Conference P r o g r a m

Transcrição

1 Conference Program

2 Brazilian Conference on Intelligent System BRACIS 2012 Conference Program Contents Welcome from General Chair... 2 BRACIS 2012 Conference Organizers... 3 Sponsored by... 4 Organized by... 4 Supported by... 4 Invited Speakers... 5 Conference Program... 9 CTDIA Program WTI Workshop BWSS Workshop Summary Technical Sessions SBIA Program SBRN Program ENIA Program... 25

3 Welcome from General Chair It is my pleasure to welcome you to Curitiba for the 2012 Brazilian Conference on Intelligent System, BRACIS We very much hope you enjoy this city and all it has to offer. BRACIS 2012 encompasses the following important scientific events: XXI Brazilian Symposium on Artificial Intelligence - SBIA XII Brazilian Symposium on Neural Network - SBRN IX Encontro de Inteligência Artificial - ENIA And some co-allocated events: VIII Best MSc Dissertation/PhD Thesis Contest in Artificial Intelligence - CTDIA IV International Workshop on Web and Text Intelligence - WTI III Brazilian Workshop on Social Simulation - BWSS All these events aim to present the state-of-the-art research on Artificial Intelligence, Neural Networks (NNs) and other models of Computational Intelligence (CI) such as Fuzzy Logic and Evolutionary Computing from Brazil, Latin-American and other parts of the world. One of highlights of BRACIS 2012 is the free tutorials and workshops held during the first two days of the conference. Thanks to Roseli Romero and Alneu Lopes for chairing the tutorials and Fabricio Enembreck, the workshops. I couldn t have planned BRACIS 2012 without the generous time commitment of all local chairs. Finally, I would like to thank you, authors, presenters, reviewers, and all other attendees. Be kindly welcome to BRACIS 2012.

4 BRACIS 2012 Conference Organizers Aurora Trinidad Ramirez Pozo (UFPR) General Chair SBRN Chair Myriam Regattieri Delgado(UTFPR) SBRN Chair Gustavo Alberto Giménez Lugo (UTFPR) and Marcos Castilho (UFPR) SBIA Chairs Emerson Cabrera Paraiso (PUCPR) and Júlio Cesar Nievola (PUCPR) ENIA Chairs Ana Carolina Lorena (Unifesp) SBRN Program Chair Carlos Eduardo Thomaz (FEI) SBRN Publicity Chair Marcelo Finger (USP) and Leliane Nunes de Barros (USP) SBIA Program Chairs Renato Tinós (FFCLRP, USP) ENIA Program Chair Alneu de Andrade Lopes (USP) and Roseli Aparecida Romero (USP) Tutorial Chairs Fabrício Enembreck Workshops Chair

5 Sponsored by Organized by Supported by

6 Invited Speakers Room Gregor Mendel Auditorium Title: Machine Learning and Computational Sustainability Mon 22/10 8:30 am chair: Prof a. Dr a. Ana Lorena Thomas G. Dietterich Computational Sustainability is the study of computer science methods for solving sustainability challenges. These challenges include sustainable cities, sustainable development, energy, transportation, ecological science, and ecosystem management. In this talk, I will focus on research opportunities in ecological science and ecosystem management and describe some of the work in my lab on these problems. This includes (a) automated data cleaning and anomaly detection in sensor data streams, (b) species distribution modeling including modeling of bird migration from citizen science data, and (c) design of optimal policies for managing wildfires and invasive species. The novel machine learning challenges include flexible anomaly detection for multiple data streams, explicit models of sampling bias and measurement processes, combining probabilistic graphical models with non-parametric learning methods, collective graphical models for reasoning with count data, and optimization of complex spatiotemporal Markov decision processes. Title: Individual and Collective Rationality in Complex Systems Mon 22/10 16:30 pm chair: Prof. Dr. Jaime Sichman Alan Kirman This talk will analyse the emergence of collective behaviour which has a certain rationality from the point of view of the aggregate but which does not correspond to that of the individual. In economics it has long been recognised that, as a result of externalities behaviour that is rational from the point of view of the individual may lead to collective disasters as in the case of the Tragedy of the Commons. However, even more interesting are the cases where individuals who interact locally and have only limited information can together produce outcomes which have regularities which could not have been predicted from the individual motives. These may be beneficial but may also, as a result of contagion, lead the system to sudden

