🚀 Humor Detection

Welcome to our project repository for the Data Science class. Our primary goal is to develop a Natural Language Processing (NLP) model capable of recognizing humorous and sarcastic text.

🌌 Introduction

In the era of digital communication, understanding the sentiment and tone of the text is crucial. Our project aims to tackle a challenging aspect of this problem: detecting humor and sarcasm. We leverage various machine learning and NLP techniques to train our models on a diverse dataset.

🛸 Features

• Multiple Machine Learning Models: We have experimented with various machine learning models, including Logistic Regression, Decision Trees, K-Nearest Neighbors (KNN), Naive Bayes, Random Forest, and XGBoost.
• Neural Networks: We have also explored more complex models such as Convolutional Neural Networks (CNN), Gated Recurrent Units (GRU), and Long Short-Term Memory (LSTM) networks.
• N-grams Analysis: We have implemented N-grams analysis for feature extraction in text data.
• BERT and SHAP: We have a dedicated section where we experiment with BERT (Bidirectional Encoder Representations from Transformers) and SHAP (SHapley Additive exPlanations) for better understanding and interpreting our models.

🤖 Technologies Used

• Python: The primary language used for implementing the models.
• Jupyter Notebook: All our code and analysis are presented in Jupyter notebooks for better readability and reproducibility.
• PyTorch: Used for implementing and training our neural network models.
• Scikit-learn: Used for implementing traditional machine learning models.
• **Tensorflow: Used in implementing neural network models with CUDA

🌠 Future Improvements

• Model Optimization: We plan to further optimize our models for better accuracy.
• Feature Engineering: We aim to explore more sophisticated feature extraction techniques for text data.
• Expand Dataset: We plan to include more diverse data to make our models more robust.
• Deployment: We aim to deploy our models as a service for real-time humor detection.

🌍 Summary

This project is a part of our Data Science class, where we aim to tackle the challenging problem of humor and sarcasm detection in text. We have experimented with various machine learning models and NLP techniques. We welcome contributions and suggestions for improving our models and techniques.