Analisis Sentimen Terhadap Kesehatan Mental Selama Pandemi Covid-19 Berdasarkan Algoritma Naïve Bayes dan Deep Learning

Abstract

The Covid-19 pandemic is an epidemic that poses a threat to both physical and mental health. The mental health disorders that many people experience during the Covid-19 pandemic are depression, stress, and excessive anxiety. Some people use Twitter and other social media to voice their problems to lessen this effect. The goal of this research is to determine how people feel about mental health amid the Covid-19 pandemic on Twitter. The sentiment data will be analyzed based on assessment findings from the testing model using the Naïve Bayes and Deep Learning algorithms using RapidMiner, and it will be divided into groups of positive sentiment and negative sentiment. This research compares the performance of two algorithms to determine which one performs better while analyzing the public's sentiment toward mental health during the Covid-19 pandemic. According to the research findings, the Deep Learning algorithm performed better with accuracy scores of 86,46%, precision scores of 89,54%, and recall scores of 95,10% than the Naive Bayes algorithm compared to accuracy scores of 76,52%, precision scores of 87,97%, and recall scores of 83,66% at analyzing public sentiment towards mental health during the Covid-19 pandemic.