Machine Learning Driven Mental Stress Detection on Reddit Posts Using Natural Language Processing

Abstract

People’s mental conditions are often reflected in their social media activity due to the internet's anonymity. Psychiatric issues are often detected through such activities and can be addressed in their early stages, potentially preventing the consequences of unattended mental disorders like depression and anxiety. In this paper, the authors have implemented machine learning models and used various embedding techniques to classify posts from the famous social media blog site Reddit as stressful and non-stressful. The dataset used contains user posts that can be analyzed to detect patterns in the social media activity of those diagnosed with mental disorders. This paper uses different NLP (Natural Language Processing) tools such as ELMo (Embeddings from Language Models) word embeddings, BERT (Bidirectional Encoder Representations from Transformers) tokenizers, and BoW (Bag of Words) approach to create word/sentence data that can be fed to machine learning models. The results of each method have been discussed. The results achieved a top F1 score of 0.76, a Precision score of 0.71, and a Recall of 0.74 using only the preprocessed texts and machine learning algorithms to classify the posts. The results achieved by this paper are significant and have the potential to be applied in real-world scenarios to analyze mental stress among social media users. Although this paper focuses on data from Reddit, the techniques used can be transferred to similar social media platforms and could help solve the growing mental health crisis.

Exploring Mental Stress Expressions in Online Communities: A Subreddit Analysis

“…All other subreddits had percentages below 10%. Redditors' scores for a post ranged from 0 to 16,029 (median [IQR]: 3 [2][3][4][5][6][7][8][9][10][11][12][13]). Four intervals ([0, 1], [2,3], [4,21], and [22,16029]) of the Redditors' scores accounted for the following percentages of total posts: 24.74% (15,043 posts before January 2020 and 18,761 posts after), 27.41% (16,106; 21,346), 29.11% (18,404; 21,373), and 18.74% (14,617; 10,988).…”

“…On social media, stressed users usually post about exhaustion, losing control, increased self-focus, and physical pain, while non-stressed users focus on topics such as breakfast, family time, and travel [3]. Moreover, features of social media (e.g., comments, likes) highly reflect the users' mental stress [4,5]. Hence, analyzing public opinion and sentiment within these online communities can offer essential insights into the prevalence and dynamics of mental stress, inform targeted mental health interventions and support strategies, and provide valuable insights into how mental stress is perceived and managed within distinct online communities.…”

TherapyBot: a chatbot for mental well-being using transformers

“…Dataset may play a very important role here as the core learning is solely dependent on it. Anger-based evaluation [35] or stress detection using social media posts [36] can also be seen as an extension to similar problems based on the availability of good datasets.…”

Natural Language Processing for Work-Related Stress Detection Among Health Professionals: Protocol for a Scoping Review

“…This approach could be efficient for the early detection and diagnosis of PTSD, especially in large volumes of patient narratives [ 24 ]. NLP could also be used as an opportunity to transform mental health conversation and help early intervention for depression [ 25 ]. It is indicated to detect stress and mental disorders in an early stage [ 25 ].…”

“…NLP could also be used as an opportunity to transform mental health conversation and help early intervention for depression [ 25 ]. It is indicated to detect stress and mental disorders in an early stage [ 25 ]. However, the use of NLP extends not only to the detection of stress in patients but also to the field of health professionals.…”