram showing an example of a prediction method performed by a link predicting unit, a specifying unit, and a compatibility determining unit according to a second exemplary embodiment of the present invention. FIG. [Figure 8] FIG. 11 is a flow diagram showing a first example of processing executed by a recruitment support device according to a second exemplary embodiment of the present invention. [Figure 9] FIG. 11 is a schematic diagram showing another example of a prediction method performed by a link predicting unit, a specifying unit, and a compatibility determining unit according to the second exemplary embodiment of the present invention. [Figure 10] FIG. 11 is a diagram for explaining an example of a recommendation method by a recommendation unit according to a second exemplary embodiment of the present invention. [Figure 11] FIG. 11 is a flow diagram showing a second example of processing executed by the recruitment support device according to the second exemplary embodiment of the present invention. [Figure 12] FIG. 11 is a diagram showing an example of presentation by an output unit according to a second exemplary embodiment of the present invention. [Figure 13] FIG. 13 is a schematic diagram illustrating an example of a prediction method performed by a link predictor according to a third exemplary embodiment of the present invention. [Figure 14]FIG. 13 is a schematic diagram illustrating another example of a prediction method performed by a link predictor according to the third exemplary embodiment of the present invention. [Figure 15] FIG. 13 is a diagram showing an overview of a recruitment support method according to a fourth exemplary embodiment of the present invention. [Figure 16] FIG. 13 is a block diagram showing a configuration of a recruitment support device according to a fourth exemplary embodiment of the present invention. [Figure 17] FIG. 13 is a flowchart showing a flow of a process executed by a recruitment support device according to a fourth exemplary embodiment of the present invention. [Figure 18] 13 is a diagram illustrating an example of predicting the characteristics of a target human resource based on feature amounts calculated from a target human resource graph and a receiving organization graph. FIG. [Figure 19] FIG. 11 is a configuration diagram for realizing a recruitment support device using software. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0012] Example 1 A first exemplary embodiment of the present invention will be described in detail with reference to the drawings. This exemplary embodiment is a basic form of the exemplary embodiments described later.

[0013] (Recruitment support device) The configuration of the recruiting support device 1 according to this exemplary embodiment will be described with reference to the drawings. Fig. 1 is a block diagram showing the configuration of the recruiting support device 1 according to this exemplary embodiment.

[0014] As shown in FIG. 1, the recruitment support device 1 includes a reception unit (reception means) 11, an estimation unit (estimation means) 12, and an output unit (output means) 13.

[0015] The reception unit 11 receives a request regarding the assignment of the target human resource. Note that information regarding the target human resource, such as the target human resource's skills and work history, is information that should be considered when deciding the assignment of the target human resource, so a request including such information is also included in the category of a request "regarding the assignment of the target human resource." The estimation unit 12 estimates candidate assignments for the target human resource that match the request received by the reception unit 11, using a trained model that has learned the relationship between at least one of the skills and work history of each of multiple people and the affiliations of each of the multiple people. The output unit 13 outputs information indicating the candidate assignments for the target human resource estimated by the estimation unit 12.

[0016] Here, the work history of a person may include, but is not limited to, the departments to which the person belongs, the departments to which the person has belonged, the number of years that the person has belonged to each department, and the order in which the departments have belonged (collectively referred to as affiliation history). In addition, departments include, but are not limited to, divisions, sections, and groups, by way of example.

[0017] According to the recruitment support device 1 having the above configuration, it is possible to present to the user candidates for assignment of the target personnel that are considered to match the request, based on at least one of the skills and work history of each of the multiple people and the relationship with the affiliation of each of the multiple people. Therefore, according to the above configuration, it is possible to obtain an effect that it is possible to appropriately perform personnel support for the target personnel by taking into account various information related to the accepting organization.

[0018] (Recruitment Support Program) The above-mentioned functions of the recruiting support device 1 can also be realized by a program. The recruiting support program according to this exemplary embodiment causes a computer to execute the following processes: accepting a request for a workplace for the target human resource; estimating candidate workplaces for the target human resource that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of people and the affiliation of each of the plurality of people; and outputting information indicating the estimated candidate workplaces for the target human resource. This recruiting support program provides the effect of enabling suitable personnel support for the target human resource by taking into account various information related to the receiving company.

[0019] (Recruitment support method) The recruiting support method according to this exemplary embodiment will be described with reference to Fig. 2. Fig. 2 is a flow chart showing the flow of the recruiting support method according to the first exemplary embodiment of the present invention.

[0020] In S11, the computer receives a request regarding the assignment of the target human resource. The request may be received via any input device. For example, the request may be received via a mouse or keyboard, a touch panel, or a voice input device.

[0021] In S12, the computer uses a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of individuals and the affiliations of each of the plurality of individuals to estimate potential assignment destinations for the target personnel that match the request.

[0022] In S13, the computer outputs information indicating the candidate assignment destinations for the estimated target personnel. The output device is arbitrary, and for example, the information may be output to a display device to display the information, or to an audio output device to output the information as audio.

[0023] As described above, the recruiting support method according to the present exemplary embodiment includes a computer having at least one processor receive a request for a workplace for a target talent (S11), estimate potential workplaces for the target talent that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of people and the affiliation of each of the plurality of people (S12), and output information indicating the potential workplaces for the target talent estimated in S12 (S13). This recruiting support method has the effect of making it possible to provide suitable personnel support for the target talent by taking into account various information related to the receiving company.

[0024] In addition, the execution subject of each step in the above-mentioned recruiting support method may be one computer (for example, the recruiting support device 1), or each step may be executed by a different computer. This also applies to the flow described in the exemplary embodiment 2 and subsequent embodiments.

[0025] [Graphs and Learning] In the following, in the exemplary embodiment 1 and the exemplary embodiments described later (hereinafter referred to as each exemplary embodiment), a graph, which is an example of information that can be used for recruitment support, will be described. Learning of the graph and prediction using the graph will also be described.

[0026] (graph) A graph here refers to data having a structure consisting of multiple nodes and links connecting the nodes. The type of link that indicates the relationship between nodes is also called a "relation." Links are also sometimes called edges. Graphs can be broadly divided into directed graphs, in which each link has directionality, and undirected graphs, in which each link does not have directionality. It is possible to use either directed graphs or undirected graphs, or to use a combination of the two.

[0027] In the exemplary embodiments, when a graph is used, the nodes may represent tangible or intangible elements related to a person. For example, Personal identification information such as name and personal ID ·age ·Personality Current job type or desired job type Current affiliation or desired affiliation ·skill A graph containing nodes representing various elements such as Personal identification information such as name and personal ID Basic attributes (job type, gender, affiliation, position, rank, etc.) Achievements and history (work history (including affiliation history), academic history, evaluation history, training attendance history, history related to superiors and subordinates, achievements, receipt history, award history, attendance history) Abilities, skills and qualifications (skill level, language ability, qualifications held, etc.) Mindset (personality, aptitude test results, personality diagnosis results, career aspirations, satisfaction survey results, interview history, supervisor notes, etc.) Job description (mission, goals, duties, etc.) - Job characteristics (whether the person has contributed or can contribute to new business, whether they are an immediate asset or a late bloomer, whether they are suited to working overseas or domestically, etc.) Behavioral data (location information, main communication partners, etc.) A graph including nodes representing various elements such as the above can be used. Here, information about main communication partners can be identified from the usage status of tools such as an in-house SNS (Social Networking Service). The above-mentioned basic attributes, abilities, skills, qualifications, and mindsets may also be expressed as characteristics. The characteristics may include other elements as described above. Each of the above-mentioned elements may also be referred to as an attribute.

[0028] The graph may also include multiple nodes corresponding to one element. For example, the nodes representing skills may be represented by individual nodes, such as a node representing a first skill, a node representing a second skill, and a node representing a third skill. The same applies to other elements.

[0029] When the above nodes exist as elements, the relationship represented by the link is as follows: The relationship between an element and a person Relationship between certain factors and age The relationship between certain elements and personality The relationship between a certain factor and the desired job type Relationship between an element and a skill For example, a link connecting a node indicating a personality and a node indicating an achievement may represent a relationship in which the personality is a factor in the achievement.

[0030] (Learning and Prediction) For the graphs described above, a machine learning technique can be applied to perform graph-based relationship learning. This learning makes it possible to perform classification and prediction processes using the graphs. In each exemplary embodiment, this learning may be performed as part of recruitment support, or a trained graph that has already undergone this learning may be used.

[0031] In graph-based relationship learning, first, the feature of each node is calculated. The feature may be calculated in the form of a feature vector, for example. By expressing the feature of each node as a feature vector, learning can be performed on a graph in which nodes of various formats are mixed. For example, graph-based relationship learning can be performed on a graph including images and numerical values ​​indicating various elements as described above. As an example, a photo of a human resource can be used as a node.

[0032] Next, the feature amount of each node is updated based on the links connected to each node and the nodes to which the links are connected. This process is similar to the convolution process in a convolutional neural network. This will be explained based on Figure 3. Figure 3 is a diagram for explaining the learning of feature amounts in graph-based relationship learning.

[0033] The graph shown in Fig. 3 includes four nodes A to D. Node A is connected to nodes B and C, and node C is connected to node D. After calculating the initial features of these four nodes, multiple convolutions are performed as described below to update the features of each node.

[0034] In the first convolution, the initial feature of node A is multiplied by a specific weight and then added to the features of nodes B and C connected to node A. For node C, the initial feature of node C is multiplied by a specific weight and then added to the features of node D. If it is a directed graph, the weights are adjusted according to the direction of the links.

[0035] In the second convolution, as in the first convolution, the feature of each node is multiplied by a predetermined weight and then added with the feature of the node linked to that node. Here, the feature of node C reflects the feature of node D due to the first convolution. Therefore, in the second convolution, not only the feature of node C but also the feature of node D is reflected in node A.

[0036] By repeating the above process a number of times according to the node hierarchy, the features of each node directly or indirectly connected by a link are mutually reflected. In graph-based relationship learning, the weight values ​​used for the above weighting are optimized based on the known relationships between nodes. By using a learned graph, it is also possible to predict the relationships between nodes and the nodes to which links will be linked, as described below.

