This project leverages on Clustering, Deep Learning techniques and Tree models to perform a comparative analysis with the aim of finding the best Machine Learning techniques to tackle credit card fraud in the e-commerce industry
This project builds a detection system for fintech transaction logs using Isolation Forest and DBSCAN models, with SHAP-based explainability and robust log parsing to identify suspicious user behavior and fraudulent patterns
Perform several key tasks in quantitative finance: calibrate the Heston stochastic volatility model to option data using both the Lewis (2001) and Carr-Madan (1999) Fourier pricing methods, then extends this to the Bates (1996) model. Finally, calibrate the CIR model to the Euribor term structure and simulate future interest rates, demonstrating the application of stochastic models to both equity derivatives and interest rates.
This project is an attempt to replicate the methodology of "An Intelligent Approach for Predicting Stock Market Movements in Emerging Markets Using Optimized Technical Indicators and Neural Networks" paper by Alma Rocío Sagaceta-Mejía, Máximo Eduardo Sánchez-Gutiérrez, and Julián Alberto Fresán-Figueroa.
I led the Data preprocessing phase of a project at Omdena that visualizes the recent price surge in Nigeria
For this Maven Environmental Challenge, Apple has a lofty goal of achieving Carbon neutrality by 2030. I worked as a journalist and Data visualization enthusiast, providing a report on their progress.
I built a classification model that predicts whether the price of a crypto asset rises or falls in a two-week window.
I gathered, accessed and cleaned data from a Twitter account @WeRateDogs. Python's libraries such as Pandas, NumPy, Matplotlib & Seaborn were used to gather data manually and programatically, access three datasets for quality and tidiness issues, fix the issues identified (data cleaning), then derive insights from the clean data.
Using mostly visualizations with Python, I explored a dataset from Prosper - a loan company. I used Matplotlib & Seaborn to create univariate, bivariate and multivariate plots to find out what factors influences the Annual Percentage Rate (APR) of a loan.
