Papers
3
Total Citations
90
H-Index
3
About
Fathma Siddique is a researcher specializing in deep learning, computer vision, and artificial neural networks, with a particular focus on handwritten digit recognition systems. Her work sits at the intersection of machine learning theory and practical implementation, leveraging powerful frameworks such as Python and TensorFlow to build and evaluate Convolutional Neural Network (CNN) architectures. Siddique's most notable contribution examines how varying the number of hidden layers in CNN models affects recognition accuracy and overall performance on handwritten digit datasets — a foundational problem in pattern recognition research. Her 2019 paper on this topic has garnered an impressive 77 citations, demonstrating its strong uptake within the machine learning and computer vision communities. Her follow-up studies further reinforce these findings, collectively establishing a body of work that offers both methodological rigor and practical guidance for practitioners designing neural network architectures. What makes Siddique's research particularly valuable for students and early-career researchers is its clear, comparative approach — systematically benchmarking architectural choices to illuminate best practices. Her contributions provide accessible yet technically meaningful insights into how deep learning models can be optimized, making her work a useful reference point for anyone entering the rapidly evolving field of AI-driven image recognition.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3