• No results found

Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R.

N/A
N/A
Info
Downloaden
Protected

Academic year: 2021

Share "Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R."

Copied!
3
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Download het nu ( 3 pagina )

Hele tekst

(1)

Data mining scenarios for the discovery of subtypes and the comparison of algorithms

Colas, F.P.R.

Citation

Colas, F. P. R. (2009, March 4). Data mining scenarios for the discovery of subtypes and the comparison of algorithms. Retrieved from

https://hdl.handle.net/1887/13575

Version: Corrected Publisher’s Version

License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden

Downloaded from: https://hdl.handle.net/1887/13575

Note: To cite this publication please use the final published version (if

applicable).

(2)

Curriculum Vitae

Fabrice Colas was born on September 14, 1981, in Laval, France.

Fabrice conducted electrical engineering studies at the ESIEA institute1. As part of ERASMUS, he spent a semester in Budapest2 and then, he went for a one year specialization in data mining research in Lyon3. For his final graduation project, he worked in text mining with Prof. Brazdil4 provided the portuguese pluri-anual funding of the Funda¸c˜ao para a Ciencia e Tecnologia under the SUMO project. In ’05, he graduated with the jury honors of the ESIEA institute and the Lyon II University, respectively with a degree of engineering and a MSc re- search oriented in data mining. After what he joined Prof. Kok to prepare a doctoral dissertation at the Leiden University5 made possible by the support of the Netherlands BioInformatics Center (NBIC).

During his three years-long PhD research, he authored seven articles in peer- reviewed conferences that totalize fourteen co-authors from ten research groups.

He was also a program committee member of IFIP AI ’08, Milan (Italy). Ad- ditionally, Fabrice prepared and made available publicly the R SubtypeDiscovery package, a data mining scenario for subtyping that was developed in the course of his PhD. This package was presented at the R BioConductor conference ’08, Seattle (USA).

This thesis summarizes his research on Data Mining Scenarios for the Discov- ery of Subtypes and the Comparison of Algorithms.

1’00-’05: ´Ecole Sup´erieur d’Informatique, ´Electronique, Automatique, Laval, France.

2’04: Budapesti M˝uszaki ´es Gazdas´agtudom´anyi Egyetem, Budapest, Hungary.

3’04-’05: DEA ECD, Lyon II University, France.

4’05: Laboratorio de Inteligˆencia Artificial e Apoio a Decis˜ao, University of Porto, Portugal.

5’06-’09: Leiden Institute of Advanced Computer Science, the Netherlands.

(3)

Referenties

Download het nu ( PDF - 3 pagina - 166.20 KB )

GERELATEERDE DOCUMENTEN

Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R.

Dataset used in the thesis The dataset is composed of substances taken from the 2008 WADA Prohibited List together with molecules having similar biological activity and

Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R.

This scenario involves techniques to prepare data, a computational approach repeating data modeling to select for a number of clusters and a particular model, as well as other

Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R.

To prevent cluster analyses that model only the time dimension in the data, we presented a method that helps to select for a type of time adjustment by assessing the cluster

Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R.

Furthermore, this number also illustrates that the models (VVI,6) tend, in average, to be more likely in terms of BIC scores than the other combinations of model type and number

Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R.

We start by presenting the design of the implementation: the data preparation methods, the dataset class, the cluster result class, and the methods to characterize, compare and

Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R.

Therefore, when running experiments on complex classification tasks involving more than two-classes, we are actually comparing n SVM classifiers (for n classes) to single

Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R.

In fact, on those tasks, small feature space SVM classifiers would, first, exhibit performances that compare with the best ones shown by the 49 nearest neighbors classifier and

Data mining scenarios for the discovery of subtypes and the comparison of algorithms Colas, F.P.R.

Furthermore, in accordance to the several performance drops observed for small C values (illustrated in Figures 8.3 (a) and (c)), the tightly constrained SVM’s can be less stable