My Projects
Python
An exploratory data analysis project on 1,000 simulated student records mapping daily lifestyle habits against final exam scores. Unlike previous EDA projects, this dataset is designed around a single actionable question — what actually determines how well a student performs academically? Every chart directly answers a piece of that question, making this one of the most story-driven projects in the portfolio.
Python
An exploratory data analysis project on 1,465 Amazon product listings across 9 categories. The dataset includes real pricing, discount, rating, and review data — making it ideal for understanding Amazon's pricing strategy, discount patterns, and customer behavior. Unlike previous EDA projects, this one focuses on business insights rather than scientific or behavioral patterns.
Python
An exploratory data analysis project on real medical data from the Breast Cancer Wisconsin Diagnostic Dataset. The dataset contains 569 breast mass samples with 30 numerical features derived from digitized fine needle aspirate (FNA) images. Unlike the previous EDA projects, this dataset has genuine predictive signal — every chart tells a clear, meaningful story about what distinguishes a malignant tumor from a benign one.
Python
An exploratory data analysis project on 8,000 Spotify users built around a single business question — why do users churn? The dataset includes behavioral, demographic, and subscription data with a target variable
is_churned, making every chart directly tied to a real retention insight
Python
An exploratory data analysis project on 120+ years of Olympic history using Python, pandas, matplotlib, and seaborn. The dataset contains 70,000 athlete-event records spanning from 1896 to 2016, covering both Summer and Winter Olympics with detailed athlete demographics, event information, and medal outcomes.
Python
A data cleaning project on a Glassdoor data science jobs dataset with 672 web-scraped job listings. The dataset had no traditional null values, making it a more investigative cleaning challenge than typical null-heavy datasets.
Python
A data cleaning project on a synthetic cafe sales dataset with 10,000 transactions. The dataset contained fake null values disguised as strings, wrong data types, and high volumes of missing data.
Python
A data cleaning project on a retail store sales dataset containing 12,575 transactions across multiple product categories, locations, and payment methods
Made with Bullet