Information processing apparatus, information processing method, and program

Abstract

To allow for associating a more likely tag with a sentence.SOLUTION: A tag specification unit 101 identifies a skill of an employee from an inputted sentence and estimates at least one tag corresponding to the skill of the employee from a plurality of tags. A score calculation unit 102 calculates scores indicating likelihoods of the at least one estimated tag. An additional question presentation unit 103 presents an additional question sentence prompting a change of contents of the sentence if the scores calculated for the at least one estimated tag do not exceed a first threshold at all. A tagging unit 104 associates a tag which the score calculated for exceeds the first threshold, with the sentence if the tag which the score calculated for exceeds the first threshold out of the scores calculated for the at least one estimated tag.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

[0002] Patent document 1 proposes a suggestion device that generates multiple tags based on product information (product image and / or descriptive text information) of the product to be tagged, outputs a score indicating the appropriateness of each tag, and suggests tags with scores above a predetermined threshold as recommended tags. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2023-101947 Summary of the Invention [Problem to be solved by the invention]

[0004] In the above-mentioned suggestion device, if the score is below a predetermined threshold, no recommended tags are suggested, and if the predetermined threshold is lowered, tags with low reliability are suggested. Therefore, even if the suggestion device is applied to human resource management, there is a problem that accurate tagging cannot be performed if the description in the text (hereinafter referred to as career summary) in which an employee describes their work history, career history, qualifications (hereinafter collectively referred to as skills), etc. is vague.

[0005] Therefore, an object of the present invention is to make it possible to associate more likely tags with sentences. [Means for solving the problem]

[0006] The information processing device of the present invention is characterized by comprising: a tag identification unit that estimates an employee's skills from an input sentence and identifies at least one tag from among a plurality of tags that corresponds to the estimated employee's skills; a score calculation unit that calculates a score indicating the likelihood of the tag for the at least one identified tag; an additional question presentation unit that presents an additional question prompting the user to change the content of the sentence if none of the scores calculated for the at least one identified tag exceeds a first threshold; and a tagging unit that associates a tag with a score that exceeds the first threshold with the sentence if the score of at least one tag among the scores calculated for the at least one identified tag exceeds the first threshold.

[0007] The information processing method of the present invention is a control method by a control unit of an equipment, and is characterized by including the steps of: estimating an employee's skills from an input sentence; identifying at least one tag from a plurality of tags that corresponds to the estimated employee's skills; calculating a score indicating the likelihood of the tag for the identified at least one tag; presenting an additional question prompting the user to change the content of the sentence if none of the scores calculated for the identified at least one tag exceeds a first threshold; and associating a tag with a score that exceeds the first threshold with the sentence if the score of at least one tag calculated for the identified at least one tag exceeds the first threshold.

[0008] The program of the present invention is characterized in that it causes a computer of a device equipped with a control unit to realize the following: a tag identification function that estimates an employee's skills from an input sentence and identifies at least one tag from multiple tags that corresponds to the estimated employee's skills; a score calculation function that calculates a score indicating the likelihood of the tag for the at least one identified tag; an additional question presentation function that presents an additional question encouraging the user to change the content of the sentence if none of the scores calculated for the at least one identified tag exceed a first threshold; and a tagging function that associates a tag with a score that exceeds the first threshold with the sentence if the score of at least one tag among the scores calculated for the at least one identified tag exceeds the first threshold.

[0009] The program of the present invention is characterized in that it causes a computer of a device having a control unit to realize the following: a skill estimation function that analyzes an input sentence and estimates at least one skill of the employee from among multiple skills; a score calculation function that calculates a score indicating the accuracy of the skill for the at least one estimated skill; an additional question presentation function that presents an additional question encouraging the employee to change the content of the sentence if none of the scores calculated for the at least one estimated skill exceed a first threshold; and a skill assignment function that associates a skill with a score exceeding the first threshold with the sentence if the score of at least one skill among the scores calculated for the at least one estimated skill exceeds the first threshold. [Effects of the Invention]

[0010] According to this invention, it is possible to associate a more likely tag with a sentence. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a block diagram showing a configuration of an information processing device according to an embodiment of the present invention. [Figure 2]10 is a flowchart for explaining the operation (tagging process) of the information processing device according to the present embodiment. [Figure 3] 10 is a conceptual diagram for explaining conditions for presenting a follow-up question calculated by the information processing device according to the present embodiment. FIG. [Figure 4] 10A and 10B are schematic diagrams illustrating an example of a display of a follow-up question presented by the information processing device according to the present embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0012] An embodiment in which a device control device of the present invention is applied to an information processing device 1 shown in Fig. 1 will be described with reference to the drawings. As shown in Fig. 1, the information processing device (server) 1 has a control unit 10, an additional question DB (database) 20, and a history information DB 30. The control unit 10 is made up of a CPU (Central Processing Unit) and the like, and functions as a tag identification unit 101, a score calculation unit 102, an additional question presentation unit 103, and a tagging unit 104.