7 collapse. Modelling this sort of phenomena usually involves starting with a formal analysis of a very simple case and then simulating a more complex and realistic system to see whether the results for the simple case still hold in the more general case. I will illustrate this with a number of examples, from some observations about the analogies with social insects, and ant and bees in particular, to recent work on the extension of the Schelling model of segregation. In the latter we include preferences for neighborhood income as well as racial composition and introduce a housing market. We analyzed segregation in terms of both race and income and the relation between house prices and these factors. I will also discuss a model of the market for mortgage backed securities and show how the rational reactions of a few individuals to a small change in the probability of default on the underlying assets can lead to a collapse in the price of the assets. In each case the self-organization of a complex adaptive system produces aggregate results which could not have been predicted from the analysis of a representative agent. Title: Computational Models of Dialogical Argumentation Tue 23/10-8:30 am Anthony Hunter Computational models of dialogical argumentation aim to reflect how humans use and resolve conflicting information, opinions, beliefs, etc. by exchanging and analyzing arguments. This may be as part of a wider process such as decision making, sense making, or negotiation. In these formal models, a dialogue between two or more agents is normally made up of a set of communicative acts, called moves, a set of rules that state which moves are legal to make at any point in a dialogue (the protocol), and a set of rules that define the effect of making a move. In addition, each agent may have a private strategy that specifies which of the allowed moves it should do in the current state of the dialogue given its goals. In this talk, we will consider some of the main approaches to modeling dialogical argumentation, and identify some of the many research questions that have been identified.

8 Title: Data Structures for Efficient Inference and Optimization in Expressive Continuous Domains Tue 23/10 16:30 pm Scott Sanner To date, our ability to perform exact closed-form inference or optimization with continuous variables is largely limited to special well-behaved cases. This talk argues that with an appropriate representation and data structure, we can vastly expand the class of models for which we can perform exact, closed-form inference. This enables novel solutions to many problems of interest in AI, machine learning, operations research and beyond, some of which have been unsolved for 50+years. This talk is in two parts. In the first part, I introduce an extension of the algebraic decision diagram (ADD) to continuous variables -- termed the extended ADD (XADD) -- to represent arbitrary piecewise functions (nb, arbitrary pieces, not just hyper-rectangular) over discrete and continuous variables and show how to define and efficiently compute elementary arithmetic operations, integrals, and maximization for various restrictions of these functions. In the second part, I briefly cover a wide range of novel applications where the XADD may be applied: (a) exact inference in expressive discrete and continuous variable graphical models, (b) factored, parameterized linear and quadratic optimization (a generalization of LP and QP solving), (c) exact solutions to piecewise convex functions that enable a number of novel applications in machine learning, and (d) exact solutions to continuous state, action, and observation sequential decision-making problems -- which includes closed-form solutions to previously unsolved problems in operations research. This is joint work with Zahra Zamani & Ehsan Abbasnejad (Australian National University), Karina Valdivia Delgado & Leliane Nunes de Barros (University of Sao Paulo), and Simon Fang (M.I.T.).

9 Title: Evolutionary Computation: Where We are and Where We're Headed Wed 24/10 8:30 pm Prof a. Dr a. Ana Lorena The field of evolutionary computation has experienced a significant growth of interest and activity in the past decade. This has resulted in fresh perspectives and a flurry of new results in both theory and applications. This presentation will summarize the recent progress and characterize some of the remaining unresolved research issues. Kenneth A. De Jong Leandro de Castro Title: Natural Computing: The Grand Challenges and Two Case Studies Wed 24/10 16:30 pm Although the origins of Natural Computing (NC) can be traced back to the early days of Computer Science in the 1940s, it only became a formal discipline around the year Since then, there has been a flood of research papers on the use of Nature to inspire the design of novel algorithms. Now, it is felt that the field faces some Grand Challenges so as to progress and mature. A new perspective of the field has emerged, broadening its scope, and three Grand Challenges have been proposed: to transform Natural Computing into a transdisciplinary science; to harness information processing in natural systems; and to formally and consistently engineer Natural Computing approaches. This talk introduces and discusses the Grand Challenges in Natural Computing Research and then, for showing the pragmatism and usefulness of NC, follows with the presentation of two real-world applications of Natural Computing, one for Social Media and another in e-commerce.

