Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM
Document Type and Number:
WIPO Patent Application WO/2019/189016
Kind Code:
A1
Abstract:
This information processing device 100 is configured so as to comprise a model generation means 110 for, on the basis of data formed from a plurality of attributes, generating a model using the attributes, and an attribute change means 120 for, on the basis of the generated model, changing the attributes used in the generation of a model, wherein the model generation means 110 generates a new model on the basis of data formed from the changed attributes.

Inventors:
SHIRAKI TAKASHI (JP)
Application Number:
PCT/JP2019/012569
Publication Date:
October 03, 2019
Filing Date:
March 25, 2019
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
NEC CORP (JP)
International Classes:
G06N20/00
Domestic Patent References:
WO2017168458A12017-10-05
Other References:
YONEYA, M.: "Intensive course in seven days! An introduction to Excel statistics", 25 August 2016 (2016-08-25), pages 165 - 196, ISBN: 978-4-274-21888-0
KHARKAR, P ., BACKWARD ELIMINATION FOR MULTIPLE LINEAR REGRESSION, 14 February 2018 (2018-02-14), pages 1 - 28, Retrieved from the Internet [retrieved on 20190424]
IWUNDU, M. P. ET AL.: "On the adequacy of variable selection techniques on model building", ASIAN JOURNAL OF MATHEMATICS AND STATISTICS, vol. 8, no. 1, 2015, pages 19 - 34, XP055638558, ISSN: 1994-5418, Retrieved from the Internet [retrieved on 20190424]
ETO, R. ET AL.: "Fully-automatic bayesian piecewise sparse linear models", PROCEEDINGS OF MACHINE LEARNING RESEARCH, vol. 33, 25 April 2014 (2014-04-25), pages 238 - 246, XP055137253, Retrieved from the Internet [retrieved on 20190422]
FUJIMAKI, R.: "Piecewise sparse linear classification via factorized asymptotic bayesian inference", JOURNAL OF THE JAPANESE SOCIETY FOR ARTIFICIAL INTELLIGENCE, vol. 32, no. 1, 1 January 2017 (2017-01-01), pages 30 - 38, XP055638564, ISSN: 2188-2266
Attorney, Agent or Firm:
BABA, Motohiro et al. (JP)
Download PDF: