Context description
Conditional Functional Dependencies (CFDs) have been recently
introduced in the context of data cleaning. They can be seen as an
unification of Functional Dependencies (FD) and Association Rules
(AR) since they allow to mix attributes and attribute/values in
dependencies. In this paper, we introduce our fist results on constant
CFD inference. Not surprisingly, data mining techniques developed
for functional dependencies and association rules can be reused for
constant CFD mining. We focus on two types of techniques inherited
from FD inference: the first one extends the notion of agree sets and
the second one extends the notion of non-redundant sets, closure and
quasi-closure. We have implemented the latter technique on which
experiments have been carried out showing both the feasibility and the
scalability of our proposition.
Objectve:
Given a relation
r, discover a cover of CFDs satisfied in
r.
Publications
T. Diallo, N. Novelli.
Découverte des dépendances fonctionnelles conditionnelles fréquentes.
In Conférence Internationale Francophone sur l'Extraction et la Gestion des Connaissances (EGC'10).
Hammamet, Tunisie. 2010.
PDF
@inproceedings{DiNo10,
author = {T. Diallo and N. Novelli},
title = {D\'ecouverte des d\'ependances fonctionnelles conditionnelles fr\'equentes},
booktitle = {10i\`emes Conf\'erence Internationale Francophone sur
l'Extraction et la Gestion des Connaissances (EGC'10)},
year = {2010},
pages = {315--326},
series = {RNTI E-19}
}
T. Diallo, N. Novelli, and JM. Petit.
Discovering (frequent) constant conditional functional dependencies.
In the International Journal of Data Mining, Modelling and Management (IJDMMM), Special issue "Interesting Knowledge Mining", 4(3):205-223, 2012.
PDF.
@article{DiNoPe10,
author = {T. Diallo and N. Novelli and JM. Petit},
title = {Discovering (frequent) constant conditional functional dependencies},
journal = {International Journal of Data Mining, Modelling and Management (IJDMMM)},
year = {2012},
pages = {205-223},
series = {},
volume = {Special issue "Interesting Knowledge Mining", 4},
number = {3},
}
Download
IMPORTANT:
The
CFun software is developed for research purposes only!
No guarantees are given.
Some bugs may be present within the software.
We would appreciate any comments, bug descriptions...
The implemtation is in C++ and the executable file can be generated with MS Visual C++ (9.0 and 10.0) or GNU g++ compilers (Linux and MinGW32).
CFun implementation V1.0
CFun implementation V1.01
Last update by Noël Novelli the 2012-09-03