10 Conference Program SCHEDULE Saturday 20/10 Sunday 21/10 Monday 22/10 Tuesday 23/10 8:30-10:00 Workshops Openning Speaker And Speaker Anthony Tutorials Thomas G. Hunter Dietterich 10:00-10:30 Coffee Break 10:30-12:30 BWSS Workshops Technical Technical And Sessions Sessions Tutorials 12:30-14:00 Lunch Time 14:00-16:00 BWSS Workshops And Tutorials Technical Sessions Technical Sessions 16:30-16:30 Coffee Break 16:30-18:00 BWSS Workshops And Tutorials Speaker Alan Kirman Speaker Scott Sanner 18:00-19:00 Openning 19:00-22:00 Cocktail Social Event Dinner Wesdnesday 24/10 Speaker Kenneth A. De Jong Technical Sessions Technical Sessions Speaker Leandro de Castro CEIA and CERN meeting Thursday 25/10 Technical Sessions Technical Sessions Speakers: Gregor Mendel Auditorium Openning Cocktail : Museu Oscar Niemeyer Social Event Dinner: Palácio Giuseppe Garibaldi

11 Tutorials Program 08:30-10:00 10:30-12:30 14:00-16:00 16:30-17:30 Sunday 21th October 2012 Vamos Construir nossos próprios Robôs? ROS - Robot Operating System Um SDK para Robótica e Ambientes Inteligentes Silvia Botelho, Paulo Dews, Luan Silveira, Diones Fischer, Felipe Almeida, Matheus Longaray. Room Lab 1 Bayesian Nonparametrics and its applications. Renato Assunção. Room Lab 2 Coffee Break Vamos Construir nossos próprios Robôs? ROS - Robot Operating System Um SDK para Robótica e Ambientes Inteligentes Silvia Botelho, Paulo Dews, Luan Silveira, Diones Fischer, Felipe Almeida, Matheus Longaray. Room Lab 1 Bayesian Nonparametrics and its applications. Renato Assunção. Room Lab 2 Lunch Time Evolutionary Computation: A Unified Approach Kenneth A. De Jong. Room Lab1 Introduction to Computational Models of Argument Anthony Hunter. Room Lab 2 Coffee Break Machine Learning in Ecological Science and Environmental Policy Thomas G. Dietterich. Room Lab 1 Recent Advances in Continuous Planning Scott Sanner. Room Lab 2

12 CTDIA Program Room Two VIII Best MSc Dissertation/PhD Thesis Contest in Artificial Intelligence Msc Dissertation Sunday 21/10 Revisão de Modelos CTL 8:30-9:10 Paulo T. Guerra Satisfazibilidade Probabilística 9:10-9:50 Glauber De Bona Coffee Break Descoberta de Regras de Conhecimento Utilizando Computação Evolutiva 10:20-11:00 Multi-Objetivo Rafael Giustii Aprendizado não supervisionado de hierarquias de tópicos a partir de 11:00-11:40 coleções textuais dinâmicas Ricardo Marcacini Representação de Coleções de Documentos Textuais por meio de Regras 11:40-12:20 de Associação Rafael Rossi Lunch GRASP with path-relinking for data clustering: a case study for biological 14:00-14:40 data Rafael Frinhani Programming Institutional Facts in Multi-Agent Systems 14:40-15:20 Maiquel Brito Non-Dominance Landscapes for Guiding Diversity Generation in Dynamic 15:20-16:00 Environments Carlos Azevedo Coffee Break Heuristics and Evolutionary Algorithms For Multicast Routing Problem 16:10-16:50 Marcos Luiz de Paula Bueno VIII Best MSc Dissertation/PhD Thesis Contest in Artificial Intelligence PhD Thesis Sunday 21/10 Aprendizado de Máquina em Redes Complexas 10:15-11:00 Fabricio Breve Classifcação de dados estacionários e não estacionários baseada em grafos 11:00-11:45 João Bertini On the Effective Revision of (Bayesian) Logic Programs from Examples 11:45-12:30 Aline Paes Lunch Time Revisão de Crenças em Lógicas de Descrição e em Outras Lógicas não 14:00-14:45 Clássicas, Marcio Ribeiro Verification of Behaviorist Multi-Agent Systems by means of Formally 14:45-15:30 Guided Simulations - PhD Thesis Overview, Paulo Salem IGMN: An Incremental Neural Network Model for On-Line Tasks 15:30-16:15 Paulo Engel

