video1.0<iframe src="https://www.loom.com/embed/93394e1094d84ddbb0c60c3daaaa3843" frameborder="0" width="1920" height="1440" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14401920Loomhttps://www.loom.com14401920https://cdn.loom.com/sessions/thumbnails/93394e1094d84ddbb0c60c3daaaa3843-d88e9de12f87429c.gif330.748In this video, I walk you through the data preparation process after exploratory data analysis. I discuss merging user and transaction tables, handling categorical variables, and the importance of creating new categorical variables to improve our fraud detection model. I also explain how I split the data into training, validation, and test sets based on time points, specifically focusing on transactions after May 1st. Please review the splitting strategy and let me know if you have any questions!