[0037] (Node relationship prediction) By carrying out the above-mentioned learning, it becomes possible to predict relationships between nodes that are not explicitly stated in the original graph. When performing node relationship prediction, the user may specify two nodes and request that the relationship between those nodes be returned. For example, when a request is input from the user asking about the relationship between the node of "person A" and the node of "person B", it is possible to predict by node relationship prediction that the relationship, i.e., the link connecting these nodes, is a "trust relationship". In addition, in node relationship prediction, the probability (certainty) of the prediction result can also be calculated. The same applies to node prediction described below.

[0038] (node ​​prediction) Furthermore, by performing the above-mentioned learning, it is also possible to predict the node connected to a certain node via a specified link. When performing node prediction, the user may specify one node and a link starting from that node, and request that the linked node be returned. For example, assume that the user inputs a request asking about the node connected to the node of "person A" via the "personality" link. In this case, by using node prediction, it is possible to predict whether the node connected to the node of "person A" via the "personality" link is "responsible" or "curious", for example.

[0039] Exemplary embodiment 2 The configuration of a recruitment support device 2 according to a second exemplary embodiment of the present invention will be described with reference to Fig. 4. Fig. 4 is a block diagram showing the configuration of the recruitment support device 2 according to this exemplary embodiment.

[0040] As shown in the figure, the recruiting support device 2 includes a reception unit 201, a graph generation unit 202, a link prediction unit 203, an identification unit 204, a compatibility determination unit 205, a recommendation unit 206, a learning unit 207, an estimation unit 208, a reason generation unit 209, and an output unit 210. In addition to these components, the recruiting support device 2 may include an input device that receives a user's input operation, an output device for data output by the recruiting support device 2, a communication device for the recruiting support device 2 to communicate with other devices, and the like. The output mode of the output device is arbitrary, and may be, for example, a display output or a voice output.

[0041] The reception unit 201 receives a request regarding the assignment of the target talent. Here, the request includes information about the target talent to whom the user wants to determine the assignment. As an example, the request includes, but is not limited to, the target talent's name (or person ID), age, personality, desired job type or desired assignment destination, and skills. For example, the request may include the target talent's current job type and current affiliation. Also, the request may include various elements exemplified in the first exemplary embodiment.

[0042] The graph generating unit 202 refers to the request received by the receiving unit 201, and generates a target talent graph that represents the target talent in a graph based on information on the target talent indicated by the request about the target talent to which the user wishes to determine a department. Specifically, the graph generating unit 202 generates a target talent graph that includes multiple nodes related to the skills or work history of the target talent and links indicating the relationships between the nodes. With this configuration, it is possible to estimate candidate departments by the estimation unit 208, which will be described later, taking into consideration not only the skills and work history of the target talent but also the relationships therebetween. Specific processing by the graph generating unit 202 will be described later.

[0043] The link prediction unit 203 uses the target talent graph generated by the graph generation unit 202, which includes multiple nodes related to the target talent, and the receiving destination graph, to perform link prediction to predict the relationship between nodes that are not connected by links in the target talent graph and the receiving destination graph, and predicts talent nodes included in the receiving destination graph that indicate talent belonging to the receiving destination, which will link to nodes included in the target talent graph.

[0044] Here, the receiving graph is, for example, a graph including receiving organizations that may receive the target human resource, a plurality of nodes related to the skills or work history of each of the plurality of people, and links indicating the relationships between the nodes. In other words, the receiving graph is a graph in which one or more people are represented by nodes representing the person's affiliation, skills, and work history, and edges representing the relationships between the nodes. The receiving graph is a learned graph regarding the relationships between the nodes, and is a learned model. The receiving graph can also be called a knowledge graph. Note that a graph corresponding to one person may be called a receiving graph, and a graph corresponding to multiple people may be called a receiving graph.

[0045] According to the above configuration, the talent nodes in the receiving organization graph that are linked to the nodes included in the target talent graph are predicted, and potential assignment destinations are estimated based on the predicted talent nodes. Since how the target talent is related to which talent in the receiving organization and how the target talent is related is useful in the personnel management of the target talent, according to the above configuration, personnel support that takes into account the talent in the receiving organization that is related to the target talent is realized.

[0046] The identifying unit 204 identifies compatible personnel who are compatible with a specific personnel from among the personnel included in the receiving organization graph.

[0047] Here, the link prediction unit 203 may be configured to predict similar talents who are similar to the target talent among talents belonging to the receiving organization by the link prediction. In this configuration, the identification unit 204, as one example, identifies compatible talents who are talents whose nodes and links included in the receiving organization graph indicate good compatibility with the similar talents among talents belonging to the receiving organization.

[0048] According to the above configuration, the talent nodes indicating similar talents to the target talent are predicted, and the nodes and links included in the receiving party graph are identified as compatible talents indicating good compatibility with the similar talents. The compatible talents that are compatible with the similar talents are likely to be compatible with the target talent.

[0049] In other words, the above-mentioned configuration makes it possible to identify, from among the personnel belonging to the hosting organization, personnel who are likely to be compatible with the target personnel. Therefore, the above-mentioned configuration makes it possible to provide useful information for determining the compatibility between the target personnel and the hosting organization.

[0050] Alternatively, the identifying unit 204 may identify, from among the human resources included in the recipient graph, human resources who are incompatible with the specific human resource.

[0051] According to the above configuration, the talent node indicating the similar talent similar to the target talent is predicted, and the nodes and links included in the receiving graph identify talent that is incompatible with the similar talent. Talent that is incompatible with the similar talent is likely to be incompatible with the target talent.

[0052] In other words, the above-mentioned configuration makes it possible to identify personnel who are likely to be incompatible with the target personnel from among the personnel belonging to the hosting organization. Therefore, the above-mentioned configuration makes it possible to provide useful information for determining the compatibility between the target personnel and the hosting organization.

[0053] The compatibility determination unit 205 determines the compatibility between the target talent and the receiving organization based on the similarity indicating the degree of similarity between each talent belonging to the receiving organization and the compatible talent identified by the identification unit 204.

[0054] A host organization that has personnel similar to the compatible personnel is likely to be compatible with the target personnel. Therefore, according to the above configuration, the compatibility between the target personnel and the host organization is determined based on the similarity indicating the degree of similarity between each of the personnel belonging to the host organization and the compatible personnel. This makes it possible to accurately determine the compatibility between the target personnel and the host organization.

[0055] Alternatively, the compatibility determination unit 205 may determine the compatibility between the target talent and the receiving organization based on a similarity indicating the degree of similarity between each talent belonging to the receiving organization and the talent with poor compatibility identified by the identification unit 204.

[0056] A host organization that has human resources similar to the incompatible human resources is likely to have a poor compatibility with the target human resources. Therefore, according to the above configuration, the compatibility between the target human resources and the host organization is determined based on the similarity indicating the degree of similarity between each human resource belonging to the host organization and the incompatible human resources. This makes it possible to accurately determine the compatibility between the target human resources and the host organization.

[0057] In addition, the compatibility determination unit 205 may be configured to identify a department to which the compatible talent identified by the identification unit 204 belongs among multiple departments included in the receiving destination, and determine the compatibility between the department and the target talent based on a similarity indicating the degree of similarity between each talent belonging to the department and the compatible talent.

[0058] The department to which the compatible talent belongs is likely to be compatible with the target talent, and if the department has talent similar to the compatible talent, the compatibility is even more likely. Therefore, with the above configuration, the department to which the compatible talent belongs is identified, and the compatibility between the department and the target talent is determined based on the similarity indicating the degree of similarity between each talent belonging to the department and the compatible talent. This makes it possible to accurately determine the compatibility between the target talent and the department where the target talent will be accepted.

[0059] Alternatively, the compatibility determination unit 205 may be configured to identify a department among multiple departments included in the receiving organization to which the incompatible personnel identified by the identification unit 204 belongs, and determine the compatibility between the department and the target personnel based on a similarity indicating the degree of similarity between each personnel belonging to the department and the incompatible personnel.

[0060] The department to which the incompatible human resource belongs is likely to be incompatible with the target human resource, and if the department contains human resources similar to the incompatible human resource, the incompatibility is even greater. Therefore, according to the above configuration, the department to which the incompatible human resource belongs is identified, and the compatibility between the department and the target human resource is determined based on the similarity indicating the degree of similarity between each human resource in the department and the incompatible human resource. This makes it possible to accurately determine the compatibility between the target human resource and the department where the target human resource will be accepted.

[0061] The recommendation unit 206 refers to the results of the processing by at least one of the link prediction unit 203, the identification unit 204, and the compatibility determination unit 205, and determines, for each of the multiple target talents, a department that is recommended as an acceptance destination for the target talent.

[0062] Here, the compatibility determination unit 205 may be configured to determine the compatibility between each of the multiple target human resources and each of the multiple departments. In this configuration, the recommendation unit 206, for example, determines a department to be recommended as a place to accept each of the multiple target human resources based on the compatibility determination result by the compatibility determination unit 205.

[0063] According to the above configuration, the compatibility between each of the multiple target human resources and each of the multiple departments is determined, and a department to be recommended as a host for each of the multiple target human resources is determined based on the result of the determination. This makes it possible to recommend a host department that takes into account the compatibility between each department and each target human resource.

[0064] It should be noted that the process by the recommendation unit 206 is not limited to the above example. For example, the recommendation unit 206 may be configured to recommend the organization to which the compatible human resource identified by the identification unit 204 belongs as a destination for the target human resource.

[0065] The learning unit 207 learns the relationships between the nodes included in the receiving destination graph based on various information on multiple existing employees, and generates a learned receiving destination graph. Unless otherwise specified, the receiving destination graph refers to one that has been learned by the learning unit 207. The learned receiving destination graph may also be loaded into the recruiting support device 2, in which case the learning unit 207 may be omitted.

[0066] The estimation unit 208 is configured to estimate candidate destinations for the target human resource that match the target human resource request, based on the trained model and a request for destinations for the target human resource received by the reception unit 201. Here, the trained model is a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of persons and the destinations to which each of the plurality of persons belongs.

[0067] In this exemplary embodiment, as an example, the trained model is a graph including multiple nodes relating to potential recipients of the target personnel, the skills or work history of each of the multiple individuals, and links indicating the relationships between the nodes, i.e., the recipient graph described above.

