Efe Ali Mert
High School Researcher in Mathematics and Computer Science
Short Bio
I am a high school student at ISTEK Bilge Kağan High School in Istanbul, Türkiye. My academic interests lie at the intersection of mathematics, computer science, and complex systems. In particular, I am interested in nonlinear dynamics, chaos theory, cryptography, and machine learning.
My work focuses on exploring how mathematical structures can be used to model complex phenomena and develop computational methods. I have conducted research on chaos-based encryption systems and machine learning approaches for analyzing neural signals such as EEG data.
Research Interests
Mathematics
- Chaos theory and nonlinear dynamical systems
- Cryptography and information security
- Machine learning and data analysis
- Computational neuroscience
Selected Achievements
- Second Prize — TÜBİTAK 2204-A National High School Research Competition (Mathematics)
- Oral Presentation — 19th International Conference on Chaotic Modeling and Simulation (CHAOS 2026), Greece
Research
Chaos-Based DNA Encryption
This project explores the use of nonlinear dynamical systems in cryptography. The goal was to design an encryption method that combines chaos theory with DNA-based encoding.
The system integrates several chaotic structures including the logistic map, Lorenz system, Mandelbrot set, and Hénon map to generate highly sensitive pseudo-random sequences. These sequences are then used together with a One-Time Pad framework to enhance encryption security.
The project investigates how deterministic chaotic systems can generate complex and unpredictable patterns that are suitable for cryptographic applications.
This research received Second Prize in the national TÜBİTAK 2204-A High School Research Competition in the field of mathematics and has been accepted for oral presentation at the 19th International Conference on Chaotic Modeling and Simulation.
ADHD Detection Using Fractal Analysis of EEG Signals
This research investigates whether fractal dimension analysis and machine learning techniques can be used to distinguish ADHD patients from healthy individuals using EEG data.
The project focuses on extracting complexity-based features such as the Higuchi fractal dimension and entropy-based measures from EEG signals. These features are then used in machine learning models including Random Forest and ensemble methods.
The current model achieves approximately 79.5% classification accuracy and explores the potential of combining nonlinear analysis methods with machine learning for neuroscience applications.
Contact
- Email: efealimert@efealimert.com
- GitHub: github.com/EfeAliMert
- LinkedIn: linkedin.com/in/efealimert