Novo artigo: Um Modelo Ontológico Probabilístico para Assistir Pessoas com Declínio Cognitivo

Autores: Gabriel Machado Lunardi, Guilherme Medeiros Machado, Alencar Machado, José Palazzo M. de Oliveira

Resumo: Fornecer lembretes a um idoso, enquanto realiza suas atividades diárias, é uma atividade de suporte ao usuário e, portanto, um tópico relevante na área de Ambientes de Vivência Assistida. Determinar tais lembretes implica na tomada de decisões, uma vez que o fluxo das ações (comportamento) de uma atividade geralmente envolve ramificações. Um sistema automatizado precisa decidir qual das próximas ações é a melhor para o usuário em determinada situação. Problemas dessa natureza envolvem níveis de incerteza que precisam ser tratados. Muitas abordagens para esse problema exploram apenas dados estatísticos, ignorando tecnologias semânticas importantes, como, por exemplo, a utilização de ontologias. Apesar de importantes, as ontologias não suportam, de forma nativa, o raciocínio sobre a incerteza. Por isso, neste artigo é apresentado um modelo ontológico, com uma extensão probabilística, que permite raciocinar sobre a incerteza sem perder informação semântica. Esse modelo é avaliado por meio de um estudo de caso no qual demonstra-se uma instanciação com dados reais.

ONTOBRAS 2018, Proceedings of the XI Seminar on Ontology Research in Brazil, São Paulo, Brazil, October 1st-3rd, 2018. Edited by: Joel Luís Carbonera, Giancarlo Guizzardi, Sandro Rama Fiorini, Mara Abel, Vol-2228, ISSN 1613-0073, p. 185-196 – http://ceur-ws.org/Vol-2228/paper13.pdf

Novas publicações: Modelos e Sistemas para Cidades Inteligentes

Durante as últimas décadas tem ocorrido uma migração massiva da população mundial em direção às cidades. Por conta disso, os grandes centros urbanos devem estar preparados para enfrentar novos desafios para gerenciar e garantir a qualidade de vida de sua população. Um desses desafios é garantia que a população tenha um bom nível de inserção social e de cidadania. Neste contexto, propõe-se a utilização de recursos presentes em cidades inteligentes para estimular a integração do cidadão. Para atingir tal objetivo será proposto uma abordagem de recomendação de recursos informacionais adaptados de acordo com o nível de conhecimentos e contexto do usuário. Acredita-se que a recomendação de recursos informacionais de tipos diversos pode estimular usuários de dispositivos móveis a tornarem-se mais inseridos socialmente em suas cidades, contribuindo assim para o aperfeiçoamento nível de participação cidadã da população.

Este número dos Cadernos de Informática destina-se à publicação de resultados do projeto Universal do CNPq “Recomendação adaptativa para cidades inteligentes” Chamada Universal MCTI/CNPq Nº 01/2016, processo: 400.954/2016-8. Novos artigos serão adicionados.

Nova tese: AwARE : an approach for adaptive recommendation of resources

Título: AwARE : an approach for adaptive recommendation of resources
Autor: Machado, Guilherme Medeiros 
Orientador: Oliveira, Jose Palazzo Moreira de 
Data: 2018
Nível: Doutorado
Instituição: Universidade Federal do Rio Grande do Sul. Instituto de Informática. Programa de Pós-Graduação em Computação.

Recommender systems were proposed in early 90’s with the goal to help users deal with cognitive overload brought by the internet and the constant increase of documents. From there to now such systems have assumed many other roles like “help users to explore”, “improve decision making”, or even “entertain”. To accomplish such new goals, the system needs to look to user characteristics that help in understand what the user task is and how to adapt the recommendation to support such task. In this direction, it is proposed in this thesis an integration between recommender and adaptive strategies into a new process of adaptive recommendation. It is shown that such integration can improve recommendation accuracy and give good results to user retention, and interaction with the systems. To validate the approach, it is implemented a prototype to recommend movies to be used in a classroom. It is also collected some statistics about the 78 users who have participated of the experiment for evaluation of the new approach.

Lume

Novo artigo: BEATnIk: an algorithm to Automatic generation of educational description of movies

Logo CBIE 2017

O artigo recebeu Menção Honrosa no Brazilian Symposium on Computers in Education (Simpósio Brasileiro de Informática na Educação – SBIE 2017)


Vinicius Woloszyn, Guilherme Medeiros Machado, José Palazzo Moreira de Oliveira, Leandro Wives, Horacio Saggion

Resumo

Teachers and professors have increasingly employed different methods to enrich the learning of a subject in class, drive other assignments, and meet curriculum standards, are some examples of it. One of such methods is the use of movies as an alternative educational experience to support class discussions. In this sense, websites such as www.TeachWithMovies.com, arise as a valuable support to the creation of lesson plans. In this website, each movie is described as a lesson plan targeting the learn of a subject, the plans comprise the movie benefits and possible problems, a helpful background, the discussion and some interesting questions to be presented to the students. However, the creation of such lesson plan or even a simple educational description of the movie can demand much work and time, since the developed text must consider educational aspects of the movie. In this paper, we describe BEATnIk (Biased Educational Automatic Text Summarization) an algorithm to automatic generation of movie’s summaries, such algorithm favors educational aspects from the text to generate a biased educational summary. The algorithm input, are users’ comments about the each movie that has a lesson plan in the TeachWithMovies (TWM) website, and the user’s comments extracted from amazon’s website. BEATnIk constructs a complete graph for each movie where each sentence from the comments’ set becomes a node, and each edge weight is defined by the value of an adapted cosine similarity between the sentences. The algorithm then employs the PageRank to compute the centrality of each node. The intuition behind this approach is that central sentences highlight aspects of a movie that many other reviews frequently mention. In addition, BEATnIk takes into account keywords extracted from the lesson plans of TWM. The final educational summary is based on the centrality score of the sentences pondered by the presence of educational keywords. The comparison of our approach to TextRank, which is a Graph-based Automatic Text Summarization, revealed that BEATnIk generate summaries closer to the description of the movies in TWM. The experiments showed that our approach statistically outperforms the baseline in precision, and achieves better results both in recall and f-score using ROUGE-n, which is a set of metrics used for evaluating automatic summarization.


Texto completo: PDF

 

Novo artigo: A systematic mapping on adaptive recommender approaches for ubiquitous environments

 

Computing

pp 1–27

  • Guilherme M. Machado
  • Vinicius Maran
  • Lorayne P. Dornelles
  • Isabela Gasparini
  • Lucinéia H. Thom
  • José Palazzo M. de Oliveira

Article: DOI: 10.1007/s00607-017-0572-7

Abstract

Recommender systems were first conceived to provide suggestions of interesting items to users. The evolution of such systems provided an understanding that a recommender system is currently used to diverse objectives. One of the current challenges in the field is to have approaches of recommendation that go beyond accuracy metrics. Since it is a very recent interest of the community, this review, also characterized as an exploratory search, provides an overview of the techniques in the area that tries to look beyond accuracy. More specifically, one of the characteristics that would provide such evolution to these systems is the adaptation. This review is then performed to find the existence and characteristics of such approaches. Of the total 438 papers returned in the submission of the search string, 57 papers were analyzed after two filtering processes. The papers have shown that the area is little explored and one of the reasons is the challenge to validate non-accuracy characteristics in such approaches.

Keywords

Systematic mapping Recommender systems (RS) Adaptive systems (AS) Ubiquitous computing Context awareness 

Mathematics Subject Classification

68-02 68U35 68M99