MLnetOiS Logo left

MLnetOiS Logo right

 Resources:DatasetsDetails

 

Index
Resources
* Bibliography
* Courses
* Datasets
* Links
* Showcases
* Software


LINUS

Add Add a dataset to the database.
Update Update the entry for this dataset.

 

 

 

Back

 

 

 

up arrow

Name (abbrev)

Name (full)

Category

Last update

 

LINUS

Samples from ILP systems

b D, Y

up arrow

Application domain

Further specifications

 

LINUS

paper, datasets

up arrow

Type

Format

Complexity

 

ILP

Linus

small examples

up arrow

WWW / FTP

 

 

ftp://ftp.mlnet.org/ml-archive/ILP/public/software/linus/
ftp://ftp.mlnet.org/ml-archive/ILP/public/papers/nonrec-learn.ps

up arrow

Contact person(s)

Related group(s)

Optional contact address

 

  1. Dzeroski, Saso
  2. Lavrac, Nada
  1. Jozef Stefan Institute, Department of Intelligent

[email protected]

up arrow

References

 

Dietterich,T.G. and Michalski,R.S. (1986). Learning to
predict sequences. In Michalski,R.S., Carbonell,J.G. and
Mitchell,T.M. (eds.) Machine learning: An artificial
intelligence approach (volume 2). Los Altos: Morgan Kaufmann.

N. Lavrac, S. Dzeroski, and M. Grobelnik.
Learning nonrecursive definitions of relations with LINUS.
In Proc. Fifth European Working Session on Learning, 265-281.
Springer, Berlin, 1991.

N. Lavrac, S. Dzeroski, and M. Grobelnik.
Experiments in learning nonrecursive definitions of relations with LINUS.
Technical Report IJS-DP 5863, Jozef Stefan Institute, Ljubljana, Slovenia,
1990.

Michalski,R.S. (1980). Pattern recognition as rule-guided
inductive infernece. IEEE Transactions on Pattern Analysis and
Machine Intelligence, 2:349-361

Quinlan,R. (1990). Learning logical definitions from
relations. Machine Learning, 5(3):239-266

Winston,P.H. (1975). Learning structural descriptions from
examples. In: Winston,P.H. (ed.) The psychology of computer
vision. New York: McGraw-Hill

up arrow

Annotations

 

This distribution includes several small datasets addopted from the ML literature in the format appropriate for LINUS [Lavrac et al. 91, Lavrac et al. 90]. The problems include:

  • learning family relationships [Michalski 80] (file dhdbmoth.pl)
  • learning the concept of an arch [Winston 75] (file dhdbarch.pl)
  • learning where trains are heading [Michalski 80] (file dhdbeast.pl)
  • learning rules governing card sequences in the game Eleusis [Dietterich 86] (files dhdbele[123m].pl)


All the mentioned references point to the original sources of the data. FOIL [Quinlan 90] was applied to all of these domains and the datasets were prepared starting from this paper, where descriptions of the domains can be found. The results obtained from LINUS are described in a Technical Report [Lavrac et al. 90].

 

Comments

 

 

 

 

Index
Resources
* Bibliography
* Courses
* Datasets
* Links
* Showcases
* Software

 

 

Supported by EU project Esprit No. 29288, University of Magdeburg Logo Uni Magdeburg and GMD Logo GMD
© ECSC - EC - EAEC, Brussels-Luxembourg, 2000