The problem of supporting end users in finding and deriving good
strategies for their information search is well known. During an
interactive search session users have to make many decisions regarding
the best way to proceed from their current situation. This is a
challenge for many inexperienced searchers.

Using the successful searches of previous users, an adaptive support
system can suggest appropriate next steps at any point during an
interactive search. Case-base reasoning techniques have been used to
implement such a  support system, which uses the situations of previous
users as cases and suggests search actions and stratagems as solutions
(adapted to the situation of the current users). The system can learn
from user participation.

Versions of the suggester have been integrated into a digital library
search system (Daffodil) as well as the open source browser Firefox,
where it is used to support Google searches with interactive
suggestions.