Data and Sentiment Analysis of Monkeypox Tweets using Natural Language Toolkit (NLTK)

In this study, the researchers aimed to gain insights into how people perceive and react to monkeypox by performing sentiment analysis on Twitter data related to the disease using natural language processing techniques, specifically the Natural Language Toolkit (NLTK). This study analyzed 95000 tweets about monkeypox that were collected using Tweepy between August 19, 2022, and September 3, 2022. The preprocessing of the tweets included formatting date columns, using regular expressions to fix the structure of the text, and removing unnecessary symbols. Feature extraction was also utilized to find typos and errors using NLTK SpellChecker() as well as word count and sentence length. NLTK SentimentIntensityAnalyzer (SIA) was used to determine the user sentiments for every tweet. Visualization libraries such as Plotly, Seaborne, and Matplotlib were used to display the analysis. Pearson Correlation was used to see correlations between the tweets and feature-engineered columns. The sentiment distribution was most neutral, then negative, and least positive. The sentence length distribution per day and the overall sentence length by language were analyzed. A word cloud was also generated. The results showed that the sentiment distribution was neutral and the sentence length was around 50 words. The word cloud revealed that the most frequently occurring words were related to the symptoms and transmission of monkeypox, as well as the measures being taken to control the disease.