**Description: **We have developed a system for learning from relational data (e.g. from graphs). The system is pretty efficient, gives a non traditional type of algorithm ("bottom up") based on the theoretical results, and is a state of the art Inductive Logic Programming (ILP) system. Current work is focused on solving large scale problems involving classification of molecules.

his work is partly supported by NSF grant IIS-0099446.

**Members:**

**Publications:**

- M. Arias, R. Khardon and J. Maloberti, Learning Horn Expressions with LogAn-H,
*Journal of Machine Learning Research*, Vol 8, pp549--587 , 2007 [+]

**Authors:**M. Arias, R. Khardon and J. MalobertiJournal of Machine Learning Research

Vol 8, pp549--587**Year:**2007**Url:**http://www.cs.tufts.edu/~roni/PUB/loganh-JMLR.pdf**Associated Research Topics:****Affiliated Tufts Members:****Tufts / Purdue Alumni:** - M. Arias and R. Khardon, Bottom-up ILP using Large Refinement Steps,
*In The Proceedings of the International Conference on Inductive Logic Programming*, pp26-42, 2004 [+]

**Authors:**M. Arias and R. KhardonIn The Proceedings of the International Conference on Inductive Logic Programming

pp26-42**Year:**2004**Associated Research Topics:****Affiliated Tufts Members:****Tufts / Purdue Alumni:** - Khardon, R., Learning Horn Expressions with LogAn-H,
*Proceedings of the International Conference on Machine Learning*, pp. 471-478, 2000 [+]

**Authors:**Khardon, R.Proceedings of the International Conference on Machine Learning

pp. 471-478**Year:**2000**Associated Research Topics:****Affiliated Tufts Members:****Tufts / Purdue Alumni:**None

**Associated Data/Software:**

- LogAn-H [+]

**Description:****LogAn-H**is a system for learning function free Horn expressions. It is based on provably correct algorithms for learning with queries. More information can be found in:

Theoretical background (correctness and complexity proofs)

Paper describing the system and some experiments

An Example Run

Two variants of the system and algorithms have been implemented (in Prolog and C) as described in the papers.

Data files and code are available (for research purposes) upon request.**Associated People:****Associated Research:**