ExoSpikeNet: A Light Curve Analysis Based Spiking Neural Network for Exoplanet Detection

Abstract

Exoplanets are celestial bodies orbiting stars beyond our Solar System. Although historically they posed detection challenges, Kepler's data has revolutionized our understanding. By analyzing flux values from the Kepler Mission, we investigate the intricate patterns in starlight that may indicate the presence of exoplanets. This study investigates a novel approach for exoplanet classification using Spiking Neural Networks (SNNs) applied to data obtained from the NASA Kepler mission. SNNs offer a unique advantage by mimicking the spiking behavior of neurons in the brain, allowing for more nuanced and biologically inspired processing of temporal data. Experimental results demonstrate the efficacy of the proposed SNN architecture, excelling in various performance metrics such as accuracy, F1 score, precision, and recall.

6 Pages, 10 Figures, 2 Tables, Accepted by the 13th IEEE International Conference on Communication Systems and Network Technologies(CSNT 2024), April 06-07,2024,India

References in corpus (5)

• Automatic Classification of Kepler Planetary Transit Candidates
• Comparative Evaluation of Metaheuristic Algorithms for Hyperparameter Selection in Short-Term Weather Forecasting
• Light Curve Analysis of Ground-Based Data from Exoplanets Transit Database
• A Comparative Analysis on Metaheuristic Algorithms Based Vision Transformer Model for Early Detection of Alzheimer's Disease
• A Possible Converter to Denoise the Images of Exoplanet Candidates through Machine Learning Techniques