[0068] According to the above configuration, it is possible to estimate, in consideration of various information related to the hosting destination, the candidate destinations for the target human resource that match the request received by the reception unit 201. Therefore, according to the above configuration, it is possible to obtain an effect that it is possible to provide personnel support for the target human resource in consideration of various information related to the hosting destination.

[0069] Furthermore, the estimation unit 208 may estimate a candidate destination for the target human resource that matches the request received by the reception unit 201, based on the node predicted by the link prediction unit 203 through the above-mentioned process. Furthermore, the estimation unit 208 may estimate a candidate destination for the target human resource that matches the request, by referring to the determination result by the recommendation unit 206.

[0070] The basis generating unit 209 generates basis information indicating the basis of the estimation by the estimation unit 208. Various methods can be applied as a method for generating the basis information. The method for generating the basis information will be described later.

[0071] As described above, the output unit 210 outputs various information generated by the recruiting support device 2, such as information indicating potential assignment destinations estimated by the estimation unit 208. The destination of the information output is arbitrary, and for example, if the recruiting support device 2 is equipped with an output device as described above, the information may be output to that output device. Also, for example, the information may be output to an output device external to the recruiting support device 2.

[0072] (How to generate a target talent graph) 5 is a schematic diagram showing an example of a method for generating a target talent graph by the graph generating unit 202. The graph generating unit 202 first refers to the request received by the receiving unit 201, and identifies information on the target talent (applicant) to whom the user wishes to determine the assignment, which is indicated by the request.

[0073] Figure 5 shows that the request Personality: Strong sense of responsibility Age: 30s Skills: Project Management Desired job type: Business planning The target talent graph generated by the graph generating unit 202 is shown when the information "The types of nodes included in the target talent graph generated by the graph generating unit 202 are not limited to those shown in the figure, and may include various types of nodes described above.

[0074] (Example 1 of processing related to identifying similar talent, identifying compatible talent, and judging the compatibility between the department and the target talent) 6 and 7 are diagrams for explaining a first processing example relating to a method for identifying similar people by the link prediction unit 203, a method for identifying compatible people by the identification unit 204, and a compatibility determination method by the compatibility determination unit 205.

[0075] Figure 6 shows an acceptance graph in which existing employees A to C are represented by nodes that represent the personality, age, skills, affiliation, and work history of the existing employees, and links that represent the relationships between the nodes. For example, the node for "Existing employee A" and the node for "Strong sense of responsibility" are connected by an edge that represents "personality," which shows that existing employee A has a strong sense of responsibility.

[0076] 6 includes nodes for multiple existing employees A to C, but a graph consisting of only a node related to one existing employee may be called an existing employee graph. In this case, the link prediction unit 203 identifies similar personnel using multiple existing employee graphs corresponding to each of the multiple existing employees.

[0077] Such a destination graph can be generated from the personality, age, skill, affiliation, and work history of each existing employee. Also, by learning the relationship between the personality, age, skill, and work history shown in the destination graph and the affiliation, it becomes possible to infer the affiliation suitable for the target human resource from the target human resource's personality, age, skill, desired job type, etc.

[0078] 6 also shows a target talent graph in which the target talent (applicant) is represented by nodes representing the personality, age, skill, or desired occupation of the target talent, and edges (links) representing the relationships between the nodes. In more detail, the target talent graph shown in FIG. 6 includes, as in FIG. 5, nodes and links indicating that the target talent has a personality of "high sense of responsibility," nodes and links indicating that the target talent has an age of "30s," nodes and links indicating that the target talent has a skill of "project management," and nodes and links indicating that the target talent has a desired occupation of "business planning." As described above, the target talent graph is generated by the graph generation unit 202.

[0079] The link prediction unit 203 uses the target talent graph and the recipient graph generated in this way to perform link prediction for predicting relationships between nodes not connected by links in the target talent graph and the recipient graph. By performing the link prediction, the link prediction unit 203 predicts, as an example, talent nodes that are linked to nodes included in the target talent graph from talent nodes included in the recipient graph that indicate talent belonging to the recipient.

[0080] For example, as shown by the dashed line in Fig. 6, there is no link between the node of "strong sense of responsibility" in the target personnel graph and the node of "strong sense of responsibility" in existing employee A. By performing link prediction, the link prediction unit 203 can predict the probability that the relationship between these nodes is "same". Then, based on the predicted probability, the link prediction unit 203 can identify a node in the existing employee graph that links to a node included in the target personnel graph. For example, the link prediction unit 203 may identify a node in the existing employee graph whose predicted probability value is equal to or greater than a threshold value as a node that links to a node in the target personnel graph.

[0081] In addition, the link prediction unit 203 can also predict nodes that link to nodes included in the target personnel graph from among nodes included in the existing employee graph that include nodes that meet pre-set conditions or conditions set by the user.

[0082] For example, the link prediction unit 203 can predict nodes that link to nodes included in the graph of the target personnel from among nodes included in the existing employee graph that includes nodes that match predetermined characteristics and skills.

[0083] Furthermore, the link prediction unit 203 can also specify existing employees who have a predetermined relationship with the target talent by using the target talent graph and the recipient graph. For example, in addition to being able to specify existing employees similar to the target talent, it is also possible to specify existing employees who are dissimilar to the target talent, existing employees who belong to the same classification as the target talent, existing employees who have characteristics in common with the target talent, etc.

[0084] The above identification can be realized by link prediction for predicting the relationship between nodes that are not connected by a link in the target talent graph and the accepting company graph. FIG. 7 is a diagram for explaining the link prediction. Like FIG. 6, FIG. 7 shows a target talent (applicant) graph and an accepting company graph including graphs of existing employees A to C. As shown by the dashed line in FIG. 7, there is no link between the "applicant" node in the target talent graph and the "existing employee A" node in the accepting company graph.

[0085] By performing link prediction, the link prediction unit 203 can predict the probability that the relationship between these nodes is "similar". Similarly, the link prediction unit 203 can predict the probability that the relationship between the "applicant" node and the nodes of existing employees B and C included in the receiving company graph is "similar". Then, the link prediction unit 203 can identify similar personnel based on the predicted probability. For example, the link prediction unit 203 may identify existing employees whose predicted probability value is equal to or greater than a threshold as similar personnel. In the example shown in FIG. 7, the link prediction unit 203 identifies existing employee A as a similar personnel who is similar to the applicant.

[0086] The link prediction unit 203 can also identify existing employees who meet preset conditions or conditions set by the user as existing employees who have a predetermined relationship with the target talent. For example, it is possible to identify an existing employee who has at least a partial commonality with the target talent as a similar talent, or to identify an existing employee who has at least a partial commonality with the target talent as a similar talent.

[0087] The identification unit 204 refers to the graph of similar talent identified as described above, and identifies from among the talent belonging to the host company, talent with good compatibility, which is talent whose nodes and links included in the host company graph indicate good compatibility with the similar talent.

[0088] In the example shown in FIG. 7, the identification unit 204 identifies that existing employee A and existing employee B, who are similar personnel, have a high affinity. In this case, the identification unit 204 identifies existing employee B as a compatible personnel. Here, the affinity calculation process by the identification unit 204 can be performed by referring to the behavioral data of each person, as an example. For example, when the similarity between the location information history of existing employee A and the location information history of existing employee B is equal to or greater than a predetermined value, it may be determined that existing employee A and existing employee B have a high affinity. Alternatively, when the behavioral data of existing employee A and existing employee B indicates that they are each other's main communication partners, it may be determined that existing employee A and existing employee B have a high affinity.

[0089] The compatibility determination unit 205 determines the compatibility between the target talent and the receiving organization based on the similarity indicating the degree of similarity between each talent belonging to the receiving organization and the compatible talent.

[0090] The link prediction unit 203 can predict the probability (similarity) that the relationship between a node of an existing employee and a node of another existing employee included in the receiving company graph is "similar." In the example shown in Fig. 7, the compatibility determination unit 205 calculates that the similarity between existing employees B, who are compatible personnel, and existing employee C is x, and calculates that the similarity between existing employees B, who are also compatible personnel, and existing employee D is y.

[0091] The compatibility determination unit 205 determines the compatibility between the target human resource and the receiving organization based on the similarity predicted by the link prediction unit 203. For example, the compatibility determination unit 205 may determine an organization to which an existing employee belongs whose similarity to the compatible human resource, existing employee B, is equal to or greater than a threshold value as a compatible receiving organization, and may determine an organization to which an existing employee belongs whose similarity to the compatible human resource, existing employee B, is less than the threshold value as a poorly compatible receiving organization.

[0092] A host organization that has personnel similar to the compatible personnel is likely to be compatible with the target personnel. Therefore, according to the above configuration, the compatibility between the target personnel and the host organization is determined based on the similarity indicating the degree of similarity between each of the personnel belonging to the host organization and the compatible personnel. This makes it possible to accurately determine the compatibility between the target personnel and the host organization.

[0093] (Processing flow according to processing example 1) The flow of the process (recruitment support method) executed by the recruitment support device 2 will be described with reference to Fig. 8. Fig. 8 is a flow diagram showing the process executed by the recruitment support device 2 and the process according to the above-mentioned process example 1.

[0094] In S201, the reception unit 201 receives a request regarding the assignment of the target human resource. In S201, a request including, for example, the age, personality, desired job type, and skills of the target human resource is received. In other words, as described above, the request includes, for example, at least one of the age, personality, desired job type, and skills of the target human resource.

[0095] In S202, the graph generating unit 202 refers to the request received in S201, and generates a target talent graph that represents the target talent in a graph based on the information on the target talent indicated by the request. For example, in S202, a target talent graph including nodes indicating the age, personality, desired occupation, or skills of the target talent, and links indicating the relationships between the nodes may be generated.

[0096] In S203, the link prediction unit 203 predicts a node that links to a node included in the target talent graph generated in S202. As described above, this node is predicted by link prediction using the learned recipient graph and the target talent graph.

[0097] In S203, the link prediction unit 203 may, for example, predict a node connected to the "target human resource" node via a "personality" or "skill" link. Alternatively, a node connected to a node indicating personality or skill may be predicted. For example, a node connected to a node indicating skill via a "qualification" link may be predicted. This makes it possible to predict a node connected to a more detailed node relating to the personality and skills of the target human resource.

