{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/853bf8dd692e4cb39e7170b4d11c9f3b\" frameborder=\"0\" width=\"1728\" height=\"1296\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1296,"width":1728,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1296,"thumbnail_width":1728,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/853bf8dd692e4cb39e7170b4d11c9f3b-78d508dad3dbe8cc.gif","duration":268.651,"title":"Data Analysis and Feature Engineering Insights from TTC SCADA Delay Records","description":"In this video, I walk through my contributions to the TTCC, focusing on phase two, which involves EDA data analysis, and phase three, feature engineering tasks using the consolidated dataset prepared by Hussain. I analyzed a TTC SCADA delay CSV file containing 12 years of data, revealing key insights such as peak delay hours and significant trends in incident behavior. I also performed feature engineering, creating time-based features and a new target variable, delayBin, which categorizes delays into low, medium, high, and severe. This transformed our predictive problem into a structural classification problem. I encourage viewers to review the EDA charts saved in our project repository and consider the implications of the findings for our modeling phase."}