Personality & Sentiment Analysis (AI Project)

Project Overview

This project is an Artificial Intelligence Lab final project that performs Sentiment Analysis and Personality Insight extraction from textual data. The system fetches real-time tweets using the Twitter API, analyzes their sentiment using Natural Language Processing (NLP) and Machine Learning, and generates sentiment-based ratings from the extracted data.

---

Objectives

• Fetch live tweets at runtime using Twitter API
• Analyze tweet text for sentiment classification
• Generate sentiment-based ratings from analyzed tweets
• Demonstrate practical implementation of NLP and ML techniques

---

Twitter API Integration

• Tweets are fetched at runtime using the Twitter API
• User provides a keyword or hashtag
• Recent tweets related to the query are collected dynamically
• These tweets act as real-world input data for analysis

---

Dataset

• Source: Live Twitter data fetched at runtime
• Format: Text data extracted from tweets
• Usage: Tweets are preprocessed and passed directly to the trained model

---

Technologies & Libraries

• Programming Language: Python
• Libraries Used:
  • pandas
  • scikit-learn
  • re (Regular Expressions)
  • joblib
  • tweepy
• ML Model: Random Forest Classifier
• Feature Extraction: CountVectorizer (Bag of Words)

---

System Workflow

1. Authenticate with Twitter API
2. Fetch tweets based on a keyword or hashtag
3. Clean and preprocess tweet text
4. Convert text into numerical features using CountVectorizer
5. Apply trained machine learning model
6. Classify sentiment (Positive / Negative / Neutral)
7. Calculate overall sentiment score
8. Generate star-based rating from sentiment results

---

Rating Generation Logic

• Positive tweets increase rating
• Negative tweets decrease rating
• Neutral tweets maintain balance
• Final output is displayed as a sentiment-based rating (e.g., 1–5 stars)

---

Model Output

• Trained sentiment model saved using joblib
• Vectorizer saved for reuse
• Sentiment predictions for fetched tweets
• Overall sentiment rating based on live Twitter data

---

How to Run

1. Open the project notebook in Jupyter Notebook or Google Colab
2. Add your Twitter API credentials
3. Enter a keyword or hashtag
4. Run all cells sequentially
5. View sentiment analysis and generated ratings

---

Academic Purpose

This project was developed as part of the Artificial Intelligence Lab to demonstrate:

• Real-time data collection using APIs
• NLP-based text preprocessing
• Machine learning–based sentiment analysis
• AI-driven decision and rating systems

---

Future Enhancements

• Personality trait detection using Big Five Model
• Deep learning models (LSTM, BERT)
• Web-based dashboard
• Visualization of sentiment trends

---

License

This project is licensed under the MIT License.

---

Author

AI Lab Final Project
BSCS – Semester 6