Bell Interview Question

How would you isolate/identify a specific pattern in a large text document?