Movie Chatbot - NLP with BERT

This project aims to develop a Telegram Chatbot based on natural language processing using the BERT model. The model has been trained using movie data from Kaggle and can provide recommendations based on user preferences via Telegram.

Project Overview

In this project, a BERT-based model has been trained to analyze movie details and provide recommendations based on user queries.

Dataset used: Movies Details Dataset

Technologies Used

• Python
• BERT
• Natural Language Processing
• Telegram Bot API
• Pandas & NumPy
• Scikit-Learn
• Sentence-Transformers
• NLTK
• Regex

Dataset

The Movies Details Dataset from Kaggle was used to train the model. The dataset includes the following information:

• Movie Title
• Overview
• Vote Average

Telegram Bot

Our bot allows users to describe the type and content of the movie they want to watch via Telegram. Based on the provided criteria, the bot lists the most suitable movies and provides recommendations.

Installation & Execution

1️⃣ Install Required Dependencies

pip install numpy pandas nltk sentence-transformers scikit-learn regex python-telegram-bot

2️⃣ Download the Kaggle Dataset

Download the Movies Details Dataset and place it in the project directory.

3️⃣ Train the Model

python bert_vectorizer.py

4️⃣ Set Up the Telegram Bot Token

To run your Telegram bot, enter the TOKEN obtained from BotFather in the telegram_bot.py file.

5️⃣ Start the Telegram Bot

python telegram_bot.py

Example Usage

User: "Recommend a science fiction movie featuring time travel and parallel universes."
Bot: "See You Yesterday (2019) - Two teenage science prodigies spend their time inventing time machines..."

Contributing

If you would like to contribute to the project, feel free to submit a pull request or open an issue.