DETAILED PROGRAM - BRACIS 2016

BRACIS Program 10th October (Monday) 11th October (Tuesday) 12th October (Wednesday)
08:30h-10:00h Bracis 1A ML/DM1
4 papers
Bracis 1B Planning
4 papers
Plenary Poster Session
Poster Session
19 papers
Plenary Talk - Roberto Santana
Coffee break
10:15h-10:30h Opening Ceremony
10:30h-12:30h Best Papers
6 papers
Bracis 3A Evol Comp
6 papers
Bracis 3B Agents-MAS
6 papers
Bracis 5A Search & Swarm
6 papers
Bracis 5B KR Reasoning
4 papers
Lunch
13:45h-14:00h Industrial Panel
14:00h-15:30h Plenary Talk - João Gama Plenary Talk - Eduardo Hruschka
15:30h-15:45h
Coffee break
16:00h-18:00h Bracis 2A Learning NN
6 papers
Bracis 2B IA/IC Applications
6 papers
Bracis 4A Pattern R. Cluster
6 papers
Bracis 4B NLP
6 papers
Bracis 6 ML/DM 2
5 papers
18:00h-19:00h CEIA and CEIC meeting
19:00h-... Dinner

Monday – 10th October


Session Bracis 1A - Machine Learning and Data Mining 1 (Monday – 10th October 8:30 – 10:00)

Chair: André Carvalho
Hyper-parameter Tuning of a Decision Tree Induction Algorithm
Rafael Gomes Mantovani, Tomas Horvath, Ricardo Cerri, André Ponce de Leon F de Carvalho, Joaquin Vanschoren
An Accurate Gaussian Process-Based Early Warning System for Dengue Fever
Julio Albinati Cortez, Wagner Meira Jr., Gisele Pappa
Supervised Microalgae Classification in Imbalanced Dataset
Iago Corrêa, Paulo Drews Jr, Márcio Silva de Souza, Virginia Tavano
Automatic selection of learning bias for active sampling
Davi Santos, André Ponce de Leon F de Carvalho

Session Bracis 1B - Planning and Scheduling (Monday – 10th October 8:30 – 10:00)

Chair: Leliane Barros
Solving Atomix with Pattern Databases
Alex Gliesch, Marcus Ritt
Using the Causal Graph to enhance Translations to solve Contingent Planning Problems
Ignasi Andres, Leliane Nunes de Barros
Improved Airport Ground Traffic Control with Domain-Dependent Heuristics
Augusto B. Corrêa, André Grahl Pereira, Marcus Ritt
Extreme Risk Averse Policy for Goal-Directed Risk-Sensitive Markov Decision Process
Valdinei Freire, Karina Valdivia-Delgado

Session Bracis - Best Papers (Monday – 10th October 10:30 - 12:30)

Chairs: Anne Canuto and Heloisa Camargo
(Deep) Learning from Frames
Jônatas Wehrmann, Rodrigo Barros, Gabriel Simões, Thomas Paula, Duncan Ruiz
Investigating selection strategies in multi-objective probabilistic model based algorithms
Andrei Strickler, Olacir Castro Júnior, Aurora Pozo, Roberto Santana
Knowledge Representation for Argumentation in Agent-Oriented Programming Languages
Alison Panisson, Rafael Bordini
Object-Oriented Reinforcement Learning in Cooperative Multiagent Domains
Felipe Silva, Ruben Glatt, Anna Costa
Tree-based Grammar Genetic Programming to Evolve Particle Swarm Algorithms
Péricles Miranda, Ricardo Prudêncio
Using Preferences over Sources of Information in Argumentation-Based Reasoning
Alison Panisson, Victor Melo, Rafael Bordini

Plenary Talk – João Gama (Monday – 10th October 14:00 – 15:30)

Title: Real-Time Data Mining
Abstract:

Nowadays, there are applications in which the data are modelled best not as persistent tables, but rather as transient data streams. In this keynote, we discuss the limitations of current machine learning and data mining algorithms. We discuss the fundamental issues in learning in dynamic environments like learning decision models that evolve over time, learning and forgetting, concept drift and change detection. Data streams are characterized by huge amounts of data that introduce new constraints in the design of learning algorithms: limited computational resources in terms of memory, processing time and CPU power. In this talk, we present some illustrative algorithms designed to taking these constrains into account. We identify the main issues and current challenges that emerge in learning from data streams, and present open research lines for further developments.

Session Bracis 2A - Learning and Neural Networks (Monday – 10th October 16:00 – 18:00)

Chair: Anna Costa
Extreme Learning Machines for Datasets with Missing Values Using the Unscented Transform
Diego Mesquita, João Paulo Pordeus, Leonardo Ramos Rodrigues
Towards Knowledge Transfer in Deep Reinforcement Learning
Ruben Glatt, Felipe Silva, Anna Costa
Co-MLM: a SSL algorithm based on the Minimal Learning Machine
Weslley Caldas, Joao Gomes, Michelle Cacais, Diego Mesquita
Evaluating Binary Encoding Techniques for WiSARD
Fabio Rangel, Fabricio Faria, Jonice Oliveira, Priscila M V Lima, Karine Da Costa, Andressa Kappaun
A Voronoi Diagram Based Classifier for Multiclass Imbalanced Data Sets
Evandro Silva, Cleber Zanchettin
Determining the Structure of Decision Directed Acyclic Graphs for Multiclass Classification Problems
Thaise Quiterio, Ana Carolina Lorena

Session Bracis 2B - IA/IC Applications (Monday – 10th October 16:00 – 18:00)

Chair: Fábio Cozman
Pixel clustering for face recognition
Tiago de Carvalho, Maria Sibaldo, Tsