Overview

The Second International Chinese Word Segmentation Bakeoff took place over the summer of 2005 and the results were presented at the 4th SIGHAN Workshop, held at IJCNLP'05, October 14-15.

Corpora from the following organizations were used:

CKIP, Academia Sinica, Taiwan
City University of Hong Kong, Hong Kong SAR
Beijing Universty, China
Microsoft Research, China

Data Collection

Four corpora are available for this bakeoff:

Corpus	Encoding	Word Types	Words	Character Types	Characters
Traditional Chinese
Academia Sinica	Unicode/Big Five Plus	141,340	5,449,698	6,117	8,368,050
City University of Hong Kong	HKSCS Unicode/Big Five	69,085	1,455,629	4,923	2,403,355
Simplified Chinese
Peking University	CP936/Unicode	55,303	1,109,947	4,698	1,826,448
Microsoft Research	CP936/Unicode	88,119	2,368,391	5,167	4,050,469

License

Custom