Analysis of Glassdoor data science job postings using PostgreSQL and R to identify the most in-demand skills, their relationship to salary, and variation across sectors.

This assignment demonstrates how HTML and JSON data can be manually created and imported into R. Both files were loaded into data frames and compared to verify that they contain the same information.

This project demonstrates how a wide-format dataset can be cleaned and transformed into tidy format using the tidyverse in R. A global country dataset containing economic, health, and demographic indicators was standardized, reshaped using pivot_longer, and analyzed to explore relationships between key variables. The analysis focuses on examining the relationship between GDP and life expectancy across countries through summary tables and visualizations.

This analysis was completed as part of Project 2 and investigates student academic performance across different types of assessments. The dataset was cleaned and transformed from a wide format into a tidy structure using the tidyverse in R, allowing for comparison of average scores across assessment categories and academic departments.

This analysis was completed as part of Project 2 and investigates laptop pricing across different manufacturers and models. The dataset was cleaned and standardized to compare average prices and identify pricing differences between brands.

This analysis applies the Elo rating formula to compare expected and actual tournament performance and rank players by deviation from rating-based predictions.

This analysis examines airline delay percentages overall and by city using tidy data principles in R. The results demonstrate how aggregated statistics can differ from grouped comparisons.

The objective is to develop an R Markdown script that converts chess tournament results from a text file into a CSV suitable for database import, including player name, state, total points, pre-rating, and average opponent pre-rating.

This project applies window functions to daily cryptocurrency price data for Bitcoin and Ethereum from September 2024 through February 2026. Using structured time-series data, a year-to-date cumulative average and a six-day moving average were calculated for each cryptocurrency. The year-to-date average resets at the beginning of each calendar year, while the six-day moving average uses a rolling window over the current observation and the previous five days. These calculations demonstrate how window functions can be used to compute dynamic summary measures within grouped and ordered data.

Implementation of a Global Baseline Estimate (GBE) recommender system in R using survey-based movie ratings. The model computes global, user, and movie bias terms to generate one recommendation per user.

Evaluating Classification Model Performance

Basic Data Loading and Transformation