A dockerized framework for hierarchical frequency-based document clustering on cloud computing infrastructures

Journal of Cloud Computing

Advances, Systems and Applications

Table 1 Summary of datasets used to evaluate the hierarchical clustering algorithms

Dataset	Domain	# of docs	# of terms	# of terms after prep.	# of classes	Source
Classic4	Abstracts	7095	7749	6576	4	[43]
Reviews	News articles	4069	22,927	12,431	5	[43]
Tr23	TREC documents	204	5833	4384	6	[43]
LATimes	News articles	6279	10,020	6389	6	[43]
Tr31	TREC documents	927	10,129	6946	7	[43]
La2s	News articles	3075	12,433	8517	7	[43]
WebKb	Web pages	8282	22,892	11,009	7	[43]
Tr12	TREC documents	313	5805	4283	8	[43]
Re8	News articles	7674	8901	5379	8	[43]
Tr11	TREC documents	414	6430	4632	9	[43]
Tr45	TREC documents	690	8262	6016	10	[43]
Tr41	TREC documents	878	7455	5406	10	[43]
Oh10	Medical documents	1050	3239	2425	10	[43]
Dmoz-Science	Web pages	6000	5011	3719	12	[43]
Dmoz-Health	Web pages	3500	4217	3172	13	[43]
Re0	Articles	1504	2886	2209	13	[43]
Dmoz-Computers	Web pages	9500	5011	3527	19	[43]
Wap	Web pages	1560	8460	5988	20	[43]
20 Newsgroups	E-mails	18,808	45,434	16,499	20	[43]
Re1	Articles	1657	3758	2863	25	[43]
ACM	Digital library	3493	60,768	16,315	40	[43]
New3	News articles	9558	26,833	14,483	44	[43]
Opinosis	Reviews	6457	2693	2201	51	[43]
NYTimes	News articles	300,000	102,660	18,001	-	[44]
PubMed	Abstracts	8,200,000	141,043	21,451	-	[44]