CAMS: An Annotated Corpus for Causal Analysis of Mental Health Issues in Social Media Posts

TL;DR

This paper introduces CAMS, a new annotated dataset for causal analysis of mental health issues in social media posts, enabling better interpretability and categorization of causal factors.

Contribution

The paper presents a novel annotation schema for causal analysis, applies it to Reddit and existing datasets, and releases a publicly available dataset with baseline model results.

Findings

Logistic Regression outperforms CNN-LSTM by 4.9% accuracy on CAMS.

CAMS dataset combines Reddit and existing data for causal analysis.

The schema improves interpretability of causal factors in mental health social media posts.

Abstract

Research community has witnessed substantial growth in the detection of mental health issues and their associated reasons from analysis of social media. We introduce a new dataset for Causal Analysis of Mental health issues in Social media posts (CAMS). Our contributions for causal analysis are two-fold: causal interpretation and causal categorization. We introduce an annotation schema for this task of causal analysis. We demonstrate the efficacy of our schema on two different datasets: (i) crawling and annotating 3155 Reddit posts and (ii) re-annotating the publicly available SDCNL dataset of 1896 instances for interpretable causal analysis. We further combine these into the CAMS dataset and make this resource publicly available along with associated source code: https://github.com/drmuskangarg/CAMS. We present experimental results of models learned from CAMS dataset and demonstrate that a classic Logistic Regression model outperforms the next best (CNN-LSTM) model by 4.9\% accuracy.