In this notebook, we perform sentiment analysis on tweets talking about COVID-19 vaccines. We experiment with several data preprocessing techniques, such as Bag of Words, TF-IDF, GloVe and Word2Vec, and plot our results on them. In addition, we classify the tweets into three classes (Neutral, Positive, Negative) using various classifiers such as, Support Vector Machines (SVM), Random Forests and K-Nearest Neighbors (KNN). We provide a detailed comparison for each method we experimented with. We also try to beat the scores of our previous classifiers in the "Beat the benchmark” section. Finally, using Latent Dirichlet Allocation (LDA) we categorize our data and generate some topics from it.
In this notebook, we analyze data from taxi rides in the greater New York City area. We perform clustering with K-Means and regression with Random Forests Regressor to predict taxi trip duration. We also create a map of New York City with the pickup and drop-off points of the trips and perform some further data analysis focusing on traffic times.
*Equal contribution.