using svm, binary classification, text reviews using svm , multi classification using lexicon+machine learning approach --> used feedforward 2-layer ann along with polarity scores of each word in vocabulary using under sampling(SMOTE) to reduce non-uniformity of data distribution dataset used #amazon-mp3 dataset #"the vinci code " book reviews dataset #"amazon products" dataset ##"sorry for the late upload"##