[0013] The tag identification unit 101 analyzes multiple sentences constituting the career summary (text) provided by the employee, estimates the skills possessed from the content of each sentence, and identifies at least one tag corresponding to the estimated skills. Furthermore, if it is estimated that multiple skills are described in a single sentence, multiple tags may be identified. The score calculation unit 102 calculates a score indicating the degree of certainty of each of the identified at least one tag. When the follow-up question presentation unit 103 determines that the career summary needs to be made more accurate based on the score of each tag and a predetermined threshold, it reads an appropriate follow-up question (text) from the follow-up question DB 20 and presents it to the employee (or their terminal). The tagging unit 104 finally associates the tag with the career summary and stores it in the history information DB 30. The follow-up question DB 20 stores multiple follow-up questions for making the career summary more accurate. The history information DB 30 stores tagged career summaries and tag scores for each employee.

[0014] The terminal 50 is used for an employee or the like to input (register) their career summary via, for example, a web browser or the like. The terminal 50 transmits the input career summary to the control unit 10 via the network 40. The network 40 consists of the Internet, an intranet, or the like.

[0015] First, an employee or the like inputs their career summary from the terminal 50. The terminal 50 transmits the career summary to the control unit 10 via the network 40. The control unit 10 stores the career summary from the terminal 50 in the history information DB 30. Next, at a predetermined timing (each time a career summary is registered, monthly, semi-annually, annually, etc.), the control unit 10 executes a tagging process according to the flowchart shown in FIG. 2. The tag identification unit 101 fetches the career summary provided by the employee from the history information DB 30 (step S10), analyzes the content described in the career summary, estimates what skills there are from the analysis result, and identifies at least one or more tags corresponding to the estimated skills (step S12). Note that a process of identifying tags may be performed according to an operation on the screen by an employee or the like (for example, a click operation on a specific button on the screen).

[0016] Next, the score calculation unit 102 calculates a score indicating the probability of at least one or more identified tags (step S14). The additional question prompt unit 103 determines whether the score of any tag exceeds the threshold Th1 (step S16). If the scores of all tags do not exceed the threshold Th1 (NO in step S16), it determines whether the scores of all tags are below the threshold Th2 (<Th1) (step S18).

[0017] The above-mentioned thresholds Th1 and Th2 will now be described with reference to FIG. 3. FIGS. 3(a) and 3(b) show examples of scores calculated by the score calculation unit 102. In this embodiment, thresholds Th1 and Th2 are set for the calculated scores. The threshold Th1 is the average score calculated based on the result of comparing the tag scores obtained from previously collected history information with the actual work capabilities (results), or the average score calculated based on accuracy information indicating whether the tags derived by the control unit 10 match the skills possessed by the employee. The threshold Th2 is a value that is lower than the threshold Th1 but close to it, such as 90% or 80%. In other words, a tag that is lower than the threshold Th1 but equal to or greater than the threshold Th2 is a tag that could have a higher score, i.e., a higher likelihood, if the content of the career summary were rewritten in more detail.

[0018] In the example shown in Fig. 3(a), only tag T2 is equal to or greater than threshold Th1, tag T1 is equal to or greater than threshold Th2, while tags T3 to T5 are below threshold Th2. In the example shown in Fig. 3(b), no tag is equal to or greater than threshold Th1, only tag T2 is equal to or greater than threshold Th2, and tags T1, T3 to T5 are below threshold Th2.

[0019] If the scores of all tags are below the threshold Th2 (YES in step S18), the additional question presenting unit 103 determines that the scoring was not performed with sufficient accuracy, and selects an additional question (text) from the additional question DB 20 that is not dependent on a specific tag and presents it to the employee's terminal 50 (step S20). In this case, since the career summary description is not specific enough, additional questions such as "Please be more specific," "What kind of products have you been involved with?", and "Please provide a description of specific projects and tasks" are presented. In accordance with instructions from the control unit 10, the terminal 50 highlights the relevant portion (a sentence, phrase, or word) of the career summary 60 displayed on the screen, as shown in FIG. 4. In other words, in accordance with instructions from the control unit 10, the terminal 50 displays the relevant portion of the career summary 60 displayed on the screen in a display mode different from the other portions. Note that the change in display mode is not limited to highlighting. For example, the color, thickness, or size of the text in the relevant portion may be changed. Furthermore, in accordance with an instruction from the control unit 10, the terminal 50 displays an additional question 61 in a speech bubble near the highlighted portion (below the highlighted portion in the example shown in FIG. 4 ). The employee inputs an answer (more detailed explanation) to the additional question 61 displayed on the terminal 50. The terminal 50 transmits the answer to the additional question 61 to the control unit 10. Thereafter, the control unit 10 returns to step S10, retrieves the employee's answer (difference) to the additional question 61 again, and repeats the above-described process. By performing the above-described process again on the employee's answer (difference), it is possible to reduce the processing time required for skill estimation, tag identification, score calculation, etc., compared to a case in which the entire text including the answer (difference) is analyzed again to estimate skills and identify tags. Note that when reducing the processing time is not a consideration, the entire text including the answer (difference) may be retrieved and analyzed again to estimate skills, identify tags, calculate scores, etc.

