In my project, I need to access and get the
contents in Wikipedia,and
then build a inverted index using these contents. But I don't know how to
get the contents in Wikipedia efficiently and easily. What can I do ? Are
there any APIs ?
Yes, see
http://mediawiki.org/wiki/API
If you need the full-text of many or even all articles you need to
download a database dump:
http://download.wikimedia.org
If the project is useful for other Wikipedians you may want to look at
Wikimedia's Toolserver which offers a live copy of the database:
http://meta.wikimedia.org/wiki/TS