The purpose of the exercise is to practice different machine learning algorithms for text classification as well as the performance evaluation. In addition, you are requried to conduct 10 fold cross validation (https://scikit-learn.org/stable/modules/cross_validation.html) in the training.
The dataset can be download from here: https://github.com/unt-iialab/INFO5731_FALL2020/blob/master/In_class_exercise/exercise09_datacollection.zip. The dataset contains two files train data and test data for sentiment analysis in IMDB review, it has two categories: 1 represents positive and 0 represents negative. You need to split the training data into training and validate data (80% for training and 20% for validation, https://towardsdatascience.com/train-test-split-and-cross-validation-in-python-80b61beca4b6) and perform 10 fold cross validation while training the classifier. The final trained model was final evaluated on the test data.
Algorithms:
(1) MultinominalNB
(2) SVM
(3) KNN
(4) Decision tree
(5) Random Forest
(6) XGBoost
Evaluation measurement:
(1) Accuracy
(2) Recall
(3) Precison
(4) F-1 score