An in-depth machine learning project which studies the complex dynamics of earthquakes through spatial-temporal data analysis. Various machine learning and data-driven techniques were utilized in the study, which includes Fast Fourier Transform, Gabor Transform, Singular Value Decomposition, Dynamic Mode Decomposition, Chaos Theory, and Neural Networks.
- Temporal analysis using Fast Fourier Transform (FFT) and Gabor Transform can be found in the MATLAB file named 'fft_gabor_transform.m'.
- Coherent structures derived using Singular Value Decomposition (SVD) and Dynamic Mode Decomposition (DMD) can be found in the MATLAB file named 'coherent_structures_dmd.m'.
- Detection of chaos within system using the Largest Lyapunov Exponent can be found in the MATLAB file named 'largest_lyapunov_exponent.m'.
- Future state prediction of earthquakes using DMD can be found in the MATLAB file named 'state_prediction_dmd.m'.
- Future state prediction of earthquakes using Neural Networks can be found in the MATLAB file named 'state_prediction_neural_networks.m'.
- Input dataset of the project is not included in this repository.
- The algorithms used and results derived in the project are thoroughly discussed in the report titled 'ME5311 Project Report.pdf'.