JP-7855951-B2 - Information processing device and information processing program

JP7855951B2JP 7855951 B2JP7855951 B2JP 7855951B2JP-7855951-B2

Inventors

姫野拓未

Assignees

富士フイルムビジネスイノベーション株式会社

Dates

Publication Date: 20260511
Application Date: 20220705

Claims (6)

Equipped with a processor, The aforementioned processor, Multiple responses to a question containing multiple intentions are classified into multiple groups based on a first similarity score indicating the degree of similarity between each response. An information processing device that, for each group, presents an answer from among the multiple answers that corresponds to each intention contained in the question, using the question, a segmented representation of the question obtained by combining two or more phrases of the question, and a second similarity score indicating the similarity between each answer included in the group.
The information processing apparatus according to claim 1, wherein the processor presents the classification results of the plurality of answers along with the answers corresponding to each of the intentions contained in the question.
The information processing apparatus according to claim 2, wherein the processor further presents the number of intentions contained in the question.
The information processing apparatus according to claim 1, wherein the processor identifies for each group a first answer which is the answer to which the second similarity of the segmented expression of the question sentence corresponds to the answer that has the largest difference in the decrease of the second similarity of the question sentence, or a second answer which is the answer to which the highest second similarity among the second similarities of each segmented expression of the question sentence corresponds, and presents them as answers corresponding to each intention contained in the question sentence.
The information processing apparatus according to claim 4, wherein the processor presents the second answer as an answer corresponding to each intention contained in the question, if the redundancy of the question, which indicates that the more redundant the expression, is greater than or equal to a predetermined value.
On the computer, Multiple responses to a question containing multiple intentions are classified into multiple groups based on a first similarity score indicating the degree of similarity between each response. An information processing program for performing a process to present, for each group, an answer corresponding to each intention contained in the question, from among the multiple answers, using a second similarity score that indicates the similarity between the question and each answer included in the group, obtained by combining the question and two or more phrases of the question.

Description

This disclosure relates to an information processing device and an information processing program. Patent Document 1 discloses a question-answering system comprising: a storage unit for storing question-answer patterns, which are pairs of question sentences and corresponding answers; a reception unit for receiving the question sentences from a user; a calculation unit for calculating the similarity between the question sentences received by the reception unit and the question sentences stored in the storage unit; an acquisition unit for acquiring the question-answer patterns from the storage unit based on the similarity calculated by the calculation unit; and a presentation unit for presenting the question-answer patterns acquired by the acquisition unit to the user. The acquisition unit, when multiple question-answer patterns containing the same answer exist, acquires only the question-answer pattern with the highest similarity from the storage unit. Furthermore, when multiple question-answer patterns containing the same answer exist, the acquisition unit applies a positive correction to the similarity of the question-answer pattern with the highest similarity, and acquires the question-answer patterns from the storage unit in order of the corrected similarity. Patent Document 2 describes a question answering program that has a comment text storage unit that stores a large number of comment texts and an answer storage unit that stores a large number of answers, and causes the computer to function to extract answers to user questions, comprising: a question text input means for inputting a question text; a question keyword extraction means for extracting multiple question keywords contained in the question text; a comment text search means for searching for comment texts containing the question keywords using the comment text storage unit; a topic classification means for classifying the searched multiple comment texts into multiple topic groups based on the distribution of predicate-argument structures by predicate-argument structure analysis; a group of comment texts contained in each topic group; and each answer. A question-answering program is disclosed that uses a computer as an answer output means to explicitly display the answers detected by the answer detection means, along with one or more corresponding sentences based on the differential predicate structures. This program includes: an answer detection means that calculates the similarity between a given sentence and a given sentence, and associates answers with each topic group whose similarity exceeds a predetermined threshold; a representative predicate structure extraction means that extracts a representative predicate structure characteristic of the topic group from among the predicate structure structures included in the associated answers; a differential predicate structure extraction means that extracts a representative predicate structure that appears only in that topic group, as a differential predicate structure; and an answer output means that explicitly displays the answers detected by the answer detection means, along with one or more corresponding sentences based on the differential predicate structures. Patent Document 3 discloses an automated response system that includes a response inference unit that uses a learning model to infer a group of responses corresponding to a query in a ranked format, based on feature quantities for each combination of a query and each of a group of responses. The response inference unit then notifies the user of at least a portion of the ranked group of responses inferred by the learning model. Patent No. 6770283Patent No. 6007088Japanese Patent Publication No. 2021-56580 This figure shows an example of the functional configuration of an information processing device.This figure shows an example of the main components of the electrical system of an information processing device.This is a flowchart illustrating an example of the response search process.This figure shows examples of how answers to questions can be categorized.This figure shows an example of how a question can be divided.This figure shows an example of how to calculate the similarity between the answer to a question and the question itself and its segmented expressions.This diagram shows examples of selections where you choose an answer according to the selection rules.This diagram shows examples of choices where you select an answer according to other selection rules.This figure shows an example of how to display an answer to a question.This figure shows an example of a display that shows the number of intentions included in a question.This figure shows an example of a classification screen that displays the results of classifying answers to a question. The embodiments of the disclosure will be described below with reference to the drawings. The same reference numerals are used throughout the drawings for the same components and process