View-based query processing involves answering queries using pre-defined views instead of raw data. This is important for tasks like data warehousing and query optimization. The main tasks include directly answering queries using views, rewriting queries in terms of views, and comparing queries and rewritings. In the context of semistructured data like web data and XML, techniques using automata and regular path queries are used to navigate and analyze the data efficiently.