13 WTI Workshop Room One Sunday 21 th October :00-12:00 Palestra - Prof. João Gama(Evolução de comunidades em redes sociais) 12:00-14:00 Lunch Time Categorização de Textos 14:00-14:20 14:20-14:40 14:40-15:00 15:00-15:20 15:20-15:40 15:40-16:00 16:30-16:50 16:50-17:10 17:10-17:30 Mineração de Textos para Organização de Documentos em Centrais de Atendimento. Maria Luiza Passini Classificação Automática de Sítios Web Usando Análise Textual Oeslei Ribas, Celso Kaestner Mineração de Opiniões e Análise de Sentimentos Análise de Sentimento em Citações Científicas para Definição de Fatores de Impacto Positivo Diana Cavalcanti, Ricardo Prudêncio, Shreyasee Pradhan, Jatin Shah, Ricardo Pietrobon Identifying Emotions in Short Texts for Brazilian Portuguese Barbara Martinazzo, Mariza Dosciatti, Emerson Paraiso SAPair: Um Processo de Análise de Sentimento no Nível de Característica Nelson Rocha da Silva, Diego Lima, Flávia Barros Expanding Opinion Lexicon with Domain Specific Opinion Words Using Semi-Supervised Approach Dusan Petkovic, Martin Ringsquandl Coffee Break Redes Complexas Um Modelo Híbrido LTM-ICM para Difusão de Influência em Redes Sociais Anália Cavalcanti, Juliano Rabelo, Flavia Barros, Ricardo Prudêncio Classificação de Alto Nível Utilizando Grafo K-associados Ótimo Murillo Carneiro. João Luís Rosa, Alneu Lopes, Liang Zhao Exploring Data Classification with K-associated Network Thiafo Faleiros, Lilian Berton, Alneu Lopes

14 BWSS Workshop Anfiteatro 10:15-10:45 10:45-12:15 BWSS III Brazilian Workshop on Social Simulation Saturday 20 th October 2012 Opening Session - Chair: Prof. Dr. Graçaliz Dimuro Long presentations Technical Session 1 - Chair: Prof. Dr. Diana Adamatti Elementary Social Functions: Concept and Interrelation to Social Dependence Relations. Antonio Carlos da Rocha Costa, Gracaliz Dimuro, Universidade Federal do Rio Grande, Brazil. Comparing complex networks An application to emergency managers mental models Nuno Caseiro, IPCB, Portugal, Paulo Trigo, Instituto Superior de Engenharia de Lisboa, Portugal. A Study on Socio-Spatial Segregation Models Based on Multi-Agent Systems Tauã Cabreira, Carlos Quadros, Felipe da Silva, Josimara Silveira, Liliane Antiqueira, Suvania Oliveira, Stephanie Brião, Leonardo M. Rodrigues, Universidade Federal do Rio Grande, Brazil. 12:15-14:00 Lunch Time Invited Talk 14:00-16:00 Dr. Armando Geller, chair: Prof. Dr. Antônio Carlos Costa Putting social simulation to use for security professionals and aid and relief practitioners in areas affected by armed conflict 16:00-16:30 Coffee break 16:30-18:00 Long presentations Technical Session 2 - Chair: Prof. Dr. Jaime Sichman Replicating Hofstede's Cultured Negotiation: A First Update João Maria Graça, Faculdade de Psicologia da Universidade de Lisboa, Portugal, Helder Coelho, Universidade de Lisboa, Portugal. How Coalitions Enhance Cooperation in the IPD over Complex Networks Ana Peleteiro, Univ. Vigo, Spain, Juan Carlos Burguillo, Universidad de Vigo, Colombia, Ana Bazzan, UFRGS, Brazil Analyzing the Evolution of Social Exchange Strategies in Social Preference-based MAS through an Evolutionary Spatial Approach of the Ultimatum Game Luís Felipe de Macedo, Universidade Federal de Rio Grande, Brazil, Gracaliz Dimuro, Universidade Federal do Rio Grande, Brazil, Marilton de Aguiar, UFPel, Brazil, Antonio Carlos da Rocha Costa, FURG, Brazil, Helder Coelho, Universidade de Lisboa, Portugal; Viviane Leite Dias de Mattos, Universidade Federal do Rio Grande, Brazil.

