MIOPIA-SD (Sistema de Minería de Opiniones en Múltiples 
Idiomas mediante Análisis Sintáctico de Dependencias)
is a library which provides you the capabilities for analysing the
perception  of the public with respect to a product, service, event or a 
celebrity, given a collection of related messages.

INSTALLATION

NOTE: Miopia requires Python 2.7

1. Install setuptools -> https://pypi.python.org/pypi/setuptools
2. Install Natural Language Toolkit (NLTK) and their dependencies -> http://nltk.org/install.html
3. Install nltk_data (the tokenizers module) -> http://nltk.org/data.html 
4. Install textblob -> http://textblob.readthedocs.org/en/dev/ (and pip-install -U textblob-aptagger)
5. Download MaltParser 1.7.2 -> http://www.maltparser.org/download.html
6. Unzip MaltParser in your desired directory (e.g /opt/maltparser/MaltParser-1.7.2)
7. Download Weka (e.g. 3-7-10) -> http://www.cs.waikato.ac.nz/ml/weka/downloading.html
8. Unzip Weka in your desired directory (e.g /opt/weka/weka-3-7-10)
9. Unzip miopia-1.0.0.tar.gz (if you didn't it yet using tar -xvzf miopia-1.0.0.tar.gz from the command line)
10. Modify the file miopia/miopia_yaml.conf to establish the path to the maltparser-1.7.2.jar and weka.jar (parameter path_maltparser)
11. Install miopia-1.0.0 executing -> sudo python setup.py install
-------------------------------------------------------------------------------------------------------
DEMO

If you have been able to complete all installation steps, take a look to the demos located in:

/usr/local/lib/python2.7/dist-packages/miope-0.1.0-py2.7.egg/demo

Models for Spanish language:

python demo_supervised_analyzer_es.py
python demo_unsupervised_analyzer.py

Models for English language:

python demo_supervised_analyzer_en.py