The following software were developed at University of Freiburg
for research purposes. They are released under
GNU GPL
(General Public License)
- FAVST. This is a fast string miner (an improvement over VST) as
described in:
Sau Dan Lee
and
Luc De Raedt.
An efficient algorithm for mining string databases under constraints.
In
Lecture Notes in Computer Science, 3377:108–129,
February 2005.
- VST. This is a string miner as
described in:
Luc De Raedt,
Manfred Jaeger,
Sau Dan Lee,
and Heikki Mannila.
A theory of inductive query answering (extended abstract).
In Proc. The 2002 IEEE International Conference
on Data Mining (ICDM'02),
pages 123–130,
Maebashi, Japan,
December 9–12 2002.
- MineSeqLog. This is a miner of SeqLog patterns on SeqLog data.
SeqLog is a language for modeling first-order logical sequences.
An older version of MineSeqLog, which has no Deepening operator, was
was used in the experiments of the following publications.
- MineSeqLog-s. This is a simplified and hence more efficient
version of MineSeqLog. It can only mine simple
sequences, though.
- molfea2. This is a molecular feature miner.
It is an enhancement of the MolFea miner in our group earlier.
This implementation allows any Boolean combination (nesting allowed)
of monotonic and anti-monotonic constraints.
molfea2 also includes a Query Optimizer based on the
theories in my PhD thesis.
The optimizer is a join work with Luc De Raedt, Manfred Jaeger and
Heikki Mannila.
- spring. This implements an optimization on VST
using an estimation-based approach.
Details can be found in my PhD thesis.
Last Updated: $Date: 2005/06/09 15:38:08 $ (UTC)