15 9:00-10:00 BWSS III Brazilian Workshop on Social Simulation Sunday 21 th October 2012 Short presentations Position Papers - Chair: Prof. Dr. Gracaliz Dimuro Conduct, Cooperation and Alliances: an Analysis of Interaction Among Firms Using an ABS Model Rogerio Morano, Edmilson Moraes, Centro Universitário da FEI, Brazil. A Model for Agent Communication Based on Imprecise Information Using Synonyms Fabio Aiub Sperotto; Diana Adamatti, Universidade Federal do Rio Grande - FURG, Brazil Competitiveness and Social Inclusion within National Programme for Production and Use of Biofuels: negative feedbacks on profitability awareness in sharp Institutional Settlements in Brazil South Region concerning soybean oil. André Zopelari, National Institute for Spatial Research, Brazil; Aldara César, Universidade Federal Fluminense, Brazil Modeling Role Interactions in a Social Organization for the Simulation of the Social Production and Management of Urban Ecosystems: the case of San Jeronimo Vegetable Garden of Seville, Spain Flávia Pereira dos Santos, Universidade Federal de Rio Grande, Brazil, Thiago Rodrigues, FURG, Brazil, Glenda Dimuro, Universidad de Sevilla, España; Diana Adamatti, Universidade Federal do Rio Grande,FURG, Brazil; Gracaliz Dimuro, Universidade Federal do Rio Grande, Brazil, Antonio Carlos da Rocha Costa, FURG, Brazil, Esteban de Manuel Jerez, Universidad de Sevilla, España 10:00-10:30 Coffee Break 10:30-12:30 Panel Chair: Prof. Dr. Helder Coelho Social Simulation: Seeing ahead scanning the future, looking to the tendencies of Computation and Complexity Sciencies, namely Agents and Artificial Intelligence Participants: Prof. Dr. Alan Kirman, Dr. Armando Geller, Prof. Dr. Bruno Reis, Prof. Dr. Jaime Sichman 12:30-14:00 Lunch Time Long presentations 14:00-16:00 Technical Session 3 - Chair: Antonio Carlos da Rocha Costa Stock prices assessment: proposal of a new index based on volume weighted historical prices through the use of computer modeling Tiago Colliri, Fernando Ferreira, University of Sao Paulo, Brazil. An ABM to monitor landscape dynamics and to undertake collective foresight investigations in the Amazon Pierre Bommel, CIRAD & PUC-Rio, France;Poccard-Chapuis René, CIRAD & EMBRAPA, France; Bendahan Amaury, EMBRAPA, Brazil, Coudel Emilie, CIRAD & EMBRAPA, France A multiagent based road pricing approach for urban traffic management Anderson Tavares, Universidade Federal do Rio Grande do Sul, Brazil, Ana Bazzan, UFRGS, Brazil. An Agent-Based Model for the Discussion of Institutions, Welfare and Capitalism Luciano Mattar, Bruno Reis, Universidade Federal de Minas Gerais, Brazil

16 BWSS III Brazilian Workshop on Social Simulation Sunday 21 th October :00-16:30 Coffee Break 16:30-18:00 Long presentations Technical Session 4 - Chair: Diana Adamatti Reduction of Coalition Structure's Search Space based on Domain Information: an Application in Smart Grids Gabriel Ramos, Universidade Federal do Rio Grande do Sul, Brazil;Ana Bazzan, UFRGS, Brazil. An Evolutionary Learning Approach for Robot Path Planning with Fuzzy Obstacle Detection and Avoidance in a Multi-agent Environment Tauã Cabreira, Universidade Federal do Rio Grande, Brazil, Marilton Aguiar, UFPEL, Brazil, Gracaliz Dimuro, Universidade Federal do Rio Grande, Brazil. A new model of agent self-regulation based on profile discovery in social networks applied to the ultimatum game Murian Ribeiro, Universidade Federal de Pelotas, Brazil, Marilton de Aguiar, UFPel, Brazil 16:30-17:20 BWSS III Brazilian Workshop on Social Simulation Monday 22 th October 2012 Invited Talk: Prof. Dr. Alan Kirman, chair: Prof. Dr. Jaime Sichman

