Title:
FEATURE SELECTION PROGRAM, DEVICE, AND METHOD
Document Type and Number:
WIPO Patent Application WO/2022/190384
Kind Code:
A1
Abstract:
This feature selection device specifies a feature of a superordinate concept having, as the subordinate concept thereof, a feature included in a set of features acquired from a knowledge graph. Further, when all the plurality of hypotheses each having a different feature of the subordinate concept and stating that a condition represented by a combination of at least one feature including a feature of the subordinate concept influences an objective variable, are true, the feature selection device determines that a hypothesis obtained by replacing the feature of the subordinate concept with the feature of the superordinate concept is also true, and selects the specified feature of the superordinate concept as a feature to be added to the set of features. When there is a hypothesis that includes a feature of the subordinate concept and is not true, the feature selection device determines that the hypothesis obtained by replacing the feature of the subordinate concept with the feature of the superordinate concept is not true, and does not select the specified feature of the superordinate concept.
More Like This:
Inventors:
FUKUDA TAKASABUO (JP)
Application Number:
PCT/JP2021/010196
Publication Date:
September 15, 2022
Filing Date:
March 12, 2021
Export Citation:
Assignee:
FUJITSU LTD (JP)
International Classes:
G06N20/00
Domestic Patent References:
| WO2020053934A1 | 2020-03-19 |
Foreign References:
| JP2017174022A | 2017-09-28 | |||
| JP2017146869A | 2017-08-24 | |||
| US20170105867A1 | 2017-04-20 | |||
| US20190138806A1 | 2019-05-09 | |||
| JP2020046888A | 2020-03-26 |
Other References:
SUZUMURA, SHINYA ET AL.: "Modified Bonferroni method for Selective Inference", IEICE TECHNICAL REPORT, vol. 116 (PRMU2016-59), no. 209 (IBISML2016-14), 13 October 2016 (2016-10-13), pages 39 - 45, XP009549313, ISSN: 0913-5685
H. AKAIKE: "Information theory and an extension of the maximum likelihood principle", 2ND INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, 1973, pages 267 - 281
R. MIYASHIROY. TAKANO: "Mixed Integer Second-Order Cone Programming Formulations for Variable Selection in Linear Regression", EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, vol. 247, 2015, pages 721 - 731
HIROAKI IWASHITATAKUYA TAKAGIHIROFUMI SUZUKIKEISUKE GOTOKOTARO OHORIHIROKI ARIMURA: "Efficient Constrained Pattern Mining Using Dynamic Item Ordering for Explainable Classification", ARXIV:2004.08015, Retrieved from the Internet
H. AKAIKE: "Information theory and an extension of the maximum likelihood principle", 2ND INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, 1973, pages 267 - 281
R. MIYASHIROY. TAKANO: "Mixed Integer Second-Order Cone Programming Formulations for Variable Selection in Linear Regression", EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, vol. 247, 2015, pages 721 - 731
HIROAKI IWASHITATAKUYA TAKAGIHIROFUMI SUZUKIKEISUKE GOTOKOTARO OHORIHIROKI ARIMURA: "Efficient Constrained Pattern Mining Using Dynamic Item Ordering for Explainable Classification", ARXIV:2004.08015, Retrieved from the Internet
Attorney, Agent or Firm:
NAKAJIMA, Jun et al. (JP)
Download PDF: