FSA/INMA -
Département d'ingénierie mathématique

 

Cristobald de Kerchove

I am currently doing a Ph.D. in the
Department of Mathematical Engineering
at the

Université catholique de Louvain-la-Neuve

You can find on this webpage
a short cv,
the main directions in my research,
my publication list,
selected talks,
a teaching part
and some links.


Bâtiment Euler, Av. Georges Lemaître, 4
B-1348 Louvain-la-Neuve, Belgium

 

Tel: +32 10 47 80 39
Fax: +32 10 47 21 80

 

Short cv

2004 - 2009 : PhD student in the Dpt of Mathematical Engineering.
2007 - 2009 :
Agrégation de l'enseignement secondaire (Postgraduate courses required to teach in secondary school)
2004 : Engineer degree in Applied Mathematics at the Université catholic de Louvain-la-Neuve

The main directions in my research

I am interested in large graphs and networks - I don't distinguish the two terms - where large means that the size of the graph, i.e. the number of nodes, is large enough to require algorithm with low complexity, typically linear complexity. My research is part of a group of several searchers and professors in my department: Large Graphs and Networks.

A first part of my research concerns the diffusion of information through large graphs. I generally work on real datasets - for example mobile phone networks - and I identify leaders who are efficient to spread some information or behavior in the network. Lastly, I look for some heuristics that could optimize the spread of information for some diffusion models.

Another trendy topic is the votes on the Web. I consider graphs of votes where users evaluate items as we can find in many web sites. Then I look for fast algorithms that are able to weight the different users according to some features of the graph and hence I obtain reputations for the users and the evaluated items.

A third point includes all sort of algorithms based on some random walk on the graph. Maybe the most famous is the PageRank algorithm. I looked at some extension of that algorithm when one allows negative links in the graph. Here I am interested in the properties of such methods and in the interpretation of the results.

Community detection and visualization of large graphs are certainly two problems that I like looking at.

Publication list

The Role of Second Trials in Cascades of Information over Networks, C. de Kerchove, G. Krings, R. Lambiotte, P. Van Dooren, V. Blondel, submitted to Physical Review E.

Geographical dispersal of mobile communication networks, R. Lambiotte, V. D. Blondel, C. de Kerchove, E. Huens, C. Prieur, Z. Smoreda, P. Van Dooren, Physica A 387,  5317-5325, 2008. arXiv

Maximizing PageRank by some Outlinks C. de Kerchove, L. Ninove and P. Van Dooren,
Accepted in the journal of Linear Algebra and its Application, vol. 429, September 2008. arXiv

Iterative Filtering for a Dynamical Reputation System C. de Kerchove and P. Van Dooren,
Posted in arXiv and a general public version accepted in Siam News, vol. 41 n°2, March 2008. Siam

The PageTrust Algorithm: How to Rank Web Pages When Negative Links are Allowed C. de Kerchove and P. Van Dooren, Proceedings of the Siam International Conference in Data Mining, 2008.

Local Leaders in Random Networks V.D. Blondel, J.-L. Guillaume, J.M. Hendrickx, C. de Kerchove and R. Lambiotte, Posted in arXiv, Phys. Rev. E 77, 036114, 2008  arXiv

Social Leaders in Graphs C. de Kerchove, E. Huens, P. Van Dooren and V. Blondel,
Lecture Notes in Control and Information Sciences, Positive Systems, vol. 341, p. 231-237, 2006.

Selected Talks

Talks given within the group Large Graphs and Networks
Maximizing PageRank via Outlinks pdf

Talk in Stanford University
Reputation systems pdf

Teaching 2008

FSAB1102: Mathématiques en bach 1, deuxième quadrimestre

Other lectures given before 2008 : Analyse numérique, projet intégré bach 1, physique bach 1.

Links


Last update : December, 2008