[0020] On the other hand, as shown in FIG. 3(b), if the scores of all tags are not equal to or greater than the threshold Th1 (NO in step S16) but any tag score is equal to or greater than the threshold Th2 (NO in step S18), the follow-up question presenting unit 103 selects at least one follow-up question related to the tag equal to or greater than the threshold Th2 from the follow-up question DB 20 (tag T2 in the example shown in FIG. 3(b)), and presents it to the terminal 50 (step S22). In this case, the follow-up question may be presented based on example sentences of tags that may be equal to or greater than the threshold Th1, or may be presented with a plurality of tags that may be equal to or greater than the threshold Th1 and allow the user to select one. For example, the follow-up question may be presented as follows: "If there is a tag below that matches your content, please select it. Tag A / Tag B / Tag D." The employee inputs a more specific answer to the follow-up question for the tag (tag T2 in FIG. 3(b)) that is equal to or greater than the threshold Th2, which is presented on the terminal 50. The terminal 50 transmits the answer to the follow-up question to the control unit 10. Thereafter, the control unit 10 returns to step S10, acquires the answer (difference) to which the employee's more specific content has been added in response to the additional question, and repeats the above-described process.

[0021] On the other hand, as shown in FIG. 3(a), if the score of any tag is equal to or greater than the threshold value Th1 (YES in step S16), the score calculation unit 102 calculates the score of the tag equal to or greater than the threshold value Th1. * (Score * The tags are sorted and output in the order of score / threshold value Th1 (step S24). * The reason for deriving is to take into account how much the score deviates from the threshold Th1 by calibrating (normalizing) it with the threshold Th1. If the score deviates significantly from the threshold Th1, * is larger (>1), and if the score is not too far from the threshold Th1, * is smaller (≒1). Score *A high score indicates that the tag (skill) is specialized. In step S16, if even one tag has a score equal to or greater than the threshold Th1, the tags are sorted and output. However, this is not limiting, and the above process may be repeated until a pre-specified number of N tags (e.g., 10) have scores equal to or greater than the threshold Th1. The tagging unit 104 associates the sorted and output tags with scores equal to or greater than the threshold Th1, i.e., tags with high likelihood, with the corresponding part of the career summary (sentence, phrase, or word) and stores them in the history information DB 30 (step S26). The process then ends. The control unit 10 may present the results, such as the sentences described as the career summary, the skills estimated from the sentences, the tags identified from the skills, and their scores, stored in the history information DB 30, to the employee's terminal 50. The employee may check the results and provide feedback to the control unit 10 as to whether they match their actual skills.

[0022] In the above-described embodiment, when skills are estimated from the contents of a career summary, at least one skill may be estimated for each of the multiple sentences that make up the career summary. However, since there may be sentences in the career summary that do not describe the career, the number of times skills are estimated for one sentence may be limited, and if a tag with sufficient certainty (score) cannot be identified even after estimation a predetermined number of times, skill estimation may be skipped, or the length (number of characters) of one sentence may be limited, and if the number of characters is less than a predetermined number, skill estimation may be skipped.

[0023] According to the above-described embodiment, if the tag score is not equal to or greater than a predetermined threshold Th1, an additional question that is not dependent on the tag is presented to enable scoring with a higher degree of certainty. Furthermore, if none of the tags have a score equal to or greater than the threshold Th1 but one of the tags has a score greater than the threshold Th2, an additional question related to the tag that exceeds the threshold Th2 is presented. In this way, if a score equal to or greater than the threshold Th1 cannot be obtained, an additional question is presented to the employee, and the score is recalculated based on the answer to the additional question. This allows more likely tags to be associated with the text written as a career summary.

[0024] In the above-described embodiment, an example in which the career summary is input via a web browser or the like has been described, but the method of inputting text is not limited to this. For example, text may be input by uploading a file containing the career summary via a web browser.

[0025] Furthermore, in the above-described embodiment, an employee's skills are estimated from the input text (career summary), a tag corresponding to the estimated employee's skills is identified from among multiple tags, a score is calculated for the identified tag, and additional questions are presented as necessary based on the score, thereby associating a more accurate tag with the text. However, this is not limited to this, and the input text may be analyzed, an employee's skills may be estimated from among multiple skills based on the analysis results (skill estimation unit (not shown)), a score is calculated for the estimated skills, and additional questions are presented as necessary based on the score, thereby associating a more accurate skill with the text (skill assignment unit (not shown)).

