Which operator form IO.Example can read the data from .data files? I would like to generate association rules from dataset in .data .names format. Could this be possible?
according to this source (http://www.fileinfo.com/extension/data) .data is a tab-delimited format. So you should have no trouble to read the file using the ReadCSV operator (see operators tab -> Import -> Data). I was on the other side not able to find a description of the .names-format.
But as soon as the data is in rapidminer, you can learn association rules (see operators tab -> Modelling -> Association and Item Set Mining).
regards,
steffen
0
Options
IngoRMAdministrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, Community Manager, RMResearcher, Member, University ProfessorPosts: 1,751 RM Founder
Just quickly jumping in: .data and .names was a file pair mainly used by the C4.5 learner by Quinlan. Actually, I am not aware of any other program using this file format and this might be the reason why the operator for reading these file pairs is named "Read C4.5". ;D
Steffen is right, however, since the .data file could also be read by "Read CSV" but in this case you would loose the meta data information from the .names file. And of course he is also right about the association rules part.
I am afraid that I have to switch to RapidMiner 5.0 to get this done. I am still using RapidMiner 4.5 I want just to do some tasks in association rules for reasearch purposes and the test data that I have are in .data format. A repository of .aml or .arff files will be a good alternative.
Contributor II
Answers
according to this source (http://www.fileinfo.com/extension/data) .data is a tab-delimited format. So you should have no trouble to read the file using the ReadCSV operator (see operators tab -> Import -> Data). I was on the other side not able to find a description of the .names-format.
But as soon as the data is in rapidminer, you can learn association rules
regards,
steffen
Steffen is right, however, since the .data file could also be read by "Read CSV" but in this case you would loose the meta data information from the .names file. And of course he is also right about the association rules part.
Cheers,
Ingo
I am afraid that I have to switch to RapidMiner 5.0 to get this done. I am still using RapidMiner 4.5
I want just to do some tasks in association rules for reasearch purposes and the test data that I have are in .data format. A repository of .aml or .arff files will be a good alternative.
Thanx again,
Maha