When the user is searching some text with a query, not all the terms that the user uses have the same importance. The more common terms are generally removed for query execution, to reduce the noise generated by them. These terms are called stop words and they are generally articles, conjunctions, and common language words (that is, the, a, so, and, or, and so on).
The list of stop words depends on the language, and is independent from your documents. Lucene provides ways to dynamically compute the stop words list based on your indexed document a query time using the common terms query.
