Merger & Acquisition

MERGER & ACQUISITION SUCCESS PREDICTION

This project leverages Machine Learning and Natural Language Processing (NLP) to enhance the prediction of successful M&A deal completions. Using Python, I built and benchmarked multiple classification models (Logistic Regression, Random Forest, Neural Networks, Decision Trees, SVM) on structured financial datasets. In parallel, I conducted sentiment analysis on pre-announcement news and rumors using VADER and TextBlob to extract market sentiment features. These, were integrated into a hybrid predictive model, combining financial and textual data: this approach improved classification performance, offering more reliable forecasts of M&A outcomes and highlighting key financial and sentiment drivers behind deal success.

Learn more
AI Chatbot for Internal Training

AI Chatbot for Internal Training

I am building an AI Chatbot trained to provide knowledge articles and immediate answers to New Joiners working in a new Logistics Team. Final project coming soon!

Artwork Dataset Cleaning

Artwork Dataset Cleaning, Integration & Analysis with Python

Worked on integrating, cleaning, and analyzing a complex artworks dataset. Identified data limitations, inconsistencies, and prepared it for public-facing consumption. Conducted exploratory data analysis (EDA) and produced visual insights to reveal patterns across institutions, artists, and artwork types.

Learn more
Products & Sales Performance: A Business Analysis

Products & Sales Performance: A Business Analysis

This interactive Google Looker Studio dashboard analyzes sales performance across Australian states from 2018 to 2020. Key insights include quarterly and monthly trends, state-wise comparisons, and best-selling products by region. The analysis highlights how sales patterns vary across states and over time, supporting data-driven decisions for targeted sales strategies.

Learn more
Fake vs True News Detection with NLP

Fake vs True COVID-Related News Detection with NLP

Developed and evaluated binary classification models to detect fake vs. real COVID-19 news using a balanced dataset of 1,164 articles. Achieved 94% accuracy with Multinomial Naïve Bayes and 100% accuracy with a Passive-Aggressive Classifier on the test set. Explored model performance through confusion matrices and validated the robustness of these approaches for text-based classification tasks.

Learn more
House Price Prediction with Regression and SVM (R)

House Price Prediction with Regression and Support Vector Machines (SVM) with R

This project investigates the relationship between house prices and key predictors using Linear Regression, Polynomial Regression, and Support Vector Machine (SVM) models. Initial linear regression assumptions (linearity, homoscedasticity) were violated, so a Polynomial Regression was employed, showing statistically significant relationships between key predictors. However, the model explained only ~8.7% of price variance (R² = 0.087) on the full dataset. After removing outliers (houses > $500k), the model performance slightly improved (R² = 12.7%). To further improve performance, SVM regression was applied on the cleaned dataset, achieving the best results with R² = 25%, outperforming both the linear and polynomial models.

Learn more
Skin Health Equity Mapping

Skin Health Equity Mapping

This project aims to identify which skin conditions disproportionately affect specific marginalized groups, and which conditions are under-served in product availability or care. Customer Segmentation will also be completed, with the creation of an African sourced Skin Ingredients Recommendation Engine. The project objective is to: raise awareness of racial or socioeconomic disparities in dermatological care and to highlight the benefits of African (Ghanian) sourced skincare botanicals. Final project coming soon!

Optimizing Marketing Campaingns: A/B Testing with ETL Pipeline in SQL

Optimizing Marketing Campaingns: A/B Testing with ETL Pipeline in SQL

This project simulates a scenario where 2 versions of an email campaing (Group A vs B) in a marketing team, are evaluated to determine which performs better. SQL is used for ETL and Python for statistical analysis. Final Project coming soon!