I have a batch of raw Facebook posts and comments that needs to be turned into clean, structured information and then explored for insights. The files arrive in a messy CSV export; before any real analysis can happen, everything must first be entered or corrected so each row lines up with its timestamp, author, and message content. Once the text is tidy, I’d like you to dig into it for recurring themes, sentiment trends, word-frequency rankings, and any notable spikes in engagement over time. A concise visual snapshot—simple bar charts or word clouds—is enough; I’m more interested in the clear takeaway points than polished infographics. Deliverables • A cleaned spreadsheet (Excel or CSV) ready for future use • A short written summary (PDF or DOC) outlining key findings, insights, and visuals • The script or notebook if you automate the cleaning/analysis in Python, R, or a similar tool Accuracy in the data entry stage is critical—misaligned rows or dropped characters will throw off the whole analysis—so I’ll run a quick spot-check expecting at least 99 % correctness. Let me know which libraries or software you prefer; I’m comfortable with pandas, NLTK, tidytext, Excel Power Query, or anything equivalent. Right now the scope is limited to Facebook, but if this first run goes smoothly there’s potential to expand into Twitter and Instagram data later.