joudahmedcodes/Customer-Service-chatbot

my customer service chatbot project for IctHub Egypt summer training # Intelligent FAQ Chatbot using NLP & Machine Learning An NLP-powered chatbot built using Python, TF-IDF vectorization, and Logistic Regression for intent classification. The chatbot preprocesses user input, predicts intent categories, and responds through an interactive Gradio interface. ## Features - NLP text preprocessing - Tokenization and stemming - Stopword removal - TF-IDF vectorization - Intent classification using Logistic Regression - Unknown input detection - Interactive chatbot GUI using Gradio ## Technologies Used - Python - Pandas - NumPy - NLTK - Scikit-learn - Gradio - Matplotlib ## Workflow 1. Load and clean dataset 2. Preprocess text - Lowercasing - Tokenization - Stopword removal - Stemming 3. Convert text using TF-IDF 4. Train Logistic Regression model 5. Predict user intent 6. Generate chatbot response 7. Display through Gradio interface ## Model Performance - Model: Logistic Regression - Vectorizer: TF-IDF - Accuracy: 96.64% ## Future Improvements - Deploy as a web application - Convert model to TensorFlow Lite - Add speech recognition - Improve intent detection - Add deep learning models - Connect to a database - Deploy using Flask or FastAPI - Add multilingual support