Paper Image

EEG-based attention type classification

Published on:

6 November 2023

Primary Category:

Human-Computer Interaction

Paper Authors:

Dhruv Verma,

Sejal Bhalla,

S. V. Sai Santosh,

Saumya Yadav,

Aman Parnami,

Jainendra Shukla

Bullets

Key Details

Proposes AttentioNet, a novel CNN + self-attention model for EEG-based attention classification

Conducts study with 20 subjects performing neuropsych tests eliciting 5 attention types

AttentioNet achieves 92.3% accuracy in classifying attention types from EEG data

Transfer learning personalization handles EEG signal variability between subjects

Outperforms EEGnet baseline, showing promise despite dataset limitations

Explore the topics in this paper

AI generated summary

EEG-based attention type classification

This paper proposes AttentioNet, a novel deep learning approach using EEG data to classify attention into 5 types: selective, sustained, divided, alternating, and relaxed. A study of 20 subjects performing neuropsych tests eliciting different attention states was conducted. AttentioNet, a CNN with self-attention, achieved 92.3% accuracy in classifying attention types. Personalization via transfer learning addressed EEG signal variability. Results show AttentioNet outperforms EEGnet, confirming its effectiveness despite dataset limitations.

Answers from this paper

Comments

No comments yet, be the first to start the conversation...

Sign up to comment on this paper

Sign Up