[0098] In S204, the identification unit 204 identifies, from among the human resources belonging to the receiving organization, human resources with good compatibility, which are human resources whose nodes and links included in the receiving organization graph indicate good compatibility with the similar human resources.

[0099] In S205, the compatibility determination unit 205 determines the compatibility between the target talent and the receiving organization based on the similarity indicating the degree of similarity between each talent belonging to the receiving organization and the compatible talent identified in S204.

[0100] In S206, the estimation unit 208 estimates candidate destinations for the target human resource that match the request based on the trained model and the request regarding the destination of the target human resource received in S201.

[0101] In S206, the estimation unit 208 may refer to the result of the determination by the compatibility determination unit 205 in S205, and estimate that receiving destinations having a degree of compatibility equal to or higher than a predetermined level are candidate destinations of assignment suitable for the request.

[0102] In S207, the basis generating unit 209 generates basis information indicating the basis of the estimation by the estimation unit 208. Specifically, the basis information may include the similarity between the attributes of the target human resource and the attributes of the people belonging to the candidate destinations estimated by the estimation unit 208 in S206.

[0103] In S208, the output unit 210 outputs the destination candidate estimated in S206. The output unit 210 may be configured to output the basis information generated in S207 together with the destination candidate estimated in S206. This ends the processing in FIG. 8.

[0104] It is also possible to omit the process of S205 and estimate the affiliation of the compatible person identified in S204 as the potential assignment destination of the target talent. Also, the link prediction unit 203 can directly predict the potential assignment destination of the target talent by link prediction using the target talent graph and the recipient graph. This is because the relationship between the talent and the recipient that is compatible with the talent is learned by taking into account the similarity and compatibility between the talents when learning the recipient graph, without going through the process of identifying similar talents or compatible talents. In this case, S204 to S205 are omitted.

[0105] (Example 2 of processing for identifying similar talent, identifying talent with good compatibility, determining compatibility between departments and target talent, and determining recommended departments) FIG. 9 is a diagram for explaining a second processing example relating to a method for identifying similar people by the link prediction unit 203, a method for identifying compatible people by the identification unit 204, and a compatibility determination method by the compatibility determination unit 205. In FIG.

[0106] Fig. 9 shows an acceptance graph including nodes indicating the affiliations of existing employees A1 to A3, existing employees B1 to B3, and existing employees C1 to C3, and links indicating the relationships between the nodes. The acceptance graph shown in Fig. 9 also shows a node indicating existing employee A. In the example shown in Fig. 9, it is expressed that existing employees A1 to A3 belong to the sales department, existing employees B1 to B3 belong to the planning department, and existing employees C1 to C3 belong to the manufacturing department.

[0107] It should be noted that there are links between existing employees A, existing employees A1 to A3, existing employees B1 to B3, and existing employees C1 to C3 shown in FIG. 9 and other nodes, but these are not shown in FIG.

[0108] 9 includes nodes for multiple existing employees A1 to C3, a graph consisting of only a node related to one existing employee may be called an existing employee graph. In this case, the link prediction unit 203 identifies similar personnel using multiple existing employee graphs corresponding to each of the multiple existing employees.

[0109] Figure 9 also shows a target talent graph, in which the target talent (applicants) are represented, similarly to Figures 5 and 7, with nodes representing the personality, age, skills, or desired job type of the target talent, and edges (links) representing the relationships between the nodes.

[0110] Even in the example shown in Fig. 9, the link prediction unit 203 can identify existing employees who have a predetermined relationship with the target talent by using the target talent graph and the receiving company graph. Moreover, the above identification can be realized by link prediction for predicting the relationship between nodes that are not connected by a link in the target talent graph and the receiving company graph. For example, as shown by the dashed line in Fig. 9, there is no link between the "applicant" node in the target talent graph and the "existing employee A" node in the receiving company graph.

[0111] By performing link prediction, the link prediction unit 203 can predict the probability that the relationship between these nodes is "similar". Similarly, the link prediction unit 203 can predict the probability that the relationship between the "applicant" node and the nodes of existing employees A1 to A3, B1 to B3, and C1 to C3 included in the employment graph is "similar". Then, the link prediction unit 203 can identify similar personnel based on the predicted probability. For example, the link prediction unit 203 may identify existing employees whose predicted probability value is equal to or greater than a threshold as similar personnel.

[0112] 9, the link prediction unit 203 can also identify existing employees who meet preset conditions or conditions set by the user as existing employees who have a predetermined relationship with the target talent. For example, it is possible to identify an existing employee who has at least a part of a personality in common with the target talent as a similar talent, or to identify an existing employee who has at least a part of a skill in common with the target talent as a similar talent.

[0113] The identification unit 204 refers to the graph of similar talent identified as described above, and identifies from among the talent belonging to the host company, talent with good compatibility, which is talent whose nodes and links included in the host company graph indicate good compatibility with the similar talent.

[0114] 9, the identification unit 204 identifies that existing employee A, who is a similar human resource, has a high affinity with existing employees A1 to A3. In this case, the identification unit 204 identifies existing employees A1 to A3 as compatible human resources. Here, the affinity calculation process by the identification unit 204 can be performed by referring to the behavioral data of each person as described in the processing example 1, for example.

[0115] The compatibility determination unit 205 identifies the department to which the compatible talent identified by the identification unit 204 belongs among the multiple departments included in the receiving company, and determines the compatibility between the department and the target talent based on the similarity indicating the degree of similarity between each talent belonging to the department and the compatible talent.

[0116] 9, the link prediction unit 203 calculates the similarity between existing employee A1, who is a compatible human resource, and existing employees A2 and A3 who belong to the same department (sales department). Then, the compatibility determination unit 205 determines the compatibility between the department (sales department) and the target human resource based on the similarity between existing employees A1 and A2 and the similarity between existing employees A1 and A3.

[0117] For example, if the similarity between existing employees A1 and A2 and the similarity between existing employees A1 and A3 are both equal to or greater than a predetermined threshold, it may be determined that the sales department and the target personnel are compatible. By carrying out the same process for existing employees B1 to B3 and existing employees C1 to C3, it is possible to determine the compatibility between the planning department and the target personnel, and the compatibility between the manufacturing department and the target personnel.

[0118] In the example shown in Figure 9, only one target talent is shown, but the link prediction unit 203, the identification unit 204, and the compatibility determination unit 205 can determine the compatibility between multiple target talents and multiple departments by processing similar to that described above.

[0119] The department to which the compatible talent belongs is likely to be compatible with the target talent, and if the department has talent similar to the compatible talent, the compatibility is even more likely. Therefore, with the above configuration, the department to which the compatible talent belongs is identified, and the compatibility between the department and the target talent is determined based on the similarity indicating the degree of similarity between each talent belonging to the department and the compatible talent. This makes it possible to accurately determine the compatibility between the target talent and the department where the target talent will be accepted.

[0120] The recommendation unit 206 determines, for each of the multiple target human resources, a department recommended as a place of employment for the target human resources, based on the compatibility determination result by the compatibility determination unit 205. FIG.

[0121] 10, the degree of compatibility between applicant 1 and the sales department is estimated to be 0.8, the degree of compatibility between applicant 1 and the planning department is estimated to be 0.7, and the degree of compatibility between applicant 1 and the manufacturing department is estimated to be 0.5 by the processing by the link prediction unit 203, the identification unit 204, and the compatibility determination unit 205. In addition, although omitted in FIG. 10, the degree of compatibility between other applicants and each department is also estimated.

[0122] With reference to such compatibility estimation results, the recommendation unit 206 determines a department to be recommended as an employment destination for each of the applicants 1 to 3. For example, for the applicant 1, the sales department, which is the department with the highest degree of compatibility, is determined to be the recommended department.

[0123] The recommendation unit 206 may determine the department that maximizes the total affinity (degree of compatibility) while satisfying the constraints (such as the number of people to be assigned to each department) specified by the user. In this case, an optimization solver such as MaxSAT (Maximum SATisfiability) may be used.

[0124] According to the above configuration, the compatibility between each of the multiple target human resources and each of the multiple departments is determined, and a department to be recommended as a host for each of the multiple target human resources is determined based on the result of the determination. This makes it possible to recommend a host department that takes into account the compatibility between each department and each target human resource.

[0125] (Processing flow according to processing example 2) The flow of the process (recruitment support method) executed by the recruitment support device 2 will be described with reference to Fig. 11. Fig. 11 is a flow diagram showing the process executed by the recruitment support device 2 and the process according to the above-mentioned process example 2.

[0126] S201 to S204 are similar to S201 to S204 shown in FIG. 8, and therefore the description will be omitted.

[0127] After S201, in S205a, the compatibility determination unit 205 identifies, among the multiple departments included in the receiving destination, the department to which the compatible talent identified by the identification unit 204 in S204 belongs, and determines the compatibility between the department and the target talent based on the similarity indicating the degree of similarity between each talent belonging to the department and the compatible talent.

[0128] After S205a, in S205b, the recommendation unit 206 determines, for each of the plurality of target human resources, a department to be recommended as a place of employment for the target human resources, based on the compatibility determination result by the compatibility determination unit 205 in S205a.

[0129] After S209, in S206, the estimation unit 208 estimates an assignment destination of the target talent that matches the target talent request based on the trained model and the request regarding the assignment destination of the target talent received in S201.

[0130] In S206, the estimation unit 208 may estimate, as an example, the recommended department determined by the recommendation unit 206 in S205b as a candidate for the assignment of the target human resource that matches the request received by the reception unit 201.

[0131] In S207, the basis generating unit 209 generates basis information indicating the basis of the estimation by the estimation unit 208. Specifically, the basis information may include the similarity between the attributes of the target human resource and the attributes of the people belonging to the candidate destinations estimated by the estimation unit 208 in S206.

[0132] In S208, the output unit 210 outputs the destination candidate estimated in S206. The output unit 210 may be configured to output the destination candidate estimated in S206 together with the basis information generated in S207. This ends the processing in FIG. 11.

[0133] Fig. 12 is a diagram showing an example of the estimation result output by the output unit 210 according to this processing example. As shown in Fig. 12, the output unit 210 presents, as an example, the recommendation level calculated based on the judgment result of the compatibility judgment unit 205 together with the assignment candidate which is the estimation result by the estimation unit 208.

