2012 Fiscal Year Final Research Report
Development of BCI speller for Japanese Language
Project/Area Number |
22500200
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
Sensitivity informatics/Soft computing
|
Research Institution | Nagoya University |
Principal Investigator |
|
Co-Investigator(Kenkyū-buntansha) |
YOSHIKAWA Tomohiro 名古屋大学, 大学院・工学研究科, 准教授 (20303753)
|
Project Period (FY) |
2010 – 2012
|
Keywords | BCI / P300 Speller / EEG / RB-ARQ / Ngram / 追加学習 |
Research Abstract |
A Brain Computer Interface (BCI) system for Japanese language, which enables us to input letters only by thoughts, was developed. Reliability Based Automatic Repeat reQuest (RB-ARQ), Error correction method based on error potential, and RB-Selective Repeat-ARQ (RB-SR-ARQ) that selects stimuli, were proposed. A method to predict candidate words based on the previous letters, a method to utilize transition probabilities from the previous letters for discriminant analysis were also proposed. Experimental results show significant enhancements of spelling speed. An incremental learning method, which does not require pre-training data at the re-start of letter input even a week later, was proposed.
|
Research Products
(11 results)