© The Brazilian Computer Society 2012
Published: 19 September 2012
This is the fourth and final issue of volume 18 of the Journal of the Brazilian Computer Society. It brings a special section on web and text intelligence plus four original paper contributions.
The special section on web and text intelligence includes four papers that extend contributions originally presented at the III International Workshop on Web and Text Intelligence (WTI 2010), which occurred on 24 October 2010, at São Bernardo do Campo, São Paulo, Brazil, co-located with the XX Brazilian Symposium on Artificial Intelligence, the XI Brazilian Symposium on Artificial Neural Networks and the IV Intelligent Robotics Meeting.
The first paper in the WTI Section is by João Vinagre and Alípio Mário Jorge, who study the role of two forgetting mechanisms, namely sliding windows and fading factors, in handling temporal effects in collaborative filtering algorithms. The second paper, by Sérgio R.P. da Silva, Marcelo R. Borth, Josiane M.P. Ferreira and Valéria D. Feltrim, introduces an approach to improve the quality of the categorizations performed by users in tagging-based systems by combining different Web resources. In the third paper, authors João Roberto Bertini, Alneu de Andrade Lopes and Liang Zhao present a graph-based semi-supervised approach that extends the K-associated Optimal Graph classifier to perform online semi-supervised classification. The final paper in this special section is also concerned with classification: Nuno Filipe Escudeiro and Alípio Mário Jorge introduce D-Confidence, an active learning approach that is effective in the presence of imbalanced class distribution in training sets, and evaluate its performance on text and tabular data.
The four original papers in this issue address multiple topics. The paper by Ricardo Piccoli, João Oliveira and Isabel Manssour describes an algorithm for automatically producing a camera-ready magazine that finds the optimal number of pages, given a set of page templates and a sequence of variable content to be placed on those templates.
In the following paper, Rodrigo V. Barros, Márcio P. Basgalupp, André C.P.L.F. de Carvalho and Marcos G. Quiles describe Clus-DTI, a new decision-tree induction algorithm based on clustering. They investigate how clustering data as a part of the tree induction process affects the accuracy and complexity of the resulting models.
Luan Silveira, Renan Q. Maffei, Silvia S.C. Botelho, Paulo L. Drews and Alessandro de L. Bicho introduce a method for planning paths for multiple robots in unknown environments, which is based on an incremental graph search and on the Space Colonization algorithm.
Finally, in their paper Victor Campos, Victor A.E. de Farias and Ana Silva show how to compute, in polynomial time, the b-chromatic number of a graph of girth at least 9, improving a previous result on trees.
In this closing issue for 2012, I also would like to thank editors and reviewers for their invaluable support. Thanks also to the authors who contributed to JBCS and to our readers for their interest and support.