8 bits

Text classifier for sentiment analysis

I just created a NaiveBayes classifier for sentiment analysis of texts in social media. I'm quite proud of the 78% over all accuracy, 100% of the "positief" precision and 100% of the "negatief" recall. You can find more information of this terms here. 

Loop through Google Spreadsheet with Python

Create a simple spreadsheet with Google Drive 

Publish the spreadsheet

(Example: https://docs.google.com/spreadsheet/pub?key=0BBZ40Xdv4Me8dEotdGVVS1dLR05KVjZMS3VpMXozSEE&output=html)

Download BeautifulSoup for Python

Put BeautifulSoup.py in your scripts folder

Start the Python terminal

Command:

import urllib2
from BeautifulSoup import BeautifulSoup
url = urllib2.urlopen("https://docs.google.com/spreadsheet/pub?key=0BBZ40Xdv4Me8dEotdGVVS1dLR05KVjZMS3VpMXozSEE&output=html")
content = BeautifulSoup(url)
items = content.findAll('td',{ 'class' : 's1'})
for item in items:
   print item.string 

 

Simple social media analysis with Google Fusion Tables

A short screen recording of one of the projects I'm workin' on

simple social media analysis from Mark Stam on Vimeo.

How to run Python 2.7 on Synology NAS

A simple walkthrough to install Python 2.7 on Synology NAS:

1) install ipkg

Read this blogposting for more information about installing ipkg:

http://setaoffice.com/2011/04/08/how-to-install-compiled-programs-on-a-synology-nas/

2) run 'ipkg update'

3) run 'ipkg install python27'
It will install Python and all dependencies

4) now, the command 'python2.7' will start the Python command line
print 'hello world';

There are many other packages which can be installed on the Synology NAS with Optware, take a look at this website:

http://www.nslu2-linux.org/wiki/Optware/HomePage

complete list: http://www.nslu2-linux.org/wiki/Optware/Packages

Why don't you install nano, sudo etc., it will definitely make your life easier!

Easy Google App Engine examples: DataStore

Google App Engine manages the scaling and maintenance of data storage automatically. GAE's abstraction for data is easy to understand, but it is not obvious how to best take advantage of it's features. In particular, it's suprisingly different from the relational database. Google doesn't call it a 'database', but a 'datastore'. It's best understood as an object database. An object in the datastore is known as an entity.