This project is a simple machine learning experiment where I trained a model using the Go-Emotions dataset to detect emotions in text.
I made this project to learn more about how machine learning and natural language processing work. The idea is to take a piece of text and try to predict what emotion it shows. I built this mainly to practice and understand how models are trained, so it is not meant to be a professional or production system.
The model was trained using the Go-Emotions dataset. This dataset contains many Reddit comments that are labeled with different emotions. The model learns from these examples and then tries to guess the emotion in new text.
This model is not perfect and it can make mistakes. Since I made this as a fun learning project, there might be:
Wrong predictions
Emotions that are classified incorrectly
Problems understanding the full context of a sentence
So it should not be used for anything serious or important.
The main purpose of this project is:
Learning about NLP and emotion classification
Practicing how to train machine learning models
Understanding how datasets and training pipelines work