Searching for authors named "Christine Decaestecker" – sorted by Relevance.

Order by: Citations | Year (Descending) | Year (Ascending) | Recency

Try your query at: Scholar | Yahoo! | Ask | Bing | CSB

8 documents found, showing 1 through 8.

Any Reasonable Cost Function Can Be Used for a Posteriori Probability Approximation
by Marco Saerens, Patrice Latinne, Christine Decaestecker — 2001 — IEEE Transactions on Neural Networks
…In this letter, we provide a straightforward proof of an important, but nevertheless little known, result obtained by Lindley in the context of subjective probability theory. This result, once interpreted in the machine learning/pattern recognition context, puts new lights on the probabilistic i…
Cited by 2 (1 self) – Add To MetaCart

Comparing RBF and Fuzzy Inference Systems on Theoretical and Practical Basis
by Hugues Bersini, Gianluca Bontempi, Christine Decaestecker — 1995 — in F.Fogelman-Soulie', P. Gallinari (Eds.) ICANN '95,International Conference on Artificial Neural Networks, Paris
…this paper is the extension of an experimental work presented in [4] comparing several forms of RBF for classification applications. An initialization strategy is used to determine the number and location of the Gaussians centers and the widths. It is linked to the detection of clusters in the patte…
Cited by 9 (0 self) – Add To MetaCart

Limiting the Number of Trees in Random Forests
by Patrice Latinne, Olivier Debeir, Christine Decaestecker — In Proceedings of MCS 2001, LNCS 2096, 2001
…Abstract. The aim of this paper is to propose a simple procedure that aprioridetermines a minimum number of classifiers to combine in order to obtain a prediction accuracy level similar to the one obtained with the combination of larger ensembles. The procedure is based on the McNemar non-parametric…
Cited by 1 (0 self) – Add To MetaCart

Adjusting the Outputs of a Classifier to New a Priori Probabilities May Significantly Improve Classification Accuracy: Evidence from a Multi-Class Problem in Remote Sensing
by Patrice Latinne, Marco Saerens, etal., Christine Decaestecker — 2001 — Neural Computation
…In the present study, we introduce a simple iterative procedure that allows to correct the outputs of a classifier with respect to the new a priori probabilities of a new data set to be scored, even when these new a priori probabilities are unknown in advance. We also show that a significant i…
Cited by 12 (1 self) – Add To MetaCart

Graph nodes clustering based on the commute-time kernel
by Luh Yen, Francois Fouss, Christine Decaestecker, Pascal Francq, Marco Saerens — 2007 — In Proceedings of the 11th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2007). Lecture notes in Computer Science, LNCS
…This work presents a kernel method for clustering the nodes of a weighted, undirected, graph. The algorithm is based on a two-step procedure. First, the sigmoid commute-time kernel (KCT), providing a similarity measure between any couple of nodes by taking the indirect links into account, is compute…
Cited by 2 (1 self) – Add To MetaCart

About Breaking the Trade Off Between Accuracy and Comprehensibility in Concept Learning
by Thierry Van De Merckt, Christine Decaestecker — 1995
…. The central issue of this paper is concerned with the knowledge representation used to encode inductively acquired concept descriptions. The central question being how to reconcile, in concept learning, the need for accurate representations (in terms of classification) as well as comprehensible on…
Cited by 2 (0 self) – Add To MetaCart

Pattern Analysis Applications (2002)5:201--209
by Ownership And Copyright, Patrice Latinne, Olivier Debeir, Christine Decaestecker — 2002
…Several ways of manipulating a training set have shown that weakened classifier combination can improve prediction accuracy.…
Add To MetaCart

Adjusting the Outputs of a Classifier to New a
by Priori Probabilities Simple, Marco Saerens, Patrice Latinne, Christine Decaestecker, Université Libredebruxelles — 2002 — Neural Computation
…It sometimes happens, for instance in case-control studies, that a classifier is trained on a data set which does not reflect the true a priori probabilities of the target classes on real-world data. This may have a negative e#ect on the classification accuracy obtained on the real-world data se…
Add To MetaCart

Showing 1 through 8.

Try your query at: Scholar | Yahoo! | Ask | Bing | CSB