[0134] (How to generate evidence information) A method for generating the basis information by the basis generating unit 209 will be described. As described above, various methods can be applied as a method for generating the basis information. As an example, the basis generating unit 209 generates basis information including the similarity between the attributes of the target human resource and the attributes of the people belonging to the candidate destination estimated by the estimation means. The generated basis information is output by the output unit 210.

[0135] Here, the attributes of the target human resources may include, but are not limited to, age, personality, skills, etc. As an example, the attributes of the target human resources may include any of the elements of the nodes described in [About Graphs and Learning].

[0136] As an example, the basis generating unit 209 can generate basis information such as "The similarity between the personality of the target personnel and the personality of existing employee A who belongs to the sales department is 0.8."

[0137] According to the above configuration, evidence information is generated that includes the degree of similarity between the attributes of the target personnel and the attributes of the people who belong to the estimated candidate destinations. This allows the user to refer to the candidate destinations based on the evidence. Since ensuring transparency is particularly important in personnel management, the ability to generate evidence information is a major advantage.

[0138] Furthermore, the basis generating unit 209 may generate basis information based on the result of link prediction by the link predicting unit 203. In this case, the link predicting unit 203 uses the destination graph and the target talent graph, which include nodes indicating attributes possessed by the target talent, to predict the probability that a node indicating the attribute will be linked to a node included in the target talent graph. Then, the basis generating unit 209 generates basis information according to the predicted probability.

[0139] As an example, the basis generating unit 209 can generate basis information such as "the probability that the skills of the target personnel are linked to the skills of existing employee A who belongs to the sales department is 0.9."

[0140] With the above-mentioned configuration, the user can refer to the candidate departments based on the reason.

[0141] (Generating evidence for link prediction results) The basis generating unit 209 can also generate basis information by analyzing the target talent graph and the receiving destination graph. The following describes a method for generating basis information by analyzing the target talent graph and the receiving destination graph.

[0142] For example, the basis generating unit 209 may mine one or more rules from the target personnel graph and the recipient graph using a PCA (Principal Component Analysis) reliability based on the OWA (Open-world assumption).The basis generating unit 209 may then generate basis information using the mined one or more rules.The method described in the following document, for example, may be applied to rule mining.

[0143] Luis Galarraga et. al, “Fast rule mining in ontological knowledge bases with AMIE +”, The VLDB Journal (2015) 24:707-730 As an example, the rule to be processed by the basis generating unit 209 is expressed as follows, using Head r(x, y) and Body {B1, . . . , Bn}:

number

number

[0144] The basis generating unit 209 sets the following as a condition for the mining process: Connected: All values ​​(variables, entities) in the rule are shared between different atoms. Closed: All variables in the rule must appear at least twice. Not reflexive: Rules that contain reflective atoms, such as r(x, x), will not be mined. Mining processing is carried out subject to the above conditions.

[0145] In addition, the reason generating unit 209

number

number

[0146] For example, suppose that the basis generating unit 209 mines a rule that, for two people who satisfy the conditions of "common personality" or "common skills", "an element contained in one person can be applied to the other person." In this case, when the link predicting unit 203 predicts an element contained in an existing employee as an element contained in the target talent, the basis generating unit 209 may generate basis information indicating that the existing employee and the target talent "have common personality" or "have common skills" as the basis for this prediction.

[0147] By presenting the above-mentioned grounds information, the user can refer to the candidate assignment destinations based on the grounds.

[0148] Exemplary embodiment 3 The third exemplary embodiment of the present invention will be described in detail with reference to the drawings. The configuration of the recruitment support device according to this exemplary embodiment is similar to that of the recruitment support device 2 according to the exemplary embodiment 2. However, in the recruitment support device according to this exemplary embodiment, the processing by the link prediction unit 203 and the estimation unit 208 is different from that of the recruitment support device 2 according to the exemplary embodiment 2.

[0149] The link prediction unit 203 in this exemplary embodiment uses a recipient graph including potential recipients of the target talent, multiple nodes related to the skills or work history of each of the multiple individuals, and links indicating the relationships between the nodes, and a target talent graph including multiple nodes related to the target talent, to perform link prediction to predict relationships between nodes that are not connected by links in the target talent graph and the recipient graph, thereby identifying talent or departments at the recipient who have a predetermined relationship with the target talent.

[0150] Then, the estimation unit 208 estimates candidate departments for the target human resource that match the request, based on the human resource or department of the receiving organization identified by the link prediction unit 203.

[0151] According to the above configuration, a person or department of the receiving company having a predetermined relationship with the target person is identified, and based on the person or department, a candidate for the target person's assignment that matches the request received by the reception unit 201 is estimated. Since information on the person or department having a predetermined relationship with the target person is useful information for the target person's personnel affairs, according to the above configuration, personnel affairs support for the target person can be performed accurately.

[0152] The processing by the link predictor 203 and the estimator 208 according to this exemplary embodiment will be described below with reference to FIGS.

[0153] (Processing example 1 regarding estimation of relationships and estimation of candidate assignments) FIG. 13 is a diagram for explaining a processing example 1 regarding estimation of a relationship between a target talent and an existing employee and estimation of a candidate assignment destination based on the estimated relationship by the link prediction unit 203 and the estimation unit 208 according to this exemplary embodiment.

[0154] FIG. 13 shows an acceptance graph in which existing employees A to C are represented by nodes that represent the personality, age, skills, affiliation, and work history of the existing employees, and edges (links) that represent the relationships between the nodes.

[0155] Note that the receiving graph shown in FIG. 13 includes nodes for multiple existing employees, existing employees A to C, but a graph consisting of only a node related to one existing employee may be called a receiving graph.

[0156] This type of recipient graph can be generated from the personality, age, skill, affiliation, and work history of each existing employee, as in the exemplary embodiment 2. In addition, by learning the relationship between the personality, age, skill, and work history shown in the recipient graph and the affiliation, it becomes possible to infer the relationship between the personality, age, skill, and desired job type of the target human resource and the affiliation of the existing employee.

[0157] Furthermore, in this exemplary embodiment, by learning the relationships between the nodes of existing employees included in the receiving graph, it is possible to infer the relationships between people. As an example, in the example shown in Fig. 13, the node of existing employee A and the node of existing employee B are connected by a link indicating that "existing employee B is a good subordinate to existing employee A", and the node of existing employee A and the node of existing employee C are connected by a link indicating that "existing employee A respects existing employee C".

[0158] Learning such relationships between existing employees in the receiving graph is possible, for example, by learning by referring to nodes that indicate the evaluation history, interview history, and behavioral data of each existing employee, but this is not a limitation of this exemplary embodiment.

[0159] Figure 6 also shows a target talent graph, which represents target talent (applicants) with nodes representing the personality, age, skills, or desired job type of the target talent, and edges (links) representing the relationships between the nodes.

[0160] The link prediction unit 203 uses the target talent graph and the recipient graph generated in this way to perform link prediction for predicting the relationship between nodes that are not connected by links in the target talent graph and the recipient graph. Then, by performing the link prediction, the link prediction unit 203 identifies, as one example, the recipient talent who has a predetermined relationship with the target talent.

[0161] For example, as shown in Fig. 13, there is no link between the node of the target talent (applicant) and the node of existing employee A. By performing link prediction, the link prediction unit 203 can predict the probability that the relationship between these nodes is a predetermined relationship. For example, the link prediction unit 203 can predict the probability that the relationship between these nodes is "the applicant is a good subordinate to existing employee A" or "the applicant respects existing employee A."

[0162] Then, the link prediction unit 203 may specify an existing employee node for which the predicted probability value for a certain relationship is equal to or greater than a threshold value as a node that links to the target personnel node through that relationship.

[0163] The estimation unit 208 estimates potential assignment destinations for the target human resource that match the request received by the reception unit 201, based on the human resources of the receiving organization identified by the link prediction unit 203 in this manner. As an example, the estimation unit 208 may estimate, as a potential assignment destination for the target human resource that matches the request, the organization of an existing employee who is predicted to have a good relationship with the target human resource (for example, "the applicant is a good subordinate to existing employee A" or "the applicant respects existing employee A").

[0164] In addition, the link prediction unit 203 can also predict existing employees who are linked to the node of the target personnel in a specified relationship from among the nodes included in the existing employee graph, which includes nodes that meet pre-set conditions or conditions set by the user.

[0165] In addition, the "predetermined relationship" is not limited to the above example. By using the target talent graph and the recipient graph, the link prediction unit 203 can specify existing employees similar to the target talent, existing employees dissimilar to the target talent, existing employees belonging to the same classification as the target talent, existing employees who have characteristics in common with the target talent, etc.

[0166] (Processing example 1 regarding estimation of relationships and estimation of candidate assignments) FIG. 14 is a diagram for explaining a processing example 1 regarding estimation of a relationship between a target talent and an existing employee and estimation of a candidate assignment destination based on the estimated relationship by the link prediction unit 203 and the estimation unit 208 according to this exemplary embodiment.

[0167] FIG. 14 shows an acceptance graph in which existing employees A to C are represented by nodes that represent the personality, age, skills, affiliation, and work history of the existing employees, and edges (links) that represent the relationships between the nodes.

[0168] Note that the receiving graph shown in FIG. 14 includes nodes for multiple existing employees, existing employees A to C, but a graph consisting of only a node related to one existing employee may be called a receiving graph.

[0169] This type of destination graph can be generated from the personality, age, skill, affiliation, and work history of each existing employee, as in the exemplary embodiment 2. In addition, by learning the relationship between the personality, age, skill, and work history shown in the destination graph and the affiliation, it becomes possible to infer the affiliation suitable for the target human resource according to the target human resource's personality, age, skill, and desired job type.

[0170] Furthermore, in this exemplary embodiment, by learning the relationship between the nodes of existing employees included in the receiving company graph and their affiliations, it is possible to infer the relationship between a person and their affiliations.

[0171] Figure 14 also shows a target talent graph, which represents target talent (applicants) with nodes representing the personality, age, skills, or desired job type of the target talent, and edges (links) representing the relationships between the nodes.

