Data process: Most of training and testing data was first processed in Excel, then parse and normalized into text.
Classifier performance: The influence of data processed with different methods was evaluated in SVM and KNN classifiers that are implemented from scikit-learn Python library.
Data analysis: Data visualization and analysis was done to observe the distribution of each data.