AI for Data & Documents · Part II · Policy documents
Topic 4

From Documents to Analysis

Deeper text-as-data work: sentiment and tone (e.g., hawkish vs. dovish), tracking how language changes over time or across sources, and synthesizing a stack of documents into an evidence-based summary — always with citations back to the original so every claim can be checked.

Open Claude Download prompts (.txt) PDF slides
1

Use the same team you used in Topic 3 (files named Doc1, Doc2, Doc3).

2

Open Claude in a new tab and upload your team's documents.

3

Work through tone, a word-count score, change over time, and synthesis.

4

Enforce the citation rule: every claim cites a document and a quote.

PDF slidesDownload the slides for this topic (PDF)

Prompts to copyCopy one, paste it into Claude with your team's documents attached

Run it in Claude

  • 1. Use your Topic 3 team's documents (Doc1, Doc2, Doc3).
  • 2. Open Claude and start a new chat; attach the documents.
  • 3. Copy a prompt on the left, paste, and send.
  • 4. Trace at least two citations back to the source yourself.
Open Claude in a new tab

Tip: put the two tabs side by side so you can copy here and paste there without losing your place.