🔧 Keywords

NLP, PySpark, AWS, Big Data, Sentiment Analysis, TextBlob, Data Visualization, text analytics, dashboards, cloud tools

🧩 Problem

Understanding public sentiment on AI helps track trends — but analyzing thousands of tweets manually isn’t scalable.

This project uses large-scale text data to extract meaningful insights from social media conversations around AI.

⚙️ What I Did

• Processed a dataset of 10K+ pre-collected tweets using PySpark on Databricks to clean and structure the text for analysis.
• Applied TextBlob to assign sentiment labels (positive, neutral, negative) to tweets, since the data had no pre-existing annotations.
• Exported processed data to Amazon S3, and queried it via Amazon Athena for flexible, scalable analysis.
• Designed an interactive dashboard using Amazon QuickSight to visualize sentiment distribution, trends over time, and key word patterns.

📊 Outcome

• Built an end-to-end cloud-based NLP pipeline for sentiment analysis using big data tools.
• Analyzed 10K+ tweets to uncover public sentiment trends around AI.
• Gained hands-on experience with PySpark, AWS services, and visual storytelling with real-world social media data.

🛠️ Tech Stack: