Next Word Predictor using LSTM with Flask

This project is a Next Word Prediction application built using an LSTM (Long Short-Term Memory) neural network.
It predicts the next likely word based on the input sentence provided by the user.
The backend is powered by Flask, and the frontend is built with Tailwind CSS for a clean, responsive UI.

🌟 Features
✅ Predicts the next word of a given sentence
✅ Built with LSTM using TensorFlow/Keras
✅ Top-5 accuracy metric included for better model evaluation
✅ Temperature sampling for diverse predictions
✅ Responsive web interface

🖥️ Environment Used
Tool/Library----------Version
Python 3.8+
TensorFlow 2.x
Flask 2.x
Numpy Latest
Tailwind CSS CDN Included

🚀 How It Works

1. Dataset of English sentences → tokenized and converted to sequences
2. Model Training happens (if model isn’t already saved)
3. On web interface, input text → /predict endpoint → next word generated
4. Displayed in the browser

👨‍💻 Author & Contact
Name: Afsar Tanvir
📫 Email: [email protected]
🌐 GitHub: https://github.com/AfsarTanvir
💼 LinkedIn: https://www.linkedin.com/in/afsar-tanvir-81ab17270/