Here's a roundup of the APIs to mine Twitter data:
Trstrank: The PageRank-esque algorithm we covered previouslyWordbag: A list of "most common words" for a user with a twist. Universally-common words are essentially filtered out because InfoChimps compares a user's word choice to all of Twitter.Influence: Gets the number of replies in and out for the user, along wiht the number of tweets and the age of the account. Analysis of influence is up to you.Conversation: Provides a summary of public interactions between two users. Results are a list of tweets in which the first user replied to the second.