Editorial

The third issue of volume 18 of the Journal of the Brazilian Computer Society brings us seven papers on diverse computer science topics. Four of the papers in this issue are extended and revised versions of papers that have been presented at conferences sponsored by the Brazilian Computer Society, and three are new original contributions.

The first paper is by Carolina Bigonha, Thiago N. C. Cardoso, Mirella M. Moro, Marcos A. Gonçalves, and Virgílio A. F. Almeida. The authors introduce a new method for ranking the most influential users in Twitter, based on a combination of factors, and report results from experimental evaluations that confirm the capability of the proposed approach to successfully identify influential users.

The second paper, by Carlos de Salles Soares Neto, Luiz Fernando Gomes Soares, and Clarisse Sieckenius de Souza introduces TAL, a Template Authoring Language for creating hypermedia document templates independently of the target hypermedia authoring language. Authors also described a TAL processor that generates complete hypermedia documents from TAL templates and content-specific data files.

The paper by Cleidson R.B. de Souza, Jean M.R. Costa, and Marcelo Cataldo addresses the design of collaborative tools. Authors studied and analyzed the scalability of coordination requirements of software developers involved in a large-scale distributed software development project, and report some findings potentially useful to the designers of such systems.

In the fourth paper, Vânia Paula de Almeida Neris and M. Cecília C. Baranauskas present and evaluate a practical approach to elicit and formalize the tailorable behavior of interactive systems, so they can more flexibly meet requirements of different user profiles. Their approach includes end user participation and formalization with a norm-based structure.

In the following paper, authors K. Jaya Priya and R.S. Rajesh handle face recognition with a local selective feature extraction approach based on Gabor filters and the Local Binary Pattern (LBP) method. Experiments using well-known face databases show that their approach is time effective and obtains promising results in the scenario of one training sample per person with significant facial variation.

Francisco de Assis Pereira Vasconcelos de Arruda, José Eustáquio Rangel de Queiroz and Herman Martins Gomes present a neural network-based technique to automatically enable non-photorealistic renderings from digital face images, which resemble semi-detailed sketches. They report results from objective and subjective evaluations of their approach, as compared to existing ones.

In the final paper, Varun Gupta and Jitender Kumar Chhabra propose a cohesion measure for re-usable components (packages) in object-oriented systems that considers internal dependencies and hierarchical structure. They validate the proposed measure theoretically as well as empirically, finding it to be a better predictor of code reusability than existing cohesion measures.

Finally, this issue brings an erratum to the paper “On the reliability and availability of replicated and rejuvenating systems under stealth attacks and intrusions”, which appeared at this volume, issue 1.