[0172] The link prediction unit 203 uses the target talent graph and the recipient graph thus generated to perform link prediction for predicting the relationship between nodes not connected by links in the target talent graph and the recipient graph. Then, by performing the link prediction, the link prediction unit 203 identifies, as an example, a recipient department that has a predetermined relationship with the target talent.

[0173] For example, as shown in Fig. 14, there is no link between the node of the target human resource (applicant) and the node of "affiliation" in the existing employee graphs A to C. The link prediction unit 203 can predict the probability that the relationship between these nodes is a predetermined relationship by performing link prediction. For example, the link prediction unit 203 can predict the probability that the relationship between these nodes is good.

[0174] Then, the link prediction unit 203 may specify an existing employee node for which the predicted probability value for a certain relationship is equal to or greater than a threshold value as a node that links to the target personnel node through that relationship.

[0175] The estimation unit 208 estimates candidate departments for the target human resource that match the request received by the reception unit 201, based on the departments of the receiving destination identified by the link prediction 203 in this manner. As an example, the estimation unit 208 may estimate a department of the receiving destination that is predicted to have a good relationship with the target human resource as a candidate department for the target human resource that matches the request.

[0176] Exemplary embodiment 4 A recruiting support device 4 according to a fourth exemplary embodiment of the present invention will be described with reference to the drawings. The recruiting support device 4 performs personnel support for a target talent. As a personnel support method, there is a case where it is determined whether or not the target talent has the job characteristics desired by the user. The recruiting support device 4 performs personnel support in such a case.

[0177] (overview) Fig. 15 is a diagram showing an overview of a recruitment support method according to this exemplary embodiment. In this exemplary embodiment, link prediction is performed using a target talent graph and a recipient graph, similar to exemplary embodiment 2. The target talent graph shown in the upper left corner of Fig. 15 includes nodes and links indicating that the target talent has a personality such as "strong sense of responsibility."

[0178] Moreover, the host graph shown in the upper right corner of Figure 15 includes nodes and links indicating that existing employee A has the personality of "high sense of responsibility" and that his / her work characteristics include "contributing to new business." Similarly, the host graph shown in the lower right corner of Figure 14 includes nodes and links indicating that existing employee B has the personality of "intense curiosity" and that his / her work characteristics include "suitable for overseas work."

[0179] By learning the above-mentioned recipient graphs for various existing employees, it becomes possible to link-predict what kind of person is likely to have what kind of work characteristics. In other words, in the recruitment support method according to the present exemplary embodiment, a target talent graph is generated, and the probability that the target talent shown in the target talent graph has the requested work characteristics is link-predicted.

[0180] For example, in the example in Figure 15, the probability that the "Applicant" node in the target personnel graph shown in the top left corner will be linked to the "Contribution to new business" node via the "Work characteristics" link is predicted to be 70%. From this, it can be said that this applicant is a person who is likely to "contribute to new business."

[0181] In this manner, according to the recruitment support method according to this exemplary embodiment, the probability that the target personnel have the desired job characteristics is presented to the user, thereby making it possible to support the recruitment of personnel.

[0182] (Device configuration) The configuration of the recruitment support device 4 according to the fourth exemplary embodiment of the present invention will be described with reference to Fig. 16. Fig. 16 is a block diagram showing the configuration of the recruitment support device 4 according to this exemplary embodiment.

[0183] As shown in the figure, the hiring support device 4 includes a reception unit 401, a graph generation unit 402, a link prediction unit 403, an estimation unit 405, a reason generation unit 406, and an output unit 407. Similarly to the hiring support device 2 of the exemplary embodiment 2, the hiring support device 4 may include, in addition to these components, an input device that receives input operations from a user, an output device for outputting data by the hiring support device 4, a communication device for the hiring support device 4 to communicate with other devices, and the like.

[0184] The reception unit 401 receives a request regarding the assignment of a target human resource. Here, the request includes information about the target human resource to which the user wants to determine the assignment. As an example, the request includes, but is not limited to, the target human resource's name (or person ID), age, personality, desired job type or desired assignment destination, and characteristics.

[0185] The graph generating unit 402 refers to the request received by the receiving unit 401, and generates a target talent graph that represents the target talent in a graph based on information about the target talent to which the user wishes to decide the assignment, which is indicated by the request. Specifically, the graph generating unit 402 generates a target talent graph that includes multiple nodes related to the skills, characteristics, or work history of the target talent, and links indicating the relationships between the nodes.

[0186] The link prediction unit 403 uses the target talent graph, which includes multiple nodes related to the target talent generated by the graph generation unit 202, and the recipient graph, to calculate the probability that a node showing a predetermined characteristic will be linked to a node included in the target talent graph by link prediction for predicting the relationship between nodes not connected by links in the target talent graph and the recipient graph. Here, the predetermined characteristic is a characteristic that matches the request received by the reception unit 401, and includes the above-mentioned business characteristics as an example.

[0187] The estimation unit 405 estimates the assignment destination of the target talent that matches the request based on the probability calculated by the link prediction unit 403. That is, the estimation unit 405 estimates the assignment destination candidate of the target talent that matches the target talent request based on the trained model and the request regarding the assignment destination of the target talent received by the reception unit 401. Here, the trained model is a trained model that has learned the relationship between at least one of the skills and work history of each of the multiple people and the affiliation of each of the multiple people. The estimation unit 405 performs the above estimation based on the result of link prediction by the link prediction unit 403, thereby performing estimation based on the trained model.

[0188] The basis generating unit 406 generates basis information indicating the basis of the estimation by the estimation unit 405. The basis generating unit 406 is similar to the basis generating unit 209 in the second exemplary embodiment, and therefore detailed description thereof will not be repeated.

[0189] The output unit 407 outputs various information generated by the recruitment support device 4, such as information indicating potential assignment destinations estimated by the estimation unit 405. As with the output unit 210 of the second exemplary embodiment, the output destination of the information output by the output unit 407 is not particularly limited.

[0190] As described above, the recruitment support device 4 is equipped with a link prediction unit 403 that calculates the probability that a node showing a specified characteristic will be linked to a node included in the target personnel graph by link prediction to predict relationships between nodes that are not connected by links in the target personnel graph and the target personnel graph, using a receiving organization graph that includes receiving organizations that may accept the target personnel, multiple nodes related to the skills or work history of each of multiple individuals, and links indicating the relationships between the nodes, and a target personnel graph that includes multiple nodes related to the target personnel, and the estimation unit 405 estimates candidate assignment destinations for the target personnel that are suitable for the request based on the probability calculated by the link prediction unit 403.

[0191] According to the above configuration, response information is generated based on the probability that a node showing a predetermined characteristic is linked to a node included in the target talent graph. The probability that a node showing a predetermined characteristic is linked to a node included in the target talent graph indicates the possibility that the target talent has the predetermined characteristic. Therefore, according to the above configuration, it is possible to provide information useful for personnel support, such as what characteristics the target talent is likely to have.

[0192] (Processing flow) Next, the flow of the process (recruitment support method) executed by the recruitment support device 4 will be described with reference to Fig. 17. Fig. 17 is a flow diagram showing the flow of the process executed by the recruitment support device 4.

[0193] In S401, the reception unit 401 receives a request regarding the assignment of a target human resource. Here, the request includes information about the target human resource to which the user wants to determine the assignment. As an example, the request includes, but is not limited to, the target human resource's name (or person ID), age, personality, desired job type or desired assignment destination, and characteristics.

[0194] In S402, the graph generating unit 402 generates a target talent graph based on the information input in S401. For example, when the characteristics of the target talent are received in S401, the graph generating unit 402 may generate a target talent graph including each node and link indicating the characteristics.

[0195] In S403, the link prediction unit 403 uses the target talent graph generated in S402, which includes multiple nodes related to the target talent, and the recipient graph, to perform link prediction to predict the relationship between nodes that are not connected by links in the target talent graph and the recipient graph, thereby calculating the probability that a node exhibiting a specified characteristic will be linked to a node included in the target talent graph.

[0196] In S406, the estimation unit 405 estimates potential affiliations of the target talent that match the request received in S401. Specifically, the estimation unit 405 estimates potential affiliations of the target talent that match the request based on the probability calculated in S403. For example, if the probability that the target talent will "contribute to a new business" is calculated in S403 and the probability is equal to or greater than a threshold, the estimation unit 405 may estimate affiliations related to new businesses as potential affiliations of the target talent.

[0197] In S407, the basis generating unit 406 generates basis information indicating the basis of the estimation in S406. Specifically, the basis generating unit 406 may generate basis information including the similarity between the attributes of the target human resource and the attributes of the people belonging to the candidate destinations estimated by the estimation unit 405 in S406.

[0198] In S408, the output unit 407 outputs information indicating the assignment destination estimated in S406. At this time, the output unit 407 may also output the basis information generated in S407. This ends the process shown in FIG.

[0199] In addition, in S402, a target talent graph including the affiliations of the target talent may be generated. This allows the link prediction in S406 to predict the characteristics that the target talent will have when the target talent is assigned to the affiliations. In other words, it is possible to simulate the results of assigning the target talent to various affiliations. In this case, the affiliation with the highest probability of obtaining the desired characteristics may be estimated as the candidate affiliation for the target talent.

[0200] [Modifications] As described above, by using the target talent graph and the recipient graph, it is possible to predict the traits that may be linked to the corresponding talent graph by link prediction. In addition, the trait prediction of the target talent can also be performed by a method other than link prediction. This will be explained based on FIG. 18. FIG. 18 is a diagram explaining an example of predicting the traits of the target talent based on the feature amount calculated from the target talent graph and the recipient graph. FIG. 18 shows the existing employee graphs of existing employees (talent) A to C and the target talent graph of the target talent. Note that the nodes and links included in these graphs are omitted from the illustration.

[0201] Here, the feature values ​​for each node in the existing employee graph can be calculated by multiplying the feature values ​​of each node by a weight corresponding to the link connected to that node and adding them up. Therefore, if learning is performed to update the weights so that the calculated feature values ​​correspond to the characteristics of the existing employees, it becomes possible to predict the characteristics of the target personnel from the feature values ​​of the target personnel graph calculated by applying the weights.

