A Gamma-Poisson Mixture Topic Model for Short Text

Table 2

Document statistics.

Dataset	Number of documents	Size of vocabulary	Number of topics	Average (Standard deviation) of document length	Minimum (Maximum) document length
Tweet	2 472	5098	89	8.5 (3.2)	2 (20)
Pascal flickr	4 821	3188	20	4.9 (1.8)	1 (19)
Search snippets	12 295	4705	8	14.4 (4.4)	1 (37)