The development of local ontologies involves a large volume
of on-line text. To analyze this text at a manageable level, this
paper focuses on information about tourist attractions in
New York City. This destination is selected because it is
one of the largest metropolitan areas in the world and
offers diverse attractions. Tourism web sites are rst selected
through a Google search using two keywords, `tourist attractions' and `New York City.' By examining the rst 100 results
of the search, 24 web sites (see Appendix B for the full list)
are selected based on two criteria. First, these web sites must
contain information about a large number of attractions.
Second, the information on these sites must be presented
under explicit titles, such as `location' and `open hours.'
Those sites that present the information in multiple levels of
titles are more preferred.