Good news, everyone!

Good and ol'news! I've finally obtained my Ph.d. on Electrical and Computer Engineering from University of Campinas.

My thesis is entitled "Biclustering on Uncertain Data Analysis" [in portuguese] and it was created from a great effort and from being stubborn about this subject.

On the upside, now I've got to decide my next steps and on the downside...well...I've got to decide my very next steps!

But, in the meantime I'm gonna post some thoughts and ideas that didn't turn out to be a paper.

A Novel Biclustering Algorithm

On 20 July I've presented my new biclustering algorithm at the WCCI'10 conference. This algorithm, named SwarmBcluster, consists of simple agents that are able to find maximal biclusters on a given region of the data set.

The results presented there, regarding an updated version of the paper's results, were much better than the six algorithms chosen as part of the experiment regarding volume and coverage while maintaining the residue value bellow the estipulated delta.

Also the time taken to find such high quality solutions were just slightly worse than the Cheng and Church algorithm, that is a constructive heuristic, and better than the other approaches. This position the SwarmBcluster as a nice choice to find high quality biclusters on any kind of set of data.

Although nowadays the biclustering algorithms are mainly used as a tool to find out previously unknown relations on microarray datasets, in my point of view this type of clustering will represent a major role in the future of data mining.

On the next few days I'm going to ellaborate a post about some topics in which the bicluster may be useful or interesting.

Back from WCCI 2010

Back from Barcelona where I was attending the World Congress on Computational Intelligence. I presented two papers at this conference: one about a study of the diversity mechanisms of opt-aiNet compared to classical approaches and the other one about a new algorithm created by me for delta-biclustering. Both presentations can be found at the links presentation and toolboxes (where you can find the pdf version)

Coming soon...

For those interested in biclustering algorithms, stay tuned! I've created an ant-based metaheuristic capable of maximizing the dataset coverage within the biclusters set while maintaining a low residue and large volume! Where will I publish still depends on how long the experiments take (and if it all goes well). I'm hoping to catch up the WCCI'2010 deadline...

Everything packed to CILAMCE

Now everything is ready and I'm on my way to CILAMCE. I'm pretty sure it will be an enjoyable experience. There's gonna be a lot of interesting papers I'm looking forward to.

The only con about this conference is the time to present...15 minutes including time for questions...and I had so much to talk about!

Well, anyway, I'm uploading right now the ICARIS'08 presentation about MOM-aiNet, a multi-objective multipopulation artificial immune algorithm for biclustering and after I'm back from the conference I'm gonna upload the CILAMCE's presentation

CILAMCE 2009

well, now it's time to prepare the presentation for my paper entitled "Coherent Recommendations Using Biclustering" that will be presented at CILAMCE 2009 (http://eventos.nacad.ufrj.br/). This is one part of the basis for my Ph.d. thesis. It is a novel methodology to use coherent biclusters (as defined by Cheng & Church) in order to impute missing values. More details and the presantation will be available soon...

Optimization under Uncertainties - Book Chapter

Today I finally delivered the final version of my chapter (in portuguese) about uncertainties under optimization scenarios. It ended up a high quality review about this subject! I'm preparing a presentation based on this chapter right now...

For other updates I added some icons under my picture and info to the left that links to my places on several social networking tools. That should make it easier to contact me if anyone has anything to chat about.

Publicar postagem