[0202] For example, in the example in Figure 18, the feature amount calculated from the existing employee graph of talent A, who is known to be suitable for sales positions, is trained to fall within a range in the feature space that corresponds to the attribute "suitable for sales positions." Also, the feature amounts calculated from the existing employee graphs of talent B and talent C, who are known to be suitable for technical positions, are trained to fall within a range in the feature space that corresponds to the attribute "suitable for technical positions."

[0203] In this case, as shown in the figure, if the feature amount calculated from the target talent graph is within a range corresponding to the attribute of "suitable for sales positions," it can be predicted that the target talent has the attribute of "suitable for sales positions." Such an attribute prediction method can be applied as an alternative method to the attribute prediction methods in each of the exemplary embodiments described above.

[0204] [Software implementation example] Some or all of the functions of the hiring support devices 1, 2, and 4 (hereinafter referred to as "hiring support device 1, etc.") may be realized by hardware such as an integrated circuit (IC chip), or may be realized by software.

[0205] In the latter case, the hiring support device 1 etc. is realized, for example, by a computer that executes instructions of a program, which is software that realizes each function. An example of such a computer (hereinafter, referred to as computer C) is shown in FIG. 19. The computer C has at least one processor C1 and at least one memory C2. The memory C2 stores a program P for operating the computer C as the hiring support device 1 etc. In the computer C, the processor C1 reads and executes the program P from the memory C2, thereby realizing each function of the hiring support device 1 etc.

[0206] The processor C1 may be, for example, a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a micro processing unit (MPU), a floating point number processing unit (FPU), a physics processing unit (PPU), a microcontroller, or a combination of these. The memory C2 may be, for example, a flash memory, a hard disk drive (HDD), a solid state drive (SSD), or a combination of these.

[0207] The computer C may further include a RAM (Random Access Memory) for expanding the program P during execution and for temporarily storing various data. The computer C may further include a communication interface for transmitting and receiving data to and from other devices. The computer C may further include an input / output interface for connecting input / output devices such as a keyboard, a mouse, a display, and a printer.

[0208] Furthermore, the program P can be recorded on a non-transitory tangible recording medium M that can be read by the computer C. Such a recording medium M can be, for example, a tape, a disk, a card, a semiconductor memory, or a programmable logic circuit. The computer C can obtain the program P via such a recording medium M. Furthermore, the program P can be transmitted via a transmission medium. Such a transmission medium can be, for example, a communication network or broadcast waves. The computer C can also obtain the program P via such a transmission medium.

[0209] [Additional Note 1] The present invention is not limited to the above-described embodiment, and various modifications are possible within the scope of the claims. For example, embodiments obtained by appropriately combining the technical means disclosed in the above-described embodiment are also included in the technical scope of the present invention.

[0210] [Additional Note 2] Some or all of the above-described embodiments can be described as follows. However, the present invention is not limited to the aspects described below.

[0211] (Appendix 1) A recruitment support device comprising: a receiving means for receiving a request regarding the assignment of a target personnel; an estimation means for estimating candidate assignment destinations for the target personnel that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of individuals and the affiliations of each of the plurality of individuals; and an output means for outputting information indicating the candidate assignment destinations for the target personnel estimated by the estimation means.

[0212] According to the above configuration, a request regarding a destination for assigning a target human resource is received. Then, using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of people and the affiliations of each of the plurality of people, candidate destinations for the target human resource that match the request are estimated. This makes it possible to estimate candidate destinations for the target human resource that match the request, taking into account various information regarding the hosting destination. Therefore, according to the above configuration, it is possible to provide suitable personnel support for the target human resource, taking into account various information regarding the hosting destination.

[0213] (Appendix 2) A recruitment support device as described in Appendix 1, further comprising a basis information generation means for generating basis information including a degree of similarity between the attributes of the target personnel and the attributes of people belonging to the candidate assignment destinations estimated by the estimation means as a basis for the estimation, and the output means further outputs the basis information.

[0214] According to the above configuration, evidence information is generated that includes the degree of similarity between the attributes of the target personnel and the attributes of the people who belong to the estimated candidate destinations. This allows the user to refer to the candidate destinations based on the evidence. Since ensuring transparency is particularly important in personnel management, the ability to generate evidence information is a major advantage.

[0215] (Appendix 3) The recruitment support device described in Appendix 1 or 2, wherein the trained model is an acceptance graph including multiple nodes related to acceptance destinations that may accept the target human resources, the skills or work history of each of the multiple individuals, and links indicating the relationships between the nodes.

[0216] According to the above configuration, the talent nodes in the receiving organization graph that are linked to the nodes included in the target talent graph are predicted, and potential assignment destinations are estimated based on the predicted talent nodes. Since how the target talent is related to which talent in the receiving organization and how the target talent is related is useful in the personnel management of the target talent, according to the above configuration, personnel support that takes into account the talent in the receiving organization that is related to the target talent is realized.

[0217] (Appendix 4) The recruitment support device described in Appendix 3 is equipped with a link prediction means for predicting, by link prediction using a target talent graph including a plurality of nodes related to the target talent and the receiving destination graph, relationships between nodes not connected by links in the target talent graph and the receiving destination graph, predicting talent nodes that link to nodes included in the graph of the target talent from among talent nodes included in the receiving destination graph that indicate talent belonging to the receiving destination, and the estimation means for estimating candidate assignment destinations for the target talent that are suitable for the request based on the talent nodes predicted by the link prediction means.

[0218] According to the above configuration, the talent nodes in the receiving organization graph that are linked to the nodes included in the target talent graph are predicted, and potential assignment destinations are estimated based on the predicted talent nodes. Since how the target talent is related to which talent in the receiving organization and how the target talent is related is useful in the personnel management of the target talent, according to the above configuration, personnel support that takes into account the talent in the receiving organization that is related to the target talent is realized.

[0219] (Appendix 5) The recruitment support device of claim 4, wherein the target talent graph includes a plurality of nodes relating to the skills or work history of the target talent and links indicating relationships between the nodes.

[0220] According to the above configuration, it is possible to estimate potential departments by taking into consideration not only the skills and work history of the target personnel but also the relationships between them.

[0221] (Appendix 6) The recruitment support device described in Appendix 4 or 5, wherein the link prediction means predicts similar talent who are similar to the target talent among the talents belonging to the receiving organization by the link prediction, and includes an identification means for identifying compatible talent from among the talents belonging to the receiving organization, who are talents whose nodes and links included in the receiving organization graph indicate that they are compatible with the similar talent.

[0222] According to the above configuration, a talent node indicating a talent similar to the target talent is predicted, and a compatible talent is identified whose nodes and links included in the hosting organization graph indicate a good compatibility with the similar talent. A compatible talent who is compatible with a similar talent is likely to be compatible with the target talent. In other words, according to the above configuration, it is possible to identify a compatible talent who is likely to be compatible with the target talent from among the talent belonging to the hosting organization. Therefore, according to the above configuration, it is possible to provide useful information for judging the compatibility between the target talent and the hosting organization.

[0223] (Appendix 7) The recruitment support device described in Appendix 6, further comprising a compatibility determination means for determining the compatibility between the target talent and the receiving organization based on a similarity indicating the degree of similarity between each talent belonging to the receiving organization and the compatible talent.

[0224] A host organization that has personnel similar to the compatible personnel is likely to be compatible with the target personnel. Therefore, according to the above configuration, the compatibility between the target personnel and the host organization is determined based on the similarity indicating the degree of similarity between each of the personnel belonging to the host organization and the compatible personnel. This makes it possible to accurately determine the compatibility between the target personnel and the host organization.

[0225] (Appendix 8) The recruitment support device described in Appendix 6, further comprising a compatibility determination means for identifying a department to which the compatible talent belongs among a plurality of departments included in the receiving company, and determining the compatibility between the department and the target talent based on a similarity indicating the degree of similarity between each talent belonging to the department and the compatible talent.

[0226] The department to which the compatible talent belongs is likely to be compatible with the target talent, and if the department has talent similar to the compatible talent, the compatibility is even more likely. Therefore, with the above configuration, the department to which the compatible talent belongs is identified, and the compatibility between the department and the target talent is determined based on the similarity indicating the degree of similarity between each talent belonging to the department and the compatible talent. This makes it possible to accurately determine the compatibility between the target talent and the department where the target talent will be accepted.

[0227] (Appendix 9) The recruitment support device described in Appendix 8, wherein the compatibility determination means determines compatibility between each of the multiple target human resources and each of the multiple departments, and includes a recommendation means for determining a department to recommend to each of the multiple target human resources as a destination for the target human resources based on the compatibility determination result.

[0228] According to the above configuration, the compatibility between each of the multiple target human resources and each of the multiple departments is determined, and a department to be recommended as a host for each of the multiple target human resources is determined based on the result of the determination. This makes it possible to recommend a host department that takes into account the compatibility between each department and each target human resource.

[0229] (Appendix 10) A recruitment support device as described in Appendix 1 or 2, comprising a link prediction means for identifying personnel or departments of the receiving organization having a predetermined relationship with the target personnel by link prediction for predicting relationships between nodes that are not connected by links in the target personnel graph and the receiving organization graph, using a receiving organization graph including receiving organizations that may accept the target personnel, a plurality of nodes related to the skills or work history of each of the plurality of persons, and links indicating the relationships between the nodes, and a target personnel graph including a plurality of nodes related to the target personnel, and the estimation means for estimating candidate assignment destinations for the target personnel that are suitable for the request, based on the personnel or departments of the receiving organization identified by the link prediction means.

[0230] According to the above configuration, a person or department of the receiving company that has a predetermined relationship with the target person is identified, and a candidate for the target person that matches the request is estimated based on the person or department. Since information on the person or department that has a predetermined relationship with the target person is useful information for the target person's personnel affairs, according to the above configuration, personnel support for the target person can be performed accurately.

[0231] (Appendix 11) The recruitment support device according to claim 1 or 2, further comprising a link prediction means for calculating the probability that a node exhibiting a predetermined characteristic will be linked to a node included in the target personnel graph by link prediction for predicting relationships between nodes not connected by links in the target personnel graph and the target personnel graph, using a destination graph including destinations that may accept the target personnel, a plurality of nodes related to the skills or work history of each of the plurality of persons, and links indicating the relationships between the nodes, and a target personnel graph including a plurality of nodes related to the target personnel, and the estimation means for estimating candidate destinations for the target personnel that are suitable for the request based on the probability calculated by the link prediction means.

[0232] According to the above configuration, response information is generated based on the probability that a node showing a predetermined characteristic is linked to a node included in the target talent graph. The probability that a node showing a predetermined characteristic is linked to a node included in the target talent graph indicates the possibility that the target talent has the predetermined characteristic. Therefore, according to the above configuration, it is possible to provide information useful for personnel support, such as what characteristics the target talent is likely to have.

[0233] (Appendix 12) A recruitment support method in which a computer receives a request regarding a workplace for a target personnel, estimates potential workplaces for the target personnel that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of individuals and the affiliations of each of the plurality of individuals, and outputs information indicating the estimated potential workplaces for the target personnel.

[0234] According to the above configuration, the same effect as that obtained by supplementary note 1 can be obtained.

[0235] (Appendix 13) A recruitment support program that causes a computer to execute the following processes: receiving a request regarding a department for assigning a target personnel; predicting candidate departments for the target personnel that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of individuals and the affiliations of each of the plurality of individuals; and outputting information indicating the predicted candidate departments for the target personnel.

[0236] [Additional Note 3] A part or all of the above-described embodiments can be further expressed as follows.

[0237] A recruitment support device having at least one processor that executes a reception process for receiving a request regarding the assignment of a target personnel, an estimation process for estimating candidate assignment destinations for the target personnel that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of individuals and the affiliations of each of the plurality of individuals, and an output process for outputting information indicating the estimated candidate assignment destinations for the target personnel.

[0238] The hiring support device may further include a memory that stores a program for causing the processor to execute the reception process, the estimation process, and the output process. The program may be recorded in a computer-readable, non-transitory, tangible recording medium. [Explanation of symbols]

[0239] 1,2,4…Recruitment support device 11…Reception 12…Estimation part 13...Output section 201,401…Reception 202,402...Graph generation section 203,403…Link prediction section 204…Specific section 205…Compatibility Judgment Section 206…Recommendation section 207…Learning Department 208,405…Estimation part 209,406…Evidence generation section 210,407…Output section

Claims

1. A receiving means for receiving a request regarding the assignment of the target human resource; An estimation means for estimating candidate destinations for the target human resources that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of persons and the affiliation of each of the plurality of persons; an output means for outputting information indicating potential assignment destinations for the target human resource estimated by the estimation means; Equipped with The learned model is a destination graph that includes multiple nodes related to destinations that may accept the target personnel, the skills or work history of each of the multiple individuals, and links indicating the relationships between the nodes.

2. a basis information generating means for generating basis information including a degree of similarity between an attribute of the target human resource and an attribute of a person belonging to the candidate destination estimated by the estimation means as a basis for the estimation, The output means further outputs the grounds information. The recruitment support device according to claim 1.

3. a link prediction means for predicting a relationship between nodes not connected by a link in the target talent graph and the recipient graph using a target talent graph including a plurality of nodes related to the target talent and the recipient graph, and predicting a talent node that links to a node included in the target talent graph from among talent nodes included in the recipient graph that indicate talent belonging to the recipient, 2. The recruitment support device according to claim 1, wherein said estimation means estimates potential assignment destinations for said target talent that matches said request based on said talent nodes predicted by said link prediction means.

4. The recruitment support device according to claim 3 , wherein the target talent graph includes a plurality of nodes related to the skills or work history of the target talent, and links indicating relationships between the nodes.

5. The link prediction means predicts, by the link prediction, human resources belonging to the receiving organization who are similar to the target human resource, and The recruitment support device of claim 3 or 4, further comprising an identification means for identifying compatible personnel from among the personnel belonging to the receiving company, the personnel being personnel whose nodes and links included in the receiving company graph indicate good compatibility with the similar personnel.

6. The recruitment support device of claim 5, further comprising a compatibility determination means for determining the compatibility between the target talent and the receiving organization based on a similarity indicating the degree of similarity between each talent belonging to the receiving organization and the compatible talent.

7. The recruitment support device of claim 5, further comprising a compatibility determination means for identifying a department to which the compatible talent belongs among a plurality of departments included in the receiving company, and determining the compatibility between the department and the target talent based on a similarity indicating the degree of similarity between each talent belonging to the department and the compatible talent.

8. The compatibility determination means determines compatibility between each of the plurality of target personnel and each of the plurality of departments, The recruitment support device according to claim 7, further comprising a recommendation means for determining, for each of the plurality of target human resources, a department to be recommended as a place of employment for the target human resources, based on the compatibility determination result.

9. A reception means for receiving a request regarding the assignment of the target human resource; An estimation means for estimating candidate destinations for the target human resources that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of persons and the affiliation of each of the plurality of persons; an output means for outputting information indicating potential assignment destinations for the target human resource estimated by the estimation means; a link prediction means for using a recipient graph including a plurality of nodes related to potential recipients of the target human resources, the plurality of nodes related to the skills or work history of each of the plurality of persons, and links showing the relationships between the nodes, and a target human resources graph including a plurality of nodes related to the target human resources, to perform link prediction for predicting relationships between nodes that are not connected by links in the target human resources graph and the recipient graph, and for identifying human resources or departments of the recipients who have a predetermined relationship with the target human resources; The estimation means estimates potential assignment destinations for the target human resource that match the request based on the human resource or department of the accepting organization identified by the link prediction means.

10. A reception means for receiving a request regarding the assignment of the target human resource; An estimation means for estimating candidate destinations for the target human resources that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of persons and the affiliation of each of the plurality of persons; an output means for outputting information indicating potential assignment destinations for the target human resource estimated by the estimation means; a link prediction means for calculating the probability that a node showing a predetermined characteristic will be linked to a node included in the target human resources graph by link prediction for predicting relationships between nodes not connected by links in the target human resources graph and the target human resources graph, using a target human resources graph including a target human resources graph including a target human resources graph including a target human resources graph including a target human resources graph including a target human resources graph including a target human resources graph including a target human resources graph, The estimation means estimates potential assignment destinations for the target human resource that match the request based on the probability calculated by the link prediction means.

11. The computer Accept requests regarding the placement of the target personnel, Using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of people and the affiliation of each of the plurality of people, estimate potential assignment destinations for the target human resource that match the request; A recruitment support method for outputting information indicating potential assignment destinations for the estimated target personnel, The trained model is a recipient graph including a plurality of nodes relating to recipients that may accept the target human resources, the skills or work history of each of the plurality of persons, and links indicating relationships between the nodes. Recruitment support methods.

12. For computers, A process for receiving requests regarding the placement of the target personnel; A process of estimating candidate destinations for the target human resources that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of people and the affiliation of each of the plurality of people; A process of outputting information indicating potential assignment destinations for the estimated target human resources; A recruitment support program that implements the following: A recruitment support program, wherein the trained model is an acceptance graph including multiple nodes relating to potential acceptance destinations for the target personnel, the skills or work history of each of the multiple individuals, and links indicating the relationships between the nodes.

13. A computer comprising: Accept requests regarding the placement of the target personnel, Using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of people and the affiliation of each of the plurality of people, estimate potential assignment destinations for the target human resource that match the request; Outputting information indicating potential assignment destinations for the estimated target personnel; A recruitment support method for identifying personnel or departments at the receiving organization having a predetermined relationship with the target personnel by link prediction for predicting relationships between nodes not connected by links in the target personnel graph and the receiving organization graph, using a receiving organization graph including a plurality of nodes related to the target personnel, the ... A recruitment support method in which, in predicting potential assignment destinations for the target personnel, the computer predicts potential assignment destinations for the target personnel that match the request based on the personnel or departments of the receiving company identified by link prediction.

14. For a computer, A process for receiving requests regarding the placement of the target personnel; A process of estimating candidate destinations for the target human resources that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of people and the affiliation of each of the plurality of people; A process of outputting information indicating potential assignment destinations for the estimated target human resources; a recipient graph including a plurality of nodes related to potential recipients of the target human resource, the plurality of nodes being related to the skills or work history of each of the plurality of persons, and links showing the relationships between the nodes; and a target human resource graph including a plurality of nodes related to the target human resource, by link prediction for predicting relationships between nodes not connected by links in the target human resource graph and the recipient graph, the recipient graph including: In the process of estimating potential assignment destinations for the target personnel, the recruitment support program causes the computer to estimate potential assignment destinations for the target personnel that match the request based on the personnel or departments of the receiving company identified by link prediction.

15. A computer comprising: Accept requests regarding the placement of the target personnel, Using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of people and the affiliation of each of the plurality of people, estimate potential assignment destinations for the target human resource that match the request; Outputting information indicating potential assignment destinations for the estimated target personnel; A recruitment support method using a destination graph including multiple nodes related to destinations that may accept the target personnel, the multiple nodes related to the skills or work histories of each of the multiple persons, and links showing relationships between the nodes, and a target personnel graph including multiple nodes related to the target personnel, the method calculates a probability that a node showing a predetermined characteristic will be linked to a node included in the target personnel graph by link prediction for predicting relationships between nodes that are not connected by links in the target personnel graph and the destination graph, A recruitment support method, in which, in predicting potential assignment destinations for the target personnel, the computer predicts potential assignment destinations for the target personnel that match the request based on the probability calculated by link prediction.

16. For a computer: A process for receiving requests regarding the placement of the target personnel; A process of estimating candidate destinations for the target human resources that match the request using a trained model that has learned the relationship between at least one of the skills and work history of each of a plurality of people and the affiliation of each of the plurality of people; A process of outputting information indicating potential assignment destinations for the estimated target human resources; a target personnel graph including a plurality of nodes related to the target personnel, the target personnel graph including a plurality of nodes related to the target personnel, and links indicating relationships between the nodes, and a target personnel graph including a plurality of nodes related to the target personnel, and a link prediction is performed to predict relationships between nodes not connected by links in the target personnel graph and the target personnel graph, the link prediction calculating a probability that a node showing a predetermined characteristic is linked to a node included in the target personnel graph, A recruitment support program in which, in the process of estimating potential assignment destinations for the target personnel, the computer is caused to estimate potential assignment destinations for the target personnel that match the request based on the probability calculated by link prediction.

Images