DATA ANALYSIS DEVICE, DATA ANALYSIS METHOD AND PROGRAM

Japanese Patent JP3813837

B2

PROBLEM TO BE SOLVED: To provide a data analysis device to obtain a concept time-series model which can be adapted to support decision making based on the sequence of text provided with time information.
SOLUTION: In a storing part 1 and a storing part 2, the text, aggregation of documents containing the time information and attribute information and a key concept dictionary are each stored. In the concept extracting part 3, a sub-class consisting of a multiple number of documents that are related to each other is selected from the documents based on the attribute information and a key concept is extracted from each of the multiple number of documents including the sub-class based on the key concept dictionary. In the document time extraction part 4, time information is extracted from each of the multiple number of documents including the sub-class. In the concept time-series data producing part 5, concept time-series data where the key concepts are lined up according to time-series produced based on the extracted key concept and the time information. In a concept time-series model learning part 6, the concept time-series data is made into a model.

Shigeaki Sakurai

JP2001157198A

August 23, 2006

May 25, 2001

Click for automatic bibliography generation  

Toshiba Corporation

G06F17/30; (IPC1-7): G06F17/30

JP2000172701A
JP2001075966A

櫻井 茂明,向井 理朗,岡 隆一,対話タスクモデルのサンプル単語時系列からの自己組織化,電子情報通信学会論文誌 (J83-D-II) 第2号,日本,社団法人電子情報通信学会,2000年 2月25日,第827-839頁

Takehiko Suzue
Sadao Muramatsu
Atsushi Tsuboi
Ryo Hashimoto
Satoshi Kono
Makoto Nakamura
Shoji Kawai

Previous Patent: INCOMPLETE FITTING PREVENTING CONNECTOR

Next Patent: FIXING STRUCTURE OF WIRE COVER FOR CONNECTOR