I conducted this case study to demonstrate how I can transform a messy dataset into actionable insights. I chose the Red Wine Quality dataset, which originally contained issues such as inconsistent column names, missing values, and unscaled variables. My goal was to walk through the full data science workflow, from cleaning to modeling to interpretation.

The study found a total of 4257 marine litter from different seagrass bed of ANIs, which later grouped into different categories based on different metrics. This Sankey diagram illustrating the flow and relative contribution of marine litter across 12 different Meadows of ANIs.