Discovery Options - Single Experiment
[General Discovery Options] [Batch
Experiments Options]
The discovery node's single experiment mode performs a single run of
the simulated annealing search engine that will search for rules
describing one class of records using the parameters detailed below.
-
- Target field
- The current target field in the database.
-
- Target class
- The name of the class that Discovery rules will describe.
-
- Rule bias
- A slider that you can use to control the bias
towards accuracy or coverage by setting the accuracy threshold.
-
- Accuracy threshold
- The desired accuracy threshold of rules generated
from the search. Initially, it can be better to set this value lower
than the required value in order to improve the path taken through
the search space. The default value is 50%. See also the tips
section for help on how to set this value.
-
- Max. fields in rule
- The maximum number of fields that are allowed to
appear in a valid rule. You can also click on the spin control to
increase or decrease the current value.
- Tip: if
you limit the number of conditions that can appear in a rule, your
rules will be easier to comprehend.
-
- Reset RNG Seed
- If you select this option, WITNESS Miner resets the
RNG each time the SA engine is started. This option allows you to
repeat experiments. The default value is true.
-
- Seed
- Allows the seed for the random number generator to
be changed using the RNG Seed Dialog.
The default value is 87326471.
-
- Advanced
- Advanced options are used to fine-tune the
simulated annealing (SA) search engine. Under most circumstances the
default options should be sufficient. There are seven pages of
advanced options for the SA engine.
Notes
For definitions of accuracy, coverage and other properties, see the definitions page.
