time-series-decomposition-using-wavelet-and-fourier-transforms-for-enhanced-solar-flare-forecasting

https://github.com/dasjar/time-series-decomposition-using-wavelet-and-fourier-transforms-for-enhanced-solar-flare-forecasting

Science Score: 26.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (6.9%) to scientific vocabulary
Last synced: 10 months ago · JSON representation

Repository

Basic Info
  • Host: GitHub
  • Owner: dasjar
  • Language: Jupyter Notebook
  • Default Branch: main
  • Size: 1.2 MB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 1 year ago · Last pushed about 1 year ago
Metadata Files
Readme Citation

readme.md

Time Series Decomposition for Enhanced Solar Flare Forecasting

This repository supports the paper:

"Time Series Decomposition Using Wavelet and Fourier Transforms for Enhanced Solar Flare Forecasting"
Victor Solomon, Omkar Rayala, Manya Rampuria, Abdul Afrid, Junzhi Wen, Rafal Angryk
Department of Computer Science, Georgia State University


Overview

This project investigates how frequency-based time series decomposition can improve multivariate solar flare prediction. We apply four transformation methods:

  • Haar Wavelet Transform (HWT)
  • Symlet Wavelet Transform (SWT)
  • Daubechies Wavelet Transform (DWT)
  • Discrete Fourier Transform (DFT)

We evaluate the effectiveness of these transforms on both lossless and lossy time series reconstructions using two classifiers:

  • Time Series Forest (TSF) for time series data
  • Random Forest (RF) for decomposed, non-time-series (NTS) data

Experiments

Experiment I: TS Reconstruction Impact

We assess the performance of TSF trained on: - Original data - Lossless reconstructed data - Lossy reconstructed data (based on 20dB SNR filtering)

Metrics: TSS and HSS2

Experiment II: TS vs NTS Representations

We compare: - TSF (trained on full and reduced TS) - RF (trained on decomposed NTS)

Time windows tested: 720, 320, 96, and 48 minutes (60, 30, 8, 4 timepoints respectively)


Installation

  1. Clone the repository: ```bash git clone https://github.com/dasjar/Time-Series-Decomposition-Using-Wavelet-and-Fourier-Transforms-for-Enhanced-Solar-Flare-Forecasting.git

Citation

@inproceedings{solomon2025wavelet, title={Time Series Decomposition Using Wavelet and Fourier Transforms for Enhanced Solar Flare Forecasting}, author={Solomon, Victor and Wen, Junzhi and Angryk, Rafal and Rampuria, Manya and Rayala, Omkar and Afrid, Abdul}, booktitle={The International FLAIRS Conference Proceedings}, volume={38}, number={1}, year={2025} }

Owner

  • Name: Victor Solomon
  • Login: dasjar
  • Kind: user
  • Location: Abuja, NIgeria

I'm a software developer. my core languages are: Javasript, Python and Java. my interests are: Big data, Analytics, Machine learning and software development.

GitHub Events

Total
  • Delete event: 2
  • Push event: 16
  • Pull request event: 2
  • Create event: 4
Last Year
  • Delete event: 2
  • Push event: 16
  • Pull request event: 2
  • Create event: 4

Issues and Pull Requests

Last synced: 10 months ago

All Time
  • Total issues: 0
  • Total pull requests: 1
  • Average time to close issues: N/A
  • Average time to close pull requests: less than a minute
  • Total issue authors: 0
  • Total pull request authors: 1
  • Average comments per issue: 0
  • Average comments per pull request: 0.0
  • Merged pull requests: 1
  • Bot issues: 0
  • Bot pull requests: 0
Past Year
  • Issues: 0
  • Pull requests: 1
  • Average time to close issues: N/A
  • Average time to close pull requests: less than a minute
  • Issue authors: 0
  • Pull request authors: 1
  • Average comments per issue: 0
  • Average comments per pull request: 0.0
  • Merged pull requests: 1
  • Bot issues: 0
  • Bot pull requests: 0
Top Authors
Issue Authors
Pull Request Authors
  • dasjar (1)
Top Labels
Issue Labels
Pull Request Labels