I already have a clean CSV file of YouTube comments taken from the animated movie trailer “Jumbo.” What I now need is a full sentiment study that goes beyond simply tagging each line as positive or negative—I want an insightful map of public opinion trends over time. The core of the project is a Random Forest–based classifier. Please build, tune, and evaluate that model in Python (pandas, scikit-learn, matplotlib/Seaborn are fine; feel free to add NLTK or SpaCy for text prep). After training, apply it to all comments, aggregate the predictions, and surface the key patterns: • How overall sentiment shifts by upload date, time of day, or any other temporal markers present • Which words or phrases drive the strongest swings, based on feature importance from the Random Forest • Visual dashboards or static charts that clearly communicate the evolving mood around the trailer Deliverables I expect: 1. Annotated Jupyter notebook or .py script that can be re-run end-to-end on my CSV 2. A short PDF or slide deck summarizing findings with the visuals embedded 3. The updated CSV containing each comment’s predicted sentiment label If something in the data needs cleaning you can handle that in-code. Otherwise, assume the CSV structure is stable and ready. Looking forward to seeing how the audience really feels about “Jumbo.”