17 Summary Technical Sessions 10:30-12:30 14:00-16:00 Monday 22 th October 2012 SBIA: Knowledge Representation SBRN: TS 1 Supervised Machine Learning. ENIA: Aplicações de Inteligência Artificial I. ENIA: Processamento de Linguagem Natural. SBIA: Kowledge Representation, Machine Learning. SBRN: TS 2 Unsupervised Machine Learning. SBRN: TS 3 Bioinformatics. ENIA: Aprendizado de Maquina I. ENIA: Mineração de Dados. Room Gregor Mendel Carlos F. F. da Costa Mário Braga de Abreu Video II Gregor Mendel Carlos F. F. da Costa Room Four Mário Braga de Abreu Video II 10:30-12:30 14:00-16:00 Tuesday 23 th October 2012 SBIA: Machine Learning. SBRN: TS 4 Semi-supervised Learning. ENIA: Inteligência Artificial Distribuída e Sistemas Multiagentes. ENIA: Redes Neurais Artificiais. SBIA: Multiagent Systems. SBRN: TS 5 Neural Networks. SBRN: TS 6 Control and Robotics. ENIA: Aprendizado de Maquina II ENIA: Computação Evolutiva e Metaheurísticas I Room Gregor Mendel Carlos F. F. da Costa Mário Braga de Abreu Video II Gregor Mendel Carlos F. F. da Costa Room Four Mário Braga de Abreu Video II

18 10:30-12:30 14:00-16:00 Wednesday 24 th October 2012 SBIA:Machine Learning and Computer Vision, Constraints SBRN: TS 7 Evolutionary Systems ENIA: Ontologias ENIA: Sistemas de Apoio à Decisão SBIA: Robotics and Language SBRN: TS 8 Swarm Intelligence SBRN: TS 9 Hybrid Intelligent Systems ENIA: Aplicações de Inteligência Artificial II ENIA: Computação Evolutiva e Metaheurísticas II Room Gregor Mendel Carlos F. F. da Costa Mário Braga de Abreu Video II Gregor Mendel Carlos F. F. da Costa Room Four Mário Braga de Abreu Video II 10:30-12:30 Thrusday 25 th October 2012 SBRN: TS 10 Multiobjective Optimization SBRN: TS 11 Pattern Recognition ENIA: Inteligência Computacional ENIA: Planejamento e Escalonamento Automático Room Carlos F. F. da Costa Room Four Mário Braga de Abreu Video II

19 SBIA Program 10:30-11:00 11:00-11:30 11:30-12:00 12:00-12:30 SBIA - XXI Brazilian Symposium on Artificial Intelligence Monday 22 th October 2012 KNOWLEDGE REPRESENTATION On the development of a formal methodology for knowledge representation in defeasible logic programming Alejandro Stankevicius, Marcela Capobianco A Framework for Empirical Evaluation of Belief Change Operators Renato Lundberg, Marcio Ribeiro, Renata Wassermann Sensorimotor Domain Approach for Artificial Autonomous Cognitive Development Mauro Muñoz, Márcio Netto A Service-Oriented Architecture for Assisting the Authoring of Semantic Crowd Maps Henrique Santos, Vasco Furtado 12:30-14:00 Lunch Time 14:00-14:30 14:30-15:00 15:00-15:30 15:30-16:00 User-centric Principles in Automated Decision Making Ingrid Nunes, Simon Miles, Michael Luck, Carlos Lucena MACHINE LEARNING Wearable Computing: Accelerometers Data Classification of Body Postures and Movements Wallace Ugulino, Débora Cardador, Katia Vega, Eduardo Velloso, Ruy Milidiú, Hugo Fuks Density Based Pattern Discovery in Distributed Time Series Josenildo da Silva, Omar Cortes, Gustavo Oliveira, Matthias Klusch Filter Approach Feature Selection Methods to Support Multi-label Learning Based on Relief and Information Gain Newton Spolaôr, Everton Alvares Cherman, Maria Carolina Monard, Huei Lee

