Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

How Long short-term memory artificial neural network, synthetic data, and fine-tuning improve the classification of raw EEG data

Albert Nasybullin, Vladimir Maksimenko, Semen Kurkin

Abstract

In this paper, we discuss a Machine Learning pipeline for the classification of EEG data. We propose a combination of synthetic data generation, long short-term memory artificial neural network (LSTM), and fine-tuning to solve classification problems for experiments with implicit visual stimuli, such as the Necker cube with different levels of ambiguity. The developed approach increased the quality of the classification model of raw EEG data.

How Long short-term memory artificial neural network, synthetic data, and fine-tuning improve the classification of raw EEG data

Abstract

In this paper, we discuss a Machine Learning pipeline for the classification of EEG data. We propose a combination of synthetic data generation, long short-term memory artificial neural network (LSTM), and fine-tuning to solve classification problems for experiments with implicit visual stimuli, such as the Necker cube with different levels of ambiguity. The developed approach increased the quality of the classification model of raw EEG data.

Paper Structure

This paper contains 13 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Data Collection and Data Structure
  3. Visual Stimuli
  4. Task
  5. EEG data acquisition
  6. Data Structure
  7. LSTM neural network, synthetic data generation, and fine-tuning
  8. LSTM neural network model
  9. LSTM classification of original data
  10. Expanding the size of the dataset with band-pass filtering and pre-training LSTM model
  11. LSTM classification with pre-trained models
  12. Discussion
  13. Conclusion

Figures (3)

  • Figure 1: Ambiguous values of the Necker cube in visual stimulus b2
  • Figure 2: The architecture of the Long short-term memory artificial neural network model used for classification.
  • Figure 3: We build our intuition behind data expanding strategy on a difference created in a signal by band-pass filtering.