Two popular inferencing approaches are forward chaining (data driven reasoning) and backward chaining (hypothesis driven reasoning). Expert systems often use these two approaches in concert to provide a natural consulting environment
Another task for the inference engine is to reason with uncertain data. The certainty associated with a value is usually described by a numeric value, and various mathematical rules are employed to derive the joint confidence across sets of values.
This concludes the Inference Methods and Uncertainty tutorial. If you would like a more general overview of expert systems, the Expert System Introduction is recommended. To learn more about the process of developing expert system knowledge bases examine the Knowledge Engineering tutorial. If you have never run the expert systems provided on this Web site, Using eXpertise2Go's Knowledge Bases is a good starting point.