This GitHub repository houses a comprehensive project focused on multiclass text classification of emotions. The baseline model employs TF-IDF vectorization and features a diverse set of classifiers, including Multinomial Logistic Regression, Decision Tree, Random Forest, Naive Bayes Classifier, and Linear SVM. In the pursuit of refining the model's performance, the final iteration incorporates a Long Short-Term Memory (LSTM) neural network with word embeddings serving as the vectorizer. This project aims to provide a robust framework for emotion classification in textual data, offering insights into the effectiveness of various machine learning algorithms and the potential enhancements gained through the utilization of deep learning techniques.
- Multiclass text classification of emotions
- TF-IDF vectorization as the baseline approach
- Diverse set of classifiers: Multinomial Logistic Regression, Decision Tree, Random Forest, Naive Bayes Classifier, and Linear SVM
- Final model: LSTM with word embedding as the vectorizer
Feel free to explore the codebase, experiment with different models, and contribute to the advancement of emotion text classification.