Sentiment Classification on Suicide Notes Using GPT, Bi-LSTM and CNN

In the modern day, where sentiments and feelings are important in every element of our daily lives, from personal to professional, sentiment classification has become a crucial field of research. This entails learning what people think about different products and how they feel about a specific topic. Relevant research has demonstrated that several models and techniques have been applied to the sentiment classification of suicide notes. Still, the state-of-the-art in this field was restricted to lower accuracy datasets. As a result, for the purpose of this study, we taken dataset for sentiment classification tasks on suicide notes utilizing information from Twitter, Reddit, blogs , and various Infomedia websites.Sentences are divided into six categories based on suicidality, including Proud, Love, Neutral, Sad, Hate and Happy. According to our results, the suggested model has achieved a 77% F1 score, a 78% recall, and a 77% precision.