Artigo de evento

URI Permanente para esta coleção

https://dspace.mackenzie.br/handle/10899/33884

Navegar

Navegando Artigo de evento por Autor "Abilhoa W.D."

Agora exibindo 1 - 2 de 2
  • Artigo de evento
    Density classification based on agents under majority rule: Connectivity influence on performance
    Abilhoa W.D.; de Oliveira P.P.B. (2020)
    © Springer Nature Switzerland AG 2020.The density classification task is a prototypical consensus problem of distributed solution, usually addressed in the field of cellular automata. In short, this problem consists of finding the most frequent state in a binary sequence, necessarily through a non-global process on which the automaton reaches uniform consensus about such state. In this regard, we formulate the task as an agent-based model, in which agents set up a connectivity pattern, here corresponding to a circulant graph, and update their internal states according to the majority rule. The performance of the model corresponds to the number of correctly classified densities, given a set of binary sequences. Therefore, our goal is to analyze the sensibility of the model’s performance in terms of the connectivity pattern associated with it, configured as a circulant graph, under different orders, average degrees and connectivity arrangements.
  • Artigo de evento
    TKG: A graph-based approach to extract keywords from tweets
    Abilhoa W.D.; de Castro L.N. (2014)
    Twitter is a microblog service that generates a huge amount of textual content daily. All this content needs to be explored by means of text mining, natural language processing, information retrieval, and other techniques. In this context, automatic keyword extraction is a task of great usefulness. A fundamental step in text mining techniques consists of building a model for text representation. This paper proposes a keyword extraction method for tweet collections that represents texts as graphs and applies centrality measures for finding the relevant vertices (keywords). The proposal is applied to two tweet collections of Brazilian TV shows and its results are compared to those of TFIDF and KEA. © Springer International Publishing Switzerland 2014.