COVID-19 Sentiment Analysis using K-Means and DBSCAN

Abstract

The analysis of sentiment towards COVID-19 plays a crucial role in understanding public opinion. This research paper proposes sentiment analysis using K-means and DBSCAN clustering algorithms on the dataset of tweets related to COVID-19. Pre-processing and extraction of features is carried out using Term Frequency-Inverse Document Frequency (Tf-idf) to capture the weight of words in the dataset. K-means clustering is explored to group similar sentiments together, enabling the identification of sentiment clusters related to COVID-19. The DBSCAN algorithm is then employed to identify outliers and noise in the sentiment clusters. The evaluation metrics considered were accuracy, recall, F1-score, and precision. It was observed that DBSCAN was more effective in identifying underlying patterns in the data more accurately.