There are many datasets available freely on the web to perform text classification. The dataset that was used is the Cornell sentiment analysis dataset.
Link: [http://www.cs.cornell.edu/people/pabo/movie-review-data/ ]
In this page you will be able to find a lot of datasets and the one that was used is "polarity_dataset_v2.0" which was introduced in Pang/Lee ACL 2004.
This dataset contains 1000 positive and 1000 negative reviews.
The corpus was created by using regular expression formatting and was converted into Bag-Of-Words(BOW) model. This was converted into a TF-IDF model. The corpus was split into train and test data and several models were trained and were tested.
Accuracy -
Logistic Regression - 0.87
SVM - 0.84
Random Forest - 0.84