Twitter Sentiment Analysis of Recession 2023: A Comparative Study of Machine Learning Approaches
DOI:
https://doi.org/10.25124/jrsi.v11i01.612Keywords:
Bernoulli Naïve Bayes,, Decision Tree, K-Nearest Neighbors, Linear Regression, Sentiment Analysis, Support Vector MachineAbstract
Sentiment Analysis helps understand public opinion on a particular topic. One recent topic that has attracted
attention is the potential for a global recession in 2023. In this study, five different algorithms - Bernoulli
Naive Bayes (BNB), Support Vector Machine (SVM), Linear Regression, K-Nearest Neighbors (KNN), and
Decision Tree - were compared to determine which algorithm provided the most accurate sentiment analysis
of Twitter data related to this topic. The results showed that the SVM algorithm had the highest accuracy,
and most Twitter users had negative sentiments towards topics related to a potential recession in 2023, with
a prediction rate of 81.7% compared to 16.3% for positive sentiments. The results of this study are expected
to be used to understand the general public's viewpoints regarding the predicted recession in 2023 and to
provide insights for developing policies and strategies to mitigate the economic downturn.
