Statistical Feature Extraction and Classification using Machine Learning Techniques in Brain-Computer Interface
1M.Kanimozhi, Ph.D research Scholar of Computer Science, Sri Sarada College for Women (Autonomous), Salem, India.
2Dr.R.Roselin, Associate Professor of Computer Science, Sri Sarada College for Women (Autonomous), Salem, India.
Manuscript received on December 16, 2019. | Revised Manuscript received on December 22, 2019. | Manuscript published on January 10, 2020. | PP: 1754-1758 | Volume-9 Issue-3, January 2020. | Retrieval Number: K23430981119/2020©BEIESP | DOI: 10.35940/ijitee.K2343.019320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Brain Computer Interface is a paralyzed system. This system is used for direct communication between brain nerves and computer devices. BCI is an imagery movement of the patients who are all unable to communicate with the people. In EEG signals feature extraction plays an important role. Statistical based features are essential feature being used in machine learning applications. Researchers mainly focus on the filters and feature extraction techniques. In this paper data are collected from the BCI Competition III dataset 1a. Statistical features like minimum, maximum, standard deviation, variance, skewnesss, kurtosis, root mean square, average, energy, contrast, correlation and Homogeneity are extracted. Classification is done using machine learning techniques such as Support Vector Machine, Artificial Neural Network and K-Nearest Neighbor. In the proposed system 90.6% accuracy is achieved.
Keywords: BCI Data, SVM, Neural Network, K-NN, Mean Min, Max, Standard Deviation, Variance, Kurtosis, Skewness, and Root mean Square
Scope of the Article: Classification