Abu Tayeb Mohammad Ayon

During my time at the NSU DIAL Lab, I had the opportunity to work on several groundbreaking projects. One of the most notable was the development of an alternative credit scoring system that leverages non-traditional data sources to provide financial inclusion for underserved populations. This project involved extensive qualitative research, including interviews and focus groups, to understand the financial behaviors and needs of low-income individuals.

Additionally, I collaborated with a multidisciplinary team to design and implement a user-friendly interface for the credit scoring system. This experience not only honed my technical skills but also deepened my understanding of the social impact of technology.

Our lightweight Deep Convolutional Neural Network (DCNN) model was designed to address the challenges of skin cancer classification. By pruning the network and incorporating Explainable AI (XAI) techniques, we achieved an accuracy of 98.07% on the HAM10000 dataset. This model not only outperformed state-of-the-art models in terms of accuracy but also significantly reduced computational complexity, making it suitable for deployment on resource-constrained devices.

The development process involved extensive experimentation with various architectures and hyperparameters. We also conducted a thorough analysis of the model's interpretability, ensuring that medical professionals could trust and understand the predictions made by the model.