How can i cluster a textual data as a positive or a negative sentiment using flat k means clustering and then store it as a dataset in which i have separate negative and postivie views just like movie_reviews dataset in corpus.I need such kind of dataset for doing further sentiment analysis.
You must be logged in to post. Please login or register an account.
Bag of words model, then cluster. Clustering is really mostly meant to be unsupervised. You'd cluster to 2 means, and just see if it happened to be positive and negative as you hope. Maybe it works, maybe not. You can then try to tweak features a bit and see if you can come up with a way to cluster to pos and neg, but clustering is really meant more for discovery, less for a classification task where you already know what you hope the attributes of the two groups to be.
-Harrison 8 years ago
You must be logged in to post. Please login or register an account.