BEED: Bangalore EEG Epilepsy Dataset

Donated on 3/10/2025

The Bangalore EEG Epilepsy Dataset (BEED) is a comprehensive EEG collection for epileptic seizure detection and classification. Recorded at a neurological research centre in Bangalore, India, it features high-fidelity EEG signals captured using the standard 10-20 electrode system at a 256 Hz sampling rate. BEED contains 16,000 segments of 20-second EEG recordings evenly distributed across four categories: Healthy Subjects (0), Generalized Seizures (1), Focal Seizures (2), and Seizure Events (3), where seizure activity occurs with events like eye blinking, nail biting, or staring. Each category includes data from 20 adult subjects (ages 21-55) with equal gender representation. The dataset comprises 16 EEG channels (X1-X16) corresponding to different brain regions, with a binary label (y) indicating seizure presence (1) or absence (0). BEED supports machine learning in seizure detection, epilepsy analysis, and EEG research with its balanced, high-resolution data.

Dataset Characteristics

Tabular, Multivariate

Subject Area

Health and Medicine

Associated Tasks

Classification

Feature Type

Integer

# Instances

8000

# Features

17

Dataset Information

Has Missing Values?

No

Introductory Paper

Feature Engineering for Epileptic Seizure Classification Using SeqBoostNet

By Najmusseher; P.K. Nizar Banu. 2024

Published in International Journal of Computing and Digital Systems

Variables Table

Variable NameRoleTypeDescriptionUnitsMissing Values
X1FeatureIntegerno
X2FeatureIntegerno
X3FeatureIntegerno
X4FeatureIntegerno
X5FeatureIntegerno
X6FeatureIntegerno
X7FeatureIntegerno
X8FeatureIntegerno
X9FeatureIntegerno
X10FeatureIntegerno

0 to 10 of 17

Additional Variable Information

Dataset Details Dataset Characteristics: Multivariate Feature Type: Integer Subject Area: Health and Medicine Instances: 8,000 (2,000 per category) Classes: 4 Format: CSV Features: 16 (EEG channels) Associated Tasks: Classification No missing Values Data Structure • Subjects: 80 individuals (20 per file across four folders) • Time Points: 200 per subject • Explanatory Variables (X1–X16): 16 features representing EEG signal characteristics • Response Variable (y) (Column 17): Categorizes the EEG recordings into four distinct classes: o 3 – Recording of seizure events (including activities such as eye blinking or constant staring) o 2 – Recording of focal seizures o 1 – Recording of generalized seizures o 0 – Recording of healthy subjects (control group without epileptic seizures)

Dataset Files

FileSize
BEED_Data.csv394.9 KB

Reviews

There are no reviews for this dataset yet.

Login to Write a Review
Download (126.5 KB)
1 citations
1578 views

Creators

NAJMUSSEHER .

najmusseher@res.christuniversity.in

CHRIST (Deemed to be University), Central Campus, Bangalore

NIZAR BANU P K

nizar.banu@christuniversity.in

CHRIST (Deemed to be University), Central Campus, Bangalore

License

By using the UCI Machine Learning Repository, you acknowledge and accept the cookies and privacy practices used by the UCI Machine Learning Repository.

Read Policy