Homework 1: The Brand & Reaction Engine

You are a digital strategy consultant. A client has handed you a CSV file of their tweet history. They want to know: What do they talk about? Who should they sponsor? And can AI help them manage their feed autonomously?

Instructions

  1. Create a Streamlit web app titled "[Your Name] Strategic Tweet Engine" that processes a CSV file of tweets and includes the following 5 Tabs:

    1. Tab 1: The Leaderboard
      Display a clean data table of the user's tweets. The table must show tweets ranked specifically by Engagement Rate (Favorites / Views).
    2. Tab 2: The Activity Heatmap
      Build a Heatmap showing posting frequency:
      • X-Axis: Hour of Day (0-23)
      • Y-Axis: Day of Week (Mon-Sun)
    3. Tab 3: Topic Modeler
      Use AI to analyze the text of all of the tweets. Identify the top 5 "Core Pillars" (topics) of the account. Display the results in a table with two columns: Topic Name and Description.
    4. Tab 4: Brand Compatibility Agent
      Create an input box where the user types a Brand Name (e.g., "Nike", "Yale University"). Use AI to analyze the brand against all of the user's tweets and output a Compatibility Score (0-100%) along with a paragraph of strategic reasoning.
    5. Tab 5: The News Reactor
      Create an input box where the user enters a News Article URL. When a URL is provided, the app should scrape the article text using newspaper3k or a similar library, then use AI to generate a reactive tweet in the user's specific voice based on all of their tweets. Display the generated tweet in a styled "Tweet Card" UI that includes a dummy Avatar icon, dummy Name, and the tweet Text.

  2. The app should be robust, professionally styled, and error-free.

Submission Checklist

Your submission should include the following files zipped into a file called hw1.zip:

  1. app.py - the main app file
  2. requirements.txt - listing all dependencies

Grading Breakdown (100 Points)

Component Points Description
Foundation & Tab 1 5 App runs smoothly; CSV loads correctly; Tab 1 correctly sorts and displays the Leaderboard by Engagement Rate.
Tab 2: Heatmap Visuals 20 Heatmap correctly aggregates data by Day vs. Hour.
Tab 3: Topic Modeler 15 AI successfully extracts meaningful topics and displays them in a structured 2-column DataFrame (not just a blob of text).
Tab 4: Brand Agent 25 The AI provides a logical numeric score and specific reasoning connecting the Brand values to the User's specific tweet content.
Tab 5: News Reactor 30 App successfully scrapes a real URL, and uses AI to generate a relevant tweet in the user's voice, and renders it in the styled "Tweet Card" UI.
Proper Submission Zip 5 hw1.zip is submitted and correctly contains app.py and requirements.txt with the expected structure and filenames.