README

FILES:
1) bigram.py: Language model classifier uses bigram model.
2) build_bigram_features.py: Creates feature vector (bigram frequency) for input to perceptron and scikit svm.
3) build_features_bigram_for_liblinear.py: Creates feature vector (bigram frequency) for input to LIBLINEAR.
4) build_features_for_liblinear_uni.py: Creates feature vector (unigram frequency) for input to LIBLINEAR.
5) build_unigram_features.py: Creates feature vector (unigram frequency) for input to perceptron and scikit svm.
6) svm_train_and_test.py: Scikit SVM classifier.
7) unigram.py: Language model classifier uses unigram model.

The folder contains a few sample csv files which are intermidiate dataset forms for SVM and Perceptron.