On Reducing the Number of Variables to Estimate Visualized Words using Event Related Potentials

M.F. Funada, M. Shibukawa, Y. Igarashi, T. Funada, and S.P. Ninomija (Japan)

Keywords

EEG, ERP, visualized word, experimental estimation, expression of ERP,

Abstract

The purpose of this research is to set up verbal communication systems between computers and persons, or between handicapped persons and normal persons. In order to make substantial progress toward the research goal, we use Event Related Potentials (ERP’s for short), which are caused by Electroencephalograms (EEG’s for short). We have investigated the estimation of visualized words, phrases, or sentences that may contain some homonyms or related words. In our estimation, we use a part of ERP’s measured from the subjects. It is desirable if we can reduce the number of variables used to store the measured data. A smaller number of variables is better, since the calculation for the estimation becomes simpler. In this paper, we show a technique of how we can reduce the number of variables. Furthermore, we propose a new method for representing ERPs.

Important Links:



Go Back