NLP: Suicide Ideation Detection Using Tweets
I develop a model to identify potential suicide ideation in tweets using sentiment analysis.
By using a pre-trained DistilBERT model, I preprocess and clean text data, fine-tune the model, and evaluate its performance using metrics like accuracy and F1 score.
Additionally, I use SHAP to interpret the model’s predictions, shedding light on which parts of the tweets influence the classification of harmful content.