Archives: Projects

This project focuses on predicting the onset of diabetes using logistic regression on
the Pima Indians Diabetes Database from Kaggle. The dataset includes variables such as glucose
concentration, blood pressure, BMI, and age. The model achieved an AUC value of 0.8396 and
an overall accuracy of 78.39%, effectively identifying key risk factors associated with diabetes.

This project explores Ridge and Lasso regression techniques in predicting college
graduation rates using the College dataset. Ridge regression slightly outperformed Lasso
regression in terms of predictive accuracy, while Lasso offered more interpretable results by
performing feature selection.

This project provides a comparative analysis of Principal Component Regression
(PCR) and Partial Least Squares Regression (PLS) for predicting benzene (C6H6) concentrations
using the Air Quality Dataset from the UCI Machine Learning Repository. The primary goal is to
address multicollinearity and dimensionality reduction to improve predictive accuracy. Results
showed that PLS had superior performance with RMSE of 0.972 and R-squared of 0.974, while
PCR had a higher RMSE of 1.572 and R-squared of 0.933.

The A/B Testing Project aims to explore and analyze the effectiveness of a new
design variant compared to an existing one through rigorous statistical analysis and
experimentation. By leveraging user interaction data, the project seeks to uncover actionable
insights into various metrics such as completion rates, time spent on steps, error rates, and
abandonment rates. Through data preparation, exploration, analysis, and statistical testing using
Python libraries such as scipy and statsmodels, the project determines whether the proposed
design changes lead to meaningful improvements in user engagement and overall user
experience.

A comprehensive data analysis project focused on the FIFA World Cup 2022,
sourced from HiCounselor.com. The project involved advanced SQL techniques, including
complex joins, subqueries, Common Table Expressions (CTEs), and stored procedures, to
analyze player and team performance, match outcomes, and key tournament insights.

This project focuses on analyzing the top 50 charts of all Spanish-speaking
countries (except Cuba) for a specific week. The analysis includes data collection from Spotify,
data storage and management using SQL, detailed data analysis through SQL queries in Python,
and visualization using Tableau.

This project provides an in-depth analysis of the electric vehicle (EV) landscape in
Washington State using Tableau. It covers data from the Washington State Department of
Licensing, detailing the distribution, growth, and characteristics of Battery Electric Vehicles
(BEVs) and Plug-in Hybrid Electric Vehicles (PHEVs).

This Streamlit app predicts monthly rent prices for apartments across various states
in the U.S. The project involved data cleaning, exploratory data analysis, and applying machine
learning models to identify the best predictors. The app allows users to input apartment features
and compare rental price distributions and averages across states.

Cryptocurrency Explorer is a comprehensive application built to provide real-time data tracking, historical analysis, and predictive modeling of cryptocurrency prices. It leverages technologies like Streamlit to offer a user-friendly experience for both casual enthusiasts and data professionals alike. Users can track real-time cryptocurrency prices, analyze historical trends, and predict future prices using the ARIMA forecasting model.