Skip to content

This Streamlit app analyzes airline tweets and predicts their sentiment—Positive, Neutral, or Negative—using a Naive Bayes model trained on the Tweets.csv dataset.

Notifications You must be signed in to change notification settings

Somenpradhan/Airline-Tweet-Sentiment-Analyzer

Repository files navigation

✈ Airline Tweet Sentiment Analyzer

This Streamlit web application analyzes the sentiment of tweets directed at airlines. It uses a Naive Bayes classifier trained on the Tweets.csv dataset to predict whether a tweet is *Positive, **Neutral, or *Negative.

🚀 Features

  • 🔍 Predict sentiment of airline-related tweets (positive, neutral, negative)
  • 📊 Shows model performance on test data
  • 🧹 Automatic text preprocessing (cleaning, lemmatization, stopword removal)
  • 🧠 Naive Bayes model trained on real Twitter data
  • 🖼 User-friendly Streamlit interface

📁 Dataset

The model uses the Tweets.csv dataset from Kaggle:

  • 14,640 tweets about major U.S. airlines.
  • Labeled as positive, neutral, or negative.
  • Pre-filtered to include only samples with sentiment confidence ≥ 0.5.

About

This Streamlit app analyzes airline tweets and predicts their sentiment—Positive, Neutral, or Negative—using a Naive Bayes model trained on the Tweets.csv dataset.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published