The DOM API has methods for retrieving sets of child elements from an element given that elements tag name.
You can also retrieve the values of attributes by attribute name.

Using those and given knowledge about the structure of the XML you can address any node/element in the parsed Document.

Some experimenting and reading of the API documentation goes a long way towards understanding how it works.

You may also look at http://xerces.apache.org/xerces2-j/ for tutorials (mind that they may need some modification in details to work with the built-in XML functionality of the core libraries, but the general idea is the same).