Data streams pose several challenges on data mining algorithm design. Limited use of resources (time and memory) is one. The necessity of dealing with data whose nature or distribution changes over time is another fundamental one. Dealing with time-changing data requires in turn strategies for detecting and quantifying change, forgetting stale examples, and for model revision. Fairly generic strategies exist for detecting change and deciding when examples are no longer relevant. Model revision strategies, on the other hand, are in most cases method-specific.