Ebenezer Addo-Kufuor

Computer Science graduate specialising in data science, machine learning, and FinTech applications. Experienced in developing predictive models for financial forecasting, risk assessment, and fraud detection, using large datasets to extract meaningful insights. Passionate about leveraging quantitative analysis and statistical modelling to improve decision-making and drive innovation. Continuously curious about uncovering patterns in data, optimising processes, and exploring new techniques to enhance performance. Dedicated to solving complex data challenges with a problem-solving mindset, ensuring impactful and measurable outcomes through innovative, data-driven approaches that create value in dynamic environments.

Applied AI – AI Model Implementation (Oct 2024 – Jan 2025)

• Developed and implemented an AI model using Neural Networks, Regression, and Deep Learning for a real-world application.
• Conducted data preprocessing, feature extraction, and model evaluation to assess performance and reliability.
• Created a Jupyter Notebook showcasing model implementation, visualisations, and evaluation metrics.
• Delivered a presentation explaining the rationale behind AI techniques used and their effectiveness.
• Technologies: Python, Jupyter Notebook, Neural Networks, Data Preprocessing, Visualisation

Applied Robotics – Reinforcement Learning for Autonomous Taxi Navigation (Dec 2024 – Jan 2025)

• Designed a robot taxi driver using Q-learning reinforcement learning to navigate an unknown environment.
• Implemented a Neural Network (MLP) model to predict optimal actions based on training data.
• Simulated real-world navigation in a 5x5 grid environment with rewards/penalties for decision-making.
• Analysed results using performance metrics, visualisations, and efficiency comparisons.
• Technologies: Python, Reinforcement Learning, Q-Learning, Neural Networks, Gymnasium API

Data Engineering 2 – Image Retrieval & Sentiment Analysis (Sep 2024 – Dec 2024)

• Developed a scalable NoSQL database (MongoDB) for storing image metadata and sentiment-labelled text.
• Built a Content-Based Image Retrieval (CBIR) system using OpenCV & feature extraction for AI-powered searches.
• Implemented a Sentiment Analysis model to classify text reviews into positive, neutral, or negative categories.
• Created interactive data visualisations to showcase sentiment distribution and retrieval performance.
• Technologies: Python, OpenCV, MongoDB, TensorFlow, TF-IDF, NLP, Machine Learning

Predicting Exchange Rates using MLP Neural Networks (Jul 2024 - Aug 2024)

• Developed a forecasting model using Multi-Layer Perceptron (MLP) Neural Networks to predict USD exchange rates.
• Preprocessed time series data, implemented data normalisation, and optimised network architecture (5-10, 5-3 layers).
• Evaluated model performance using RMSE, MAE, MAPE, sMAPE, and visualised results with ggplot2.
• Technologies: R, Neural Networks, Data Normalisation, ggplot2, Model Evaluation Metrics