[0026] Furthermore, it goes without saying that the detailed configuration and detailed operation of each component of the information processing device 1 in the above embodiment can be appropriately modified without departing from the spirit of the present invention. Although the embodiments of the present invention have been described, the scope of the present invention is not limited to the above-described embodiments, but includes the scope of the invention set forth in the claims and their equivalents. [Explanation of symbols]

[0027] 1...information processing device, 10...control unit, 101...tag identification unit, 102...score calculation unit, 103...additional question presentation unit, 104...tagging unit

Claims

1. a tag identification unit that estimates the employee's skills from an input sentence and identifies at least one tag corresponding to the estimated employee's skills from among a plurality of tags; a score calculation unit that calculates a score indicating the likelihood of the tag for the at least one identified tag; an additional question presenting unit that presents an additional question prompting the user to change the content of the text when none of the scores calculated for the at least one identified tag exceeds a first threshold; a tagging unit that, when the score of at least one tag among the scores calculated for the at least one identified tag exceeds the first threshold, associates the tag having the score exceeding the first threshold with the sentence; An information processing device comprising:

2. the additional question presenting unit presents an additional question prompting the user to describe the sentence more specifically, or an additional question related to a tag having a score that does not exceed the first threshold; 2. The information processing apparatus according to claim 1, wherein:

3. When presenting an additional question related to a tag having a score that does not exceed the first threshold, the additional question presenting unit presents the additional question in a manner that the relevant part of the sentence is displayed differently from other parts.

3. The information processing apparatus according to claim 2, wherein:

4. the additional question presenting unit presents an additional question prompting the user to describe the sentence more specifically when the scores of all tags among the at least one identified tag do not exceed the first threshold and do not exceed a second threshold that is smaller than the first threshold.

3. The information processing apparatus according to claim 2, wherein:

5. the additional question presenting unit presents, as the additional question, an example sentence related to the specific tag when there is a specific tag among the at least one identified tag whose score calculated by the score calculation unit does not exceed the first threshold value and is equal to or greater than the second threshold value; 5. The information processing apparatus according to claim 4,

6. the additional question presenting unit, when there is a specific tag among the at least one identified tag whose score calculated by the score calculation unit does not exceed the first threshold value and is equal to or greater than the second threshold value, presents at least one or more candidate tags related to the specific tag in a selectable manner; 5. The information processing apparatus according to claim 4,

7. the score calculation unit outputs, from the at least one identified tag, tags whose scores calculated by the score calculation unit are equal to or greater than the first threshold in order of score; 2. The information processing apparatus according to claim 1, wherein:

8. The text is a career summary.

2. The information processing apparatus according to claim 1, wherein:

9. the additional question presentation unit repeats presenting an additional question a predetermined number of times until the score calculated for the at least one identified tag exceeds the first threshold; the tag identification unit and the score calculation unit identify tags and calculate scores for additional sentences in response to the additional question sentence; 2. The information processing apparatus according to claim 1, wherein:

10. A control method by a control unit of a device, A step of estimating employee skills from an input sentence and identifying at least one tag corresponding to the estimated employee skills from among a plurality of tags; calculating a score indicating the likelihood of the tag for the at least one identified tag; If none of the scores calculated for the at least one identified tag exceeds a first threshold, presenting an additional question prompting the user to change the content of the sentence; a step of associating a tag having a score exceeding the first threshold with the sentence when the score of at least one tag among the scores calculated for the at least one identified tag exceeds the first threshold; An information processing method comprising:

11. The computer of the device that has the control unit a tag identification function that estimates an employee's skills from an input sentence and identifies at least one tag corresponding to the estimated employee skills from among a plurality of tags; a score calculation function for calculating a score indicating the likelihood of the tag for the at least one identified tag; a function of presenting an additional question that prompts the user to change the content of the text when none of the scores calculated for the at least one identified tag exceeds a first threshold; a tagging function that, when the score of at least one tag among the scores calculated for the at least one identified tag exceeds the first threshold, associates the tag having the score exceeding the first threshold with the sentence; A program characterized by realizing the above.

12. The computer of the device that has the control unit A skill estimation function that analyzes input text and estimates at least one of an employee's skills from multiple skills; a score calculation function for calculating a score indicating the likelihood of the estimated at least one skill; an additional question suggesting function that suggests an additional question prompting the user to change the content of the sentence when none of the scores calculated for the estimated at least one skill exceeds a first threshold; a skill assignment function that, when the score of at least one skill among the scores calculated for the estimated at least one skill exceeds the first threshold, associates the skill having a score exceeding the first threshold with the sentence; A program characterized by realizing the above.

Images