Title:
自然言語テキストからアイディアを自動合成する方法、コンピュータプログラム及び装置
Document Type and Number:
Japanese Patent JP7239959
Kind Code:
B2
Abstract:
PROBLEM TO BE SOLVED: To provide a method, a device, and a computer program capable of fully automatically synthesizing a completely novel idea responding to a request from a user from a completely free natural language text in multiple fields relating to the request from the user.SOLUTION: A computer program that contains a part of or all of a PROLOG function allows the utilization of a literal that has a variable embedded in a natural language in a logical program, enables pattern matching of the natural language in conjunction with a recursive syllogism, places, includes, in the logical program, a rule (hereinafter, "synthesis rule") that does not contain all kinds of variable contained in a left-side head where the variable contained in each condition of a right-side body of a rule which is the rule of PROLOG is contained, causes different literals that are respective conditions of the synthesis rule (however, the scope of the contained variable is common within the synthesis rule) to individually match multiple texts in multiple fields, and enables coupling of the concepts of the multiple fields by constructing a new concept in the literal of the head of the synthesis rule.SELECTED DRAWING: Figure 1
More Like This:
JP7306951 | Adaptable transporter monitoring system |
WO/2024/084097 | SELECTIVE ACQUISITION FOR MULTI-MODAL TEMPORAL DATA |
WO/2022/075897 | ROOT CAUSE ANALYSIS |
Inventors:
Keisuke Nakamura
Application Number:
JP2017092076A
Publication Date:
March 15, 2023
Filing Date:
May 07, 2017
Export Citation:
Assignee:
Narel System Co., Ltd.
International Classes:
G06N5/022; G06F9/448; G06F16/30; G06F40/186
Domestic Patent References:
JP2006024045A | ||||
JP2014501988A |
Previous Patent: Self-adaptive indoor fusion positioning method based on dynamic environment
Next Patent: PAPER MONEY DISCRIMINATION DEVICE
Next Patent: PAPER MONEY DISCRIMINATION DEVICE