The Sentiment Analysis on Mental Health Awareness by Non-Governmental Organisation's Twitter

Abstract

The Ministry of Health Malaysia's statistical analysis showed increasing mental health problems among Malaysians. However, Malaysian society’s stigma causes ignorance of mental health problems and lack awareness of the issues. The main purpose of this study was to determine the Malaysian community's awareness of mental health issues using data from the Non-Governmental Organization Twitter. The data were taken from the Twitter application for NGOs related to mental health in Malaysia. NGOs often disseminate information such as health statistics, causes, and ways to manage mental health on their Twitter application. The data collected from the Twitter API application require permission and application from Twitter even though the data are accessible publicly. The study was implemented using experimental methods from which the sentiment analysis is an appropriate way to study the Malaysian community's awareness of mental health problems. A few experiments were conducted, such as data collection, preprocessing, Sentiment Analysis, machine learning techniques, Support Vector Machine (SVM), Neural Network (NN), and Naive Bayes (NB). The analysis showed that each NGO's total number of positive tweets was more than the number of negative snippets. The analysis of machine learning using the three techniques showed the highest percentage of positive data for Precision, Recall, and F1-Score. Therefore, the awareness of mental health problems should be created using more positive text posts by the NGOs on social media to educate people.