Abstract:
Coronavirus is a group of viruses that can infect both animals and humans. Coronavirus
usually causes respiratory tract infections in humans, ranging from coughs, colds to
Middle East Syndrome (MERS), and Severe Acute Respiratory Syndrome (SARS). Many
studies have been conducted to determine sentiment, especially regarding the effect of
delivery services during the COVID-19 period. This study aims to compare the
presentation of public sentiment regarding GoFood services in the early days of the
pandemic and the decline of COVID-19 and to determine the level of accuracy in
conducting sentiment analysis. This study used the Naïve Bayes method which is useful
for classifying data and used a Confusion Matrix to see how well the accuracy level of
the classification results is. In the process of data classification is divided into two,
positive and negative. The final result of this research is that at the beginning of the
pandemic, the sentiment was in the form of a positive sentiment was 52% and a negative
value was 48%. Then during the pandemic decline, positive sentiment was 74% and a
negative value was 26%. Based on this process, the accuracy rate at the beginning of the
pandemic was 82.67% and during the decline in the pandemic, the accuracy level was
70.58%. So analyzing sentiment data using the Naïve Bayes method achieved satisfactory
results.