20 SBIA - XXI Brazilian Symposium on Artificial Intelligence Tuesday 23 th October 2012 MACHINE LEARNING Automatic Analysis of Leishmania Infected Microscopy Images via 10:30-11:00 Gaussian Mixture Model Pedro Nogueira, Luís Teófilo 11:00-11:30 Link prediction in complex networks based on cluster information Jorge Carlos Valverde Rebaza, Alneu Lopes 11:30-12:00 A Parallel Approach to Clustering with Ant Colony Optimization Guilherme Ramos On the Use of Consensus Clustering for Incremental Learning of Topic 12:00-12:30 Hierarchies Ricardo Marcacini, Eduardo Hruschka, Solange Rezende 12:30-14:00 Lunch Time MULTIAGENT SYSTEMS 14:00-14:30 Orchestrating Multiagent Learning of Penalty Games Ana Bazzan 14:30-15:00 An Architectural Model for Autonomous Normative Agents Baldoino Santos Neto, Viviane Torres da Silva, Carlos Lucena A coalition formation mechanism for trust and reputation-aware multiagent systems 15:00-15:30 Bruno Hoelz, Celia Ralha Profile Management and Recommendation in Communities of Practice 15:30-16:00 based on Multiagent Systems João Luis Tavares da Silva, Sidinei Lubenow, Alexandre Ribeiro SBIA - XXI Brazilian Symposium on Artificial Intelligence Wednesday 24 th October 2012 MACHINE LEARNING and COMPUTER VISION Image retrieval by content based on a visual attention model and genetic 10:30-11:00 algorithms Milton Heinen, Paulo Engel A Symbolic Representation Method to Preserve the Characteristic Slope of 11:00-11:30 Time Series Willian Zalewski, Fabiano Silva, Feng Chung Wu, Huei Lee, André Gustavo Maletzke CONSTRAINTS 11:30-12:00 A New Ecology-based Heterogeneous Approach for Cooperative Search Rafael Parpinelli, Heitor Lopes Providing trade-off techniques subsets to improve software testing 12:00-12:30 effectiveness: Using evolutionary algorithm to support software testing techniques selection by a web tool Aurélio Grande, Arilo Dias Neto, Rosiane Rodrigues 12:30-14:00 Lunch Time ROBOTICS and LANGUAGE Knowledge-Intensive Word Disambiguation via Common-Sense and 14:00-14:30 Wikipedia Vasco Furtado, Vladia Pinheiro, Lívio Melo Freire, Caio Ferreira 14:30-15:00 Context-sensitive ASR for controlling the navigation of mobile robots Gabriel Araújo, Hendrik Macedo An evaluation of the model of stigmergy in a RoboCup Rescue multiagent 15:00-15:30 system Gabriel Rigo da Cruz Jacobsen, Carlos Barth, Fernando dos Santos

21 SBRN Program 10:30-11:00 11:00-11:30 11:30-12:00 12:00-12:30 SBRN - XII Brazilian Symposium on Neural Network Monday 22 th October 2012 TS 1: Supervised Machine Learning Chair: André C. P. L. F. Carvalho Combining Meta-Learning with Multi-Objective Particle Swarm Algorithms for SVM Parameter Selection: An Experimental Analysis Péricles Miranda (Universidade Federal Rural Pernambuco), Ricardo Prudêncio (Universidade Federal de Pernambuco), André Ponce de Leon F de Carvalho (Universidade de São Paulo), Carlos Soares (Universidade do Porto) Meta-learning for Periodic Algorithms Selection in Time-changing Data André Rossi (Universidade de São Paulo), André Ponce de Leon F de Carvalho (Universidade de São Paulo), Carlos Soares (Universidade do Porto) A Study on Class Noise Detection and Elimination Luís Paulo Garcia (Universidade de São Paulo), Ana Carolina Lorena (Universidade Federal do ABC, Universidade Federal de São Paulo), André Ponce de Leon F de Carvalho (Universidade de São Paulo) Negative Selection with High-dimensional Support for Keystroke Dynamics Paulo Pisani (Universidade Federal do ABC), Ana Carolina Lorena (Universidade Federal do ABC, Universidade Federal de São Paulo) 12:30-14:00 Lunch Time TS 2: Unsupervised Machine Learning Chair: Murilo C. Naldi 14:00-14:30 14:30-15:00 15:00-15:30 15:30-16:00 14:00-14:30 Variable-Wise Kernel-Based Clustering Algorithms for Interval-Valued Data Francisco De Carvalho (Universidade Federal de Pernambuco), Marcelo Ferreira (Universidade Federal da Paraiba), Gibson Barbos (Universidade Federal de Pernambuco) An Energy Exchanging Mechanism for Data Clustering Roberto Gueleri (Universidade de São Paulo), Liang Zhao (Universidade de São Paulo) A Density-Based Clustering Approach for Behavior Change Detection in Data Streams Rosane Vallim (Universidade de São Paulo), Jose Augusto Andrade Filho (Universidade de São Paulo), André Ponce de Leon F de Carvalho (Universidade de São Paulo), Joao Gama (Universidade do Porto) Combining Information from Distributed Evolutionary k-means Murilo Naldi (Universidade Federal de Viçosa), Ricardo Campello (Universidade de São Paulo) TS 3: Bioinformatics Chair: Marcilio C. P. Souto A comparison of external clustering evaluation indices in the context of imbalanced data sets Marcilio de Souto (Universidade Federal de Pernambuco), André Coelho (Universidade de Fortaleza), Katti Faceli (Universidade Federal de São Carlos), Tiemi Sakata (Universidade Federal de São Carlos),

