Skip to content

Latest commit

 

History

History
11 lines (9 loc) · 1010 Bytes

README.md

File metadata and controls

11 lines (9 loc) · 1010 Bytes

NLP-Sentiment-Analysis For Suicidal Ideation Detection

This project (SAS) uses the Facebook FastText library to classify texts for the analysis of sentiments that show suicidal ideation. There is a standard tutorial on using FastText for text classification> The dataset used was acquired from the authors of the research, ‘Supervised Learning for Suicidal Ideation Detection in Online User Content’ Shaoxiong Ji et al., 2018 These datasets contain suicidal ideation texts collected from Reddit.

Results:

Image of the result of prediction carried out on some texts. Image of the comparison of SAS' result to another research's.