Short Message Service Application by Using Brain Control System and Support Vector Machine (SVM) on Single Channel Electroencephalography (EEG)

Abstract

Most people who suffer from physical and sensory disabilities have limited activities. They need communication tools to facilitate communication activities with others. The purpose of this research is to create an application that translates thoughts to text which will be implemented in SMS feature by taking raw EEG from Emotiv EPOC, filtering, and applying machine learning algorithm, which is Support Vector Machine. There are two research steps: analysis and implementation. In the analysis step, the EEG samples taken from respondents are used for analyzing the most dominant channel. Then, EEG signal extraction uses Emotiv EPOC SDK, filters EEG signal taken from the most dominant channel and applies SVM algorithm for data training. C# based UI application is used as interactive media, so user can see the extraction result. The result of this research is an application that translates human thoughts to SMS.