Classification models comparison for a Sentiment Analysis task on a Twitter dataset by Nedescu I. C., Parolin A

Data Science Lab course @ Politecnico di Torino [Project Year 2021/2022]

In this competition for the exam course, we propose an entire classification pipeline, after a short exploration of the dataset, we apply a few pre-processing steps and use the clean data to feed different classification algorithms in order to compare their results based on the F1-score metric.