CAUSALITY ESTIMATION DEVICE, CAUSALITY ESTIMATION METHOD, AND PROGRAM

WIPO Patent Application WO/2019/160138

A1

This causality estimation device comprises: an input unit that inputs data of a time-series multidimensional numeric vector; a regression model learning unit that uses the inputted data of the time-series multidimensional numeric vector to learn a nonlinear regression model that predicts data at a given time on the basis of data from a past time; a causality estimation unit that uses the nonlinear regression model to calculate the strength of causality of a dimension j in relation to a dimension i in the data of the time-series multidimensional numeric vector; and an output unit that outputs the strength of the causality calculated by the causality estimation unit.

IKEDA YASUHIRO (JP)
MATSUO YOICHI (JP)
NAKANO YUSUKE (JP)
ISHIBASHI KEISUKE (JP)
WATANABE KEISHIRO (JP)
KAWAHARA RYOICHI (JP)

PCT/JP2019/005857

August 22, 2019

February 18, 2019

Click for automatic bibliography generation  

NIPPON TELEGRAPH & TELEPHONE (JP)

G06N20/00; G06N3/02

WO2018025361A1

2018-02-08

CN107126193A

2017-09-05

N AGATANI, TOMOHIRO ET AL.: "Review of wind speed prediction considering the causal relationship", PROCEEDINGS OF THE 2016 IEICE ENGINEERING SCIENCES SOCIETY/NOLTA SOCIETY CONFERENCE, 6 September 2016 (2016-09-06), pages 116, ISSN: 2189-700X
ABE, HITOSHI ET AL.: "Lasso Granger Causality Estimation Considering Smoothness of Causality from Time Series Data", IEICE TECHNICAL REPORT, vol. 115, no. 112, 16 June 2015 (2015-06-16), pages 55 - 62, ISSN: 0913-5685

ITOH, Tadashige et al. (JP)

View/Download PDF      PDF Help

Previous Patent: LUMINOUS FLUX CONTROL MEMBER, LIGHT EMISSION DEVICE, AND ILLUMINATION DEVICE

Next Patent: MULTILAYER BOARD, LOW-PASS FILTER, HIGH-PASS FILTER, MULTIPLEXER, HIGH-FREQUENCY FRONT-END CIRCUIT, ...