Sentiment analysis, analyzing user’s textual reviews, to understand if a comment includes a positive or negative review (i.e. binary classification model). The dataset has been scraped from the tripadvisor.it Italian web site and contains 41077 textual reviews written in the Italian language.

In tripadvisor.ipynb different classifiers have been implemented : Decision Tree, Random Forest, Stochastic Gradient Descent (SGD) and Linear Support Vector Machine (SVM).

To handle the imbalance of the dataset, we apply cross-validation during the training and validation steps. This approach consists of dividing all the samples in groups of subsamples. Then the prediction function is learnt using k-1 folds and the remaining fold is used for testing. The default value for the number of folds is set to 10.

Overall the best accuracy (0.94) has been obtained by applying Linear SVM model.