With the increasing rate of credit card fraud, there is a need to develop a robust system that can accurately detect anomalies (fraud) in the transaction dataset. In collaboration with 2 of my colleagues, we investigated the fraud in a typical credit card transaction by comparing the performance of Deep Neural Network with Logistic Regression.
The tools, algorithms and the socio-ethical risk evaluation of implementing this algorithm is broadly expatiated in the poster below.
First of all, I didn’t start as a programmer or computer scientist. I had my background in studying Medical Sciences. But, ever since I wrote my first ‘hello’ program in Python, and manipulating some variables to produce a desired output, I have been captivated with the idea of using software to solve practical problems.
I am currently an Applied Artificial Intelligence Student and a beneficiary of the prestigious OFS scholarship Award at Teeside University. My passion for Artificial intelligence is attached to the application of cutting edge technology in accurately proposing solutions to data related problems.
My career ambitions are :
- Contribution of distinctive knowledge to the field of data engineering via continuous research and publications of findings;
- Eagerly anticipating teaching as a professional in an Industrial and University settings as well as private practice and consultation.
- Programming Languages and tools: Python, R studio, Jupyter, WordPress, JSON, DAX
- Machine Learning| Deep Learning – Regression, Neural Networks, Naive Bayes, SVM, Decision trees, K means clustering, KNN, gradient descent, Tensor flow
- Statistical Programming and Packages -R, Python, NumPy, Matplotlib, Plotly, Scikit-learn, Pandas, ggplot2, dplyr, Keras
- Business Intelligence and Visualization: Excel, power Bi, Google Analytics, Tableau, AI ethics and research
- Operating System: Window, MacOS