Datasets

Note: Some of the links are broken. You can download (some) of the original raw data files from here.
NameCommentFormatToolWWWDownloadContributorDate
alzheimers GolemFMRhttpftp Peter Reutemann07.02.2004
cancer ProgolRftpftp Peter Reutemann07.02.2004
chess GolemRhttpftp Peter Reutemann07.02.2004
drug_datadata for Proper can be found in 'ilp' subdirectoriesPrologFMRftpftp Peter Reutemann07.02.2004
eastwestpart of RSD from Filip ZeleznyPrologFMR   Filip Zelezny07.02.2004
eastwest PrologFMRftpftp ftp Peter Reutemann07.02.2004
GenesKDD01 datasetsCSVFMRhttphttp http Peter Reutemann02.05.2004
geosystem PrologFMRftpftp Peter Reutemann07.02.2004
jobsystem PrologFMRftpftp Peter Reutemann07.02.2004
krkpart of RSD from Filip ZeleznyPrologFMR   Filip Zelezny07.02.2004
KRKthe predicates 'positive' and 'negative' have to be changed to 'position' and ':-position' (s. prepare.sh)PrologFMRftpftp Peter Reutemann27.02.2004
mesh GolemFMRhttpftp Peter Reutemann07.02.2004
musk1classes are changed to '0' and '1' with prepare.sh; already propositionalized, therefore the script also creates "relational" dataCSVFMRftpftp Peter Reutemann07.02.2004
musk2see 'musk1'CSVFMRftpftp Peter Reutemann07.02.2004
mutagenesispart of RSD from Filip Zelezny (only friendly, atom-bond-relationship is added to other functors, e.g. to anthracene!)PrologFMR   Filip Zelezny07.02.2004
mutagenesisthe friendly inst. are in dir 188, the unfriendly in 42ProgolFMRhttpftp Peter Reutemann27.02.2004
mutagenesisseveral relations from original data: atom, atom-bond, atom-bond-atomCSVFMR   Bernhard Pfahringer10.05.2004
proteins GolemFMRhttpftp Peter Reutemann07.02.2004
RELAGGSdata from Mark Krogel ([email protected])MySQL-dumpR   Peter Reutemann07.02.2004
restsystemOut Of Memory Exception in Weka with 400mPrologRftpftp Peter Reutemann18.02.2004
satellite GolemFMRhttpftp Peter Reutemann07.02.2004
suramin ProgolFMRhttpftp Peter Reutemann07.02.2004
Tic2000column descriptions added, file for classified and unclassified created, also learning + unclassified (s. prepare.sh)CSV (Separator = \t)httphttp Peter Reutemann26.02.2004

Key:

  • Tool: R=RELAGGS, F=Flattening, M=REMILK

Note:

If the script prepare.sh is mentioned, it is referring to the script in the Datasets-Archive of the Proper-Distribution.

Last Updated: <!-- updated -->18 May 2020<!-- updated -->