Sentiment-Analysis-Using-Neural-Network

• Discription:

  A full-stack web application for analyzing the sentiment of news headlines using machine learning techniques. The application allows users to input news headlines and receive predictions on whether the sentiment is neutral, positive, or negative.

• Functionality:

  • Developed a web-based interface using HTML, CSS, and Flask, providing a user-friendly platform for inputting news headlines.
  • Implemented a machine learning model using TensorFlow and Scikit-learn to predict sentiment based on headline content.
  • Utilized CountVectorizer for text vectorization and SMOTE for data oversampling to handle imbalanced classes.
  • Deployed the application on a Flask server, allowing for real-time interaction with users.
  • Enabled seamless integration between the front-end interface and the back-end machine learning model for efficient sentiment analysis.

• Technologies Used:

  • Front-End: HTML, CSS
  • Back-End: Flask
  • Machine Learning: TensorFlow, Scikit-learn
  • Data Processing: Pandas, NumPy
  • Deployment: Google Colab, pickle
  • Additional Libraries: Matplotlib, Imbalanced-learn