NLP for Suicide and Depression Identification with Noisy Labels

A review of our recent work for using deep learning for suicide and depression classification in the presence of noisy labels

Image by Authors

This article is authored by Ayaan Haque and Viraaj Reddi

In this article, we will review our recent work titled “Deep Learning for Suicide and Depression Identification with Unsupervised Label Correction” by Ayaan Haque, Viraaj Reddi, and Tyler Giallanza. This paper is a method that uses deep learning for suicide vs depression identification in the presence of noisy labels, and is currently under review at a top conference. Our paper is available on ArXiv, the code is available on Github, and the project website is here. This article will review our method, titled SDCNL, as well as some of the results.

Overview

SDCNL, which stands for Suicide Depression Classification with Noisy Labels, is a method for distinguishing between suicide and depression using deep learning and noisy label correction. Early detection of suicidal ideation in depressed individuals can allow for adequate medical attention and support, which in many cases is life-saving. Recent NLP research focuses on classifying, from a given piece of text, if an individual is suicidal or…