# Introduction This is the unlabelled content of IMDB's movie reviews. The full dataset can be found here. [Stanford AI](http://ai.stanford.edu/~amaas/data/sentiment/). Since this is unlabelled, we can use this for our test set to predict if the reviews are `positve` or `negative`. Data Scraper script is here, [in this notebook](https://www.kaggle.com/venkatramani/imdb-datascraper) which reads the unlabelled reviews and saves as a `.csv` file as `testset.csv` # Data dimension This is a simple `n x 2` dimention dataset with `review` and `sentiment` as columns. `Sentiment` is as mentioned not labelled. # Task The task is to see if an effiecient classifcation algorithm can be developed.