View on GitHub ↗
Exploratory Data Analysis - E-Commerce Dataset
An in-depth EDA of the Olist Brazilian e-commerce dataset, uncovering key drivers of customer reviews, delivery performance, and revenue trends.
A curated collection some of my Python dashboards. Click into a section to explore the work.
No projects match your filters.
Try clearing a filter or searching a different term.
View on GitHub ↗
An in-depth EDA of the Olist Brazilian e-commerce dataset, uncovering key drivers of customer reviews, delivery performance, and revenue trends.
View on GitHub ↗
An EDA of NYC Airbnb listings combining price outlier analysis, geospatial mapping, and host behaviour clustering to uncover what drives pricing across the city.
View on GitHub ↗
An EDA of the Telco Customer Churn dataset uncovering the contract types, tenure patterns, and pricing behaviours that most strongly drive customer churn.
View on GitHub ↗
This is a Python project that explores the famous Titanic Dataset.
View on GitHub ↗
This is a Python project around PharmaTech who aims to create a Predictive model.
View on GitHub ↗
This is just a small Python code to simplyfy using multiple ML algorithms on a basic dataset with some useful visuals to interpret the models performance against each other.
View on GitHub ↗
A regression project comparing Linear Regression, Random Forest, Gradient Boosting, and Ridge models on scikit-learn's diabetes dataset to predict disease progression, finding that plain Linear Regression performs best due to the dataset's largely linear signal.
View on GitHub ↗
Binary classification project detecting fraudulent credit card transactions from anonymized PCA-transformed features, comparing tuned Logistic Regression, Random Forest, and XGBoost models under severe class imbalance (~0.17% fraud).
View on GitHub ↗
This project uses K-Means clustering on mall customer data to segment shoppers into distinct groups based on age, annual income, and spending score, using the elbow method to determine the optimal number of clusters and visualizing the resulting segments in 2D and 3D scatter plots.
View on GitHub ↗
This project applies PCA, t-SNE, and UMAP to a 20,000+ gene RNA-Seq dataset to visualize cancer subtype clusters and compares classifier performance on raw versus PCA-reduced features.
View on GitHub ↗
A binary classification project comparing linear and RBF-kernel Support Vector Machines on a non-linearly separable dataset, using grid search to tune hyperparameters and visualizing the resulting decision boundary.
View on GitHub ↗
A reinforcement learning agent (PPO) learns to play a custom-built Breakout clone purely through trial and error, improving from random flailing to reliably clearing bricks.