22 14:30-15:00 15:00-15:30 15:30-16:00 Ivan Gesteira Costa (Universidade Federal de Pernambuco) On the ensemble prediction of gene regulatory networks: a comparative study Mariana Mendoza (Universidade Federal do Rio Grande do Sul), Ana Bazzan (Universidade Federal do Rio Grande do Sul) Predicting gene functions using semi-supervised clustering algorithms with objective function optimization Valmir Macário (Universidade Federal de Pernambuco), Ivan Gesteira Costa (Universidade Federal de Pernambuco), Joao Oliveira (Universidade Federal de Pernambuco), Francisco De Carvalho (Universidade Federal de Pernambuco) An epsilon-greedy mutation operator based on prior knowledge for GA convergence and accuracy improvement: an application to networks inference Mariana Mendoza (Universidade Federal do Rio Grande do Sul), Adriano Werhli (Universidade Federal do Rio Grande), Ana Bazzan (Universidade Federal do Rio Grande do Sul) 10:30-11:00 11:00-11:30 11:30-12:00 12:00-12:30 SBRN - XII Brazilian Symposium on Neural Network Tuesday 23 th October 2012 TS 4: Semi-supervised Learning Chair: Liang Zhao A Semi-Supervised Approach to Estimate the Number of Clusters per Class Davidson Sestaro (Universidade Católica de Santos), Thiago Covões (Universidade de São Paulo), Eduardo Hruschka (Universidade de São Paulo) Particle Competition and Cooperation to Prevent Error Propagation from Mislabeled Data in Semi-Supervised Learning Fabricio Breve (Universidade Estadual Paulista), Liang Zhao (Universidade de São Paulo) Robustness Analysis of Network-Based Semi-Supervised Learning Algorithms Bilzã Araújo (Universidade de São Paulo), Liang Zhao (Universidade de São Paulo) Using Interacting Forces to Perform Semi-Supervised Learning Thiago Cupertino (Universidade de São Paulo), Liang Zhao (Universidade de São Paulo) 12:30-14:00 Lunch Time TS 5: Neural Networks Chair: Teresa Ludermir 14:00-14:30 14:30-15:00 15:00-15:30 15:30-16:00 Firing activity induced by nonidentical signal phases in two coupled excitable neurons Xiaoming Liang (Universidade de São Paulo) Zhao Liang (Universidade de São Paulo) On the universality of quantum logical neural networks Adenilton da Silva (Universidade Federal de Pernambuco), Teresa Ludermir (Universidade Federal de Pernambuco), Wilson de Oliveira (Universidade Federal Rural de Pernambuco) Hardware/Software Co-design Implementation of On-Chip Backpropagation Mauricio Dias (Universidade de São Paulo), Fernando Osorio (Universidade de São Paulo) A Hybrid RNN Model for Cursive Offline Handwriting Recognition Byron Leite Dantas Bezerra (Universidade Federal de Pernambuco), Cleber Zanchettin (Universidade Federal de Pernambuco),