2006-07 SIGHAN Report 
 
There is a new slate of officer bearers as indicated in 2 below.   Since 
SIGHAN’s focus is on NLP research involving the prime example of a 
non-Alphabetic language in the world, efforts have been made to improve and 
augment ties between the ACL community and the large community of 
researchers and research organizations in places such as China. 
The  following is a summary report. 

1. Workshop Organization 

One major event for the year was the 5th SIGHAN Workshop, held 
in  conjunction with COLING/ACL-2006 in Sydney, Australia, 22-23 July 2006. 
The workshop was co-chaired by Hwee Tou Ng and Olivia Kwong.  There 
were 24 submissions received, 8 of which were accepted for oral paper 
presentations and 6 as posters.

The 3rd International Chinese Language Processing Bakeoff was also 
organized in conjunction with SIGHAN-5, coordinated by Gina Levow and 
Olivia Kwong. There were two shared tasks, namely automatic Chinese word 
segmentation and named entity recognition.  Altogether 29 teams 
representing 29 organizations in 8 territories participated in the Bakeoff, which 
have set a new record. This year the benchmark corpora used in the 
evaluation also drew a new contributor : Microsoft [China], in addition to the 
regulars such as Academia Sinica [Taipei], Peking University, University of 
Colorado [originating from U Penn], and City University of Hong Kong. The 6th 
SIGHAN Workshop is intended to be held in conjunction with IJCNLP-08 in 
Hyderabad, India, 11-12 January 2008.  SIGHAN-6 will be co-chaired 
by Haizhou Li and Olivia Kwong.  A workshop proposal has recently 
been submitted.  We also plan to concurrently organise the 4th International 
Chinese Language Processing bakeoff, and we have reached 
initial agreement with the Verification Center on Chinese Language and 
Character Standards of the State Language Commission of PRC, and the 
Chinese LDC, to  co-organise the event.  Additionally, advanced discussions 
are underway for a panel on the status quo of automatic word segmentation of 
Chinese and related undertakings, to be jointly conducted by the Chinese 
Academy of Sciences, SIGHAN, and the Chinese Information Processing 
Society at the biennial conference of the latter in Dalian, China in August 
2007. 

The issues relevant to tokenization and automatic segmentation of texts 
in mostly west and southeast Asian languages and transferable experience 
from competitive events such as the SIGHAN Bakeoffs were shared with 
participants of the conference on Regional Consultation on Local Language 
Computing Policy in Developing Asia, organized in Bhutan by the PAN 
localization project based in Pakistan.

2. SIG Compliance 

The current SIGHAN officers are: 
  Benjamin Tsou (Chair) 
  Dan Jurafsky (Vice-Chair) 
  Maosong Sun (Treasurer) 
  Gina Levow (Secretary / Web Master)

The webpage of SIGHAN is located at http://www.sighan.org, containing 
a statement of purpose and the list of officers and information officers 
from individual regions, and the constitution can be found 
at  http://www.sighan.org/constitution.htm. We are proposing a constitutional 
change on adding an at-large-member and are in the process of holding a vote 
among members.  The issue will be further discussed in the SIGHAN 
business meeting in the next workshop.