About this Repository ℹ️

This repository contains the project I did as a part of the coursework for ECS7005P - Risk and Decision-Making for Data Science and AI.
The assignment consisted of several individual questions that required me to solve probabilistic models, Bayesian networks, and decision analysis using Python and PyAgrum.

Key Takeaways 🔍

1. Data Analysis and Probability: Processed and categorised 3D printing data, computing key probabilities and statistical insights.

2. Machine Learning: Developed a regression model to predict material robustness, enhancing predictive analytics skills.

3. Bayesian Networks: Modelled quality assurance effectiveness and product reliability, leveraging PyAgrum for probabilistic reasoning.

4. Decision Optimisation: Designed influence diagrams to assess financial trade-offs in QA processes, improving strategic decision-making.

5. Causal Inference: Analysed Simpson’s Paradox in medical treatment data, highlighting the impact of confounding variables.

Environment 👩🏻‍💻

    

Stack 🛠️

Repository Structure 🌲

.
├── .gitattributes
├── 3d_printing.csv
├── RDMDSAI_CW1.ipynb
└── README.md

Reflection 🪞

This coursework was an invaluable opportunity to apply statistical and probabilistic models in decision-making under uncertainty. Through hands-on experimentation with real-world datasets, I deepened my understanding of Bayesian inference, machine learning, and decision optimisation. The challenges of modelling complex dependencies, evaluating risk, and making data-driven decisions reinforced the importance of combining theoretical knowledge with practical implementation.

This coursework has been instrumental in enhancing my data-driven decision-making skills and has motivated me to explore more advanced probabilistic AI techniques. The insights gained here will be foundational as I continue to build expertise in machine learning, uncertainty quantification, and AI-powered decision systems.