公开(公告)号：US20180092567A1

公开(公告)日：2018-04-05

申请号：US15564071

申请日：2016-04-05

申请人： NATIONAL INSTITUTE OF INFORMATION AND COMMUNICATIONS TECHNOLOGY

发明人： Shinji NISHIMOTO ,  Hideki KASHIOKA

IPC分类号： A61B5/0484 ,  G06F3/01 ,  A61B5/16

CPC分类号： A61B5/0484 ,  A61B5/04842 ,  A61B5/055 ,  A61B5/165 ,  A61B5/7267 ,  G06F3/015

摘要： A perceptual semantic content estimation method includes: (A) inputting, to data processing means, brain activity induced in a subject by a training stimulation and detected as an output of a brain activity detection means and an annotation of a perceptual content; (B) associating a sematic space representation of the training stimulation and the output of the brain activity detection means in a stored semantic space and storing the association in a training result information storage means; (C) inputting, to the data processing means, an output when the brain activity detection means detects brain activity induced by a novel stimulation, and obtaining a probability distribution in the semantic space which represents perceptual semantic contents for the output of the novel stimulation-induced brain activity by the brain activity detection means on the basis of the association; and (D) estimating a highly probable perceptual semantic content on the basis of the probability distribution.

公开(公告)号：US20180314687A1

公开(公告)日：2018-11-01

申请号：US15740256

申请日：2016-12-22

申请人： National Institute of Information and Communications Technology ,  NTT Data Corporation ,  NTT DATA INSTITUTE OF MANAGEMENT CONSULTING, Inc.

发明人： Shinji NISHIMOTO ,  Satoshi NISHIDA ,  Hideki KASHIOKA ,  Ryo YANO ,  Naoya MAEDA ,  Masataka KADO ,  Ippei HAGIWARA ,  Takuya IBARAKI

IPC分类号： G06F17/27 ,  G06F17/16 ,  G06T7/00 ,  G06Q30/02 ,  G06F17/28

CPC分类号： G06F17/30

摘要： A viewing material evaluating method includes: a brain activity measuring step of measuring a brain activity of a test subject who views a viewing material by using a brain activity measuring unit; a first matrix generating step of generating a first matrix estimating a semantic content of perception of the test subject on the basis of a measurement result acquired in the brain activity measuring step by using a first matrix generating unit; a second matrix generating step of generating a second matrix by performing natural language processing for text information representing a planning intention of the viewing material by using a second matrix generating unit; and a similarity calculating step of calculating similarity between the first matrix and the second matrix by using a similarity calculating unit.

公开(公告)号：US20160110642A1

公开(公告)日：2016-04-21

申请号：US14787903

申请日：2014-05-15

申请人： NATIONAL INSTITUTE OF INFORMATION AND COMMUNICATIONS TECHNOLOGY

发明人： Shigeki MATSUDA ,  Xugang LU ,  Chiori HORI ,  Hideki KASHIOKA

IPC分类号： G06N3/08 ,  G06N3/04

CPC分类号： G06N3/08 ,  G06N3/04 ,  G06N3/0445 ,  G06N3/0454 ,  G06N3/084 ,  G10L15/16

摘要： Provided is a DNN learning method that can reduce DNN learning time using data belonging to a plurality of categories. The method includes the steps of training a language-independent sub-network 120 and language-dependent sub-networks 122 and 124 with training data of Japanese and English. This step includes: a first step of training a DNN obtained by connecting neurons in an output layer of the sub-network 120 with neurons in an input layer of sub-network 122 with Japanese training data; a step of forming a DNN by connecting the sub-network 124 in place of the sub-network 122 to the sub-network 120 and training it with English data; repeating these steps alternately until all training data ends; and after completion, separating the first sub-network 120 from other sub-networks and storing it as a category-independent sub-network in a storage medium.

摘要翻译： 提供了一种可以使用属于多个类别的数据来减少DNN学习时间的DNN学习方法。 该方法包括以日语和英语的训练数据训练与语言无关的子网络120和与语言相关的子网络122和124的步骤。 该步骤包括：训练通过将子网络120的输出层中的神经元与子网络122的输入层中的神经元连接到日本训练数据而获得的DNN的第一步骤; 通过将子网络124代替子网络122连接到子网络120并用英语数据进行培训来形成DNN的步骤; 交替重复这些步骤，直到所有训练数据结束; 并且在完成之后，将第一子网120与其他子网分离，并将其作为类别独立的子网存储在存储介质中。