Common collocation matching

Greetings,

I was wondering if anybody could suggest some methods of discovering the most common collocations in a body of text? I have a body of data that I wish to process. I can think of ways to generate the rankings, but my problem lies in the preprocessing and tagging.

I have tried searching for software to accomplish this, but have been unsuccessful so far. Sorry if this sounds a little vague, but I'm currently climbing a steep learning curve on this subject.

Many thanks,

Tony McNulty.