., Vincentius (2019) Klasifikasi Sinyal P300 Menggunakan Principal Component Analysis, Linear Discriminant Analysis, dan Support Vector Machine. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
05111540000159-Undergraduate_Theses.pdf Download (2MB) | Preview |
Abstract
Brain Computer Interface merupakan salah satu metode interaksi yang dapat digunakan oleh penyandang cacat untuk dapat berinteraksi dengan komputer. Interaksi terjadi secara langsung antara otak pengguna dengan komputer melalui pengambilan sinyal electroencephalogram (EEG) yang kemudian diklasifikasi untuk mendeteksi Sinyal P300. Pada tugas akhir ini, dilakukan eksperimen untuk mengklasifikasi Sinyal P300 dengan menerapkan variasi pemilihan channel sinyal EEG dengan Principal Component Analysis (PCA). Kemudian, untuk mempercepat proses training, dilakukan reduksi dimensi temporal dan channel lebih lanjut menggunakan Linear Discriminant Analysis (LDA). Proses klasifikasi sinyal EEG dilakukan dengan algoritma Support Vector Machine (SVM). Uji coba dilakukan menggunakan dataset Wadsworth BCI Dataset (P300 Evoked Potentials), BCI Competition III Challenge 2004. Hasil dari uji coba menghasilkan rata-rata f1-score terbaik sebesar 86% untuk subjek A dan 90% untuk subjek B. Proses Linear Discriminant Analysis (LDA) berhasil mengurangi waktu training hingga dibawah 1 detik untuk semua skenario uji coba. Rata-rata f1-score terbaik setelah dilakukan LDA untuk subjek A adalah sebesar 84% dan untuk subjek B sebesar 89%.
================================================================================================
Brain Computer Interface is one of the interaction method that can be used by people with physical disability to interact with computers. Interaction occurs directly between the user’s brain to the computer by collecting the electroencephalogram (EEG) signals that can be classified to detect the P300 Signal. In this final project, an experiment is done to classify P300 Signal by applying various channel selection of the EEG signal using Principal Component Analysis (PCA). To speed up training process, further dimensionality reduction is applied to reduce channels and temporal signal using Linear Discriminant Analysis (LDA). The classification process is done using the Support Vector Machine (SVM) algorithm. The dataset used for this experiment is from Wadsworth BCI Dataset (P300 Evoked Potentials), BCI Competition III Challenge 2004. From this experiment, the best f1-score obtained for subject A is 86% and for subject B is 90%. The LDA process successfully reduce training time to under 1 second for all experiment scenarios, resulting the f1-score of 84% for subject A and 89% for subject B.
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSIf 621.398 1 Vin k-1 2019 |
| Uncontrolled Keywords: | Brain Computer Interface; Electroencephalogram; Linear Discriminant Analysis; Principal Component Analysis; Sinyal P300; Support Vector Machine |
| Subjects: | R Medicine > RC Internal medicine > RC386.5 Electroencephalography. T Technology > T Technology (General) > T174 Technological forecasting |
| Divisions: | Faculty of Information and Communication Technology > Informatics > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Vincentius Vincentius |
| Date Deposited: | 09 Jul 2021 03:53 |
| Last Modified: | 09 Jul 2021 03:53 |
| URI: | http://repository.its.ac.id/id/eprint/60800 |
Actions (login required)
 |
View Item |