W-GeneralizedSequentialPatterns example

szymek

Hi all,

Has anybody managed to use W-GeneralizedSequentialPatterns? I tried to use it with data that worked in WEKA, e.g.

@relation sequential_test_set
@attribute day {1, 2, 3}
@attribute power_consumption {power=base, power=peak}
@attribute wind_speed {wind=calm, wind=breeze}
@data
1,power=base,wind=calm
1,power=peak,wind=breeze
2,power=base,wind=calm
2,power=peak,wind=breeze
3,power=base,wind=calm
3,power=peak,wind=breeze

and Rapid Miner generates an error saying that W-GeneralizedSequentialPatterns Cannot handle missing values. It is strange because there are no missing values in this file.

If this component does not work in RM maybe there is some other possibility to process sequential patterns in RMr?

regards,
Szymon

land

Hi,
one of the next updates of RapidMiner will contain a new GSP operator as part of the core.

Greetings,
  Sebastian

B_Miner

So is the current defective?

Thanks

land

Hi,
I must admit, that I don't know how this operator works. It's take from weka and their parameter settings are a little bit of magic, because they don't tell what they mean if you cannot guess it from one character.
The update with an own GSP operator is currently deployed and should be available in a few hours.

Greetings,
  Sebastian