Software

It is my true pleasure to announce the first public release of Carrot2, an open source component-based framework for search results clustering.

Carrot2 includes novel algorithm for clustering textual content -- LINGO, but it also provides several implementations of well known clustering techniques such as Suffix Tree Clustering (STC) and agglomerative hierarchical methods (AHC).

The project is aimed at researchers willing to experiment with various forms of data pre- and post- processing in order to increase the information impact the user gets from search results.

Carrot2 consists of various components, which can be easily deployed, distributed, combined into different configurations and modified to one's needs (source code is provided).

The public demo of the framework is available at our server (please allow for long delays - the server is a low-end machine):

http://ophelia.cs.put.poznan.pl:2001/index.html

Several precached queries (clustered at runtime, however) are available on "demo" page. For example:

results for "data mining"

Main project pages are located at (CVS access, downloads, manual, etc:)

http://www.cs.put.poznan.pl/dweiss/carrot/index.php/index.xml

Regards,

Dawid Weiss, http://www.cs.put.poznan.pl/dweiss
Laboratory of Intelligent Decision Support Systems, Poznan UT, Poland