Text Mining: Sentiment Classifier
Jump to navigation
Jump to search
Sentiment Classifier menggunakan Word Sense Disambiguation menggunakan WordNet dan statistik terjadinya kata dari corpus movie review NLTK. Mengklasifikasikan ke dalam kategori positif dan negatif.
Persyaratan
- Python 2.6.
- NLTK http://www.nltk.org 2.0
- NumPy http://numpy.scipy.org
- SentiWordNet http://sentiwordnet.isti.cnr.it
Jika belum di instalasi, tampaknya sentiment classifer akan men-download & meng-compile semua yang dibutuhkan.
Cara Install
Akan butuh akses ke folder
/usr/local/lib/python2.7/dist-packages/
Perintah shell
sudo python setup.py install
akan terinstalasi di
/usr/local/bin
Download
https://github.com/downloads/kevincobain2000/sentiment_classifier/sentiment_classifier-0.5.tar.gz
Dokumen
http://readthedocs.org/docs/sentiment_classifier/en/latest/ Try Online
Penggunaan
Perintah shell
senti_classifier -c file/with/review.txt
Penggunaan Python
Perintah shell
cd sentiment_classifier/src/senti_classifier/ python senti_classifier.py -c reviews.txt
Penggunaan Library
from senti_classifier import senti_classifier sentences = ['The movie was the worst movie', 'It was the worst acting by the actors'] pos_score, neg_score = senti_classifier.polarity_scores(sentences) print pos_score, neg_score