Carrot2 - Open Source Search Results Clustering Engine

Below is a summary of clustering algorithms that work within the Carrot² framework.


Algorithm	Author	Speed [s]*			Hierarchical clustering	Other features	Papers	Example results
Algorithm	Author	100	200	400	Hierarchical clustering	Other features	Papers	Example results
FuzzyAnts	Steven Schockaert	2.17	8.70	16.93	yes		[1]	london
HAOG-STC	Karol Gołembniak	0.04	0.11	0.28	yes			london
Lingo**	Stanislaw Osinski	0.34	0.52	0.84	no	multilingual clustering	[2][3]	london
Rough k-means	Ngo Chi Lang	1.38	6.76	27.73	no		[4]	london
STC	Oren Zamir (impl: Dawid Weiss)	0.04	0.10	0.23	no		[5]	london
Lingo3G***	Stanisław Osiński (Carrot Search)	0.03	0.06	0.13	yes	multilingual clustering, synonyms, advanced tuning, scalability (5000 snippets in 1.3s*)		london

*) Clustering speed measurements were done for 100, 200, 400 snippets downloaded from Yahoo! for query 'london', using the Carrot² standalone GUI application. Benchmark environment: Pentium M 1.3 GHz, 768 MB RAM, Windows XP. Java Virtual Machine: Sun JDK 1.4.2, JVM switches: -Xmx512m -Xms128m -XX:NewRatio=1 -server. Time presented in the table is an average of 75 runs, for each algorithm time measurement was followed by 25 untimed warm-up runs.

**) Lingo is the default clustering algorithm used in the Carrot² live demos.

***) Lingo3G is a commercial document clustering engine and is not available in the Open Source part of Carrot². Please contact Carrot Search for details.


© 2002-2008 Stanislaw Osinski, Dawid Weiss, updated: 18 June 2008

Algorithms