I wish you had just gone ahead and asked the whole question at once as we requested. Breaking a single question into pieces and only feeding us one piece at a time might seem to be a good approach to you, but trust me, we can take bigger bites. We don't want to write your script for you, but if we're going to answer a question, at least let us answer the complete question.

You really should be using HTML::TokeParser. Nevertheless, the following will use a fragile regexp to find a keyword, and grab the digits that immediately follow it (whitespace optional).

if( $page =~ m/keyword\s+(\d+)/ ) {
   print "Found the keyword, and retrieved a value of $1\n";
}
[download]

Now if your HTML has multiple instances of this keyword, you'll have to ask us another question, or read perlretut and perlrequick.

By the way, if you're screen-scraping Google you are violating their Terms of Service, and exposing yourself to civil liability. From the Google Terms of Service page:

No Automated Querying
You may not send automated queries of any sort to Google's system without express permission in advance from Google. Note that "sending automated queries" includes, among other things:

• using any software which sends queries to Google to determine how a website or webpage "ranks" on Google for various queries;
• "meta-searching" Google; and
• performing "offline" searches on Google.

Please do not write to Google to request permission to "meta-search" Google for a research project, as such requests will not be granted.