Projects

memeswimmer

various codes for natural language text processing, esp. Thai text. memeswimmer is intended to be a replacement of "digiboard" (deployed on sites like http://siit.net/webboard/ and http://daytag.org/webboard/ ) Why Meme Swimmer ?Meme (/miːm/) consists of any unit of cultural ... [More] information, such as a practice or idea, that gets transmitted verbally or by repeated action from one mind to another. Gene is for biological, meme is for cultural. Gene pool is for diversity of life, biologically. Meme pool is for the same, culturally. And we are all the "meme swimmer". [Less]

pylatinam

PyLatinam is Python module that deals with linguistic processing of Latin language words. This includes for time being declension and conjugation (morphology), with plans to upgrade code to support more processing, even on syntactic level.

textpro

In this project focused on the text processing and nature language processing with python power. In my project to describe the basic and advanced level of NLP using python. Still i am working as NLP core developer in serene informatics. Also contributing a more nlp related project both python and java platform.

wilkens

Code and scripts here are related to my ongoing work on literary text processing and analysis. Some pieces are cleaner than others. For more information, please contact me at mattwilkens@gmail.com.

yulu

Include CutWord, etc.

kltk

Korean Language Toolkit

weboca

Introduction to WeBoCaWeBoCa is an advanced and altered implementation of JBootCat (a Java implemention of the BootCat scripts written by Marco Baroni et al for acquiring corpora from the Internet). Written by Michael Drayson, WeBoCa will allow users to create a corpus from a range of search ... [More] engines, and then conduct processing on the corpus in order to tidy up / manipulate the corpus in a range of ways. The BootCat scripts are of great interest to linguists, translators and anyone researching such techniques for academic purposes. While the main goal of JBootCat was 'to encapsulate the BootCat functionality within a user-friendly desktop application', WeBoCa looks to improve upon the open-source application, and increase its functionality in terms of both corpus collection, and knowledge discovery from within the corpus created. The application is now in a state ready for public release. WeBoCa FeaturesWeBoCa includes the following features: Vertical / Horizontal corpus creation Google / Yahoo search engine implementation Define additional search parameters Define a word limit Define a page size limit Save URLs used in downloading Advanced URL processing including; Remove stored URLs as terms Remove non alpha-numerical terms Sort corpus Convert corpus terms to lower case Remove non-unique corpus terms Generate frequency count Running WeBoCaIn order to run WeBoCa, first please ensure you have the Java Virtual Environment 1.4 or greater installed. Then, if using Windows download the latest WeBoCa Windows executable archive. If prefered, or using non-Windows system, download the latest WeBoCa Distribution release, navigate to extracted folder and type 'java -jar WeBoCa.jar' at the terminal. You can also access the latest WeBoCa source code by connecting to it's Subversion (SVN) repository from your favourite IDE. WeBoCa was developed in Netbeans with the SVN pluggin and this is the recommended IDE. Any issues or problems? Contact Michael Drayson at weboca.info@gmail.com Version History1.2 Fixed String bug on Get URLs button, and corrected GUI errors 1.1 Fixed String bug, optimised file loading and processing algorithms 1.0 Initial public release [Less]

murdermystery

Murder Mystery Interactive Drama being created at CCL (Georgia Tech)

nlptest

Natural language processing, in Java

opencicero

An API for development of natural language processing applications. Modularized into distinct components that may run as individual services (such as web services) and combined to form conventional or new systems.