• No results found

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

N/A
N/A
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)

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)

Acknowledgements

Between my colleagues, because of their constant support over the first two years of my PhD, I am particularly grateful to my two Czech friends, Pavel Pacl´ık and Professor Pavel Brazdil.

Since they contributed so much to my PhD, I am also very thankful to Ingrid Meulenbelt, Jeanine Houwing-Duistermaat and Eline Slagboom, Stephanie van Rooden, Martine Visser, Han Marinus and Bob van Hilten, and more recently Andreas Bender. What a challenge to work with you all!

My former colleagues of the LIAAD in Porto also played a very important role as, during my MSc thesis, I adapted to their way of conducting research.

In the last years, each time I visited them, they left me with a unique feeling of availability and friendliness.

Other than my project colleagues, I wish to thank Wim Aspers and Thijs Dijk for counselling, chatting and arranging daily things at the LIACS. I also shared office with Yanju Zhang over the last two years and however distant our respective countries, I did much appreciate the understanding which underlied our regular talks, especially those not related to our jobs. I further enjoyed the cultural diversity at the LIACS having neighbour colleagues from the Netherlands, Morocco, China, Russia, Israel, Italy, etc.

Job accounting for 50% to my coming to the Netherlands, I want to thank those who, by their teaching, contributed to a today being which I feel more confortable with than before. Thanks to Ren´e and Jelena, Aharona and Pauline, Peter, Natalia.

Already unrooted for a couple of years before starting my PhD and living all that time with a single backpack, I went with some relief for a few years to the Netherlands. Being relatively close to my motherland, I enjoyed returning to my family who, from my parents to my grandma’s, from my sister to my cousins and

(3)

146 Acknowledgements

from my uncles to my nephews, has been so welcoming and caring each time that I visited them. What a family, merci.

Fabrice, July-November 2008.

Referenties

GERELATEERDE DOCUMENTEN

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

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

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

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

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

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

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

To conclude, our comparison of algorithms data mining scenario offers a new view on the problem of classifying text documents into categories. This focus en- abled to show that SVM