Big Data and Mental Health: Social Media Sentiment Analysis

With mental health problems affecting 1 in every 8 people globally, leveraging technologies such as big data and sentimental analysis techniques for detection and monitoring is crucial. This paper explores how big data and sentiment analysis techniques can be used to predict mental health issues from posts of social media users. The research finding shows that sentimental analysis is an innovative solution for detecting and monitoring mental health issues among social media users through the collected big data sets. Key findings indicate that lexicon-based approaches such as VADER and machine learning models like SVM and Naïve Bayes present potential advantages and limitations in identifying and monitoring mental health issues. Also, hybrid approaches that synergises machine learning methods and lexicon-based approaches presents potential benefits in addressing mental health concerns among social media users. Current advances in sentimental analysis techniques for identifying and monitoring mental health issues from various data sources are also presented in the paper. The research also identified challenges such as inherent subjectivity in labelling, interpretational complexities, data quality and privacy concerns. Looking forward, future researchers should explore multilingual sentiment analysis, cross-platform analysis, integrate wearables and mobile apps and conduct longitudinal studies. In summary, while sentimental analysis presents promising avenue for mental health detection, there is need for continuous refinement and ethical adherence.