Sentiment Analysis

Classification of books reviews, the first step is a binary classification(positive or negative), the second step is multiclass classification, contain four grades (1,2,4,5)

Data

Amazon provide a complete information of score of different productos, there are two database, the first contain a binary classification and the second a multiclass classification. link data

Usage Data

The format of the data is tar, in Python is possible decompress, so:

import tarfile

file = tarfile.open('processed_acl.tar.gz')
file.extractall('./data_reviews')
file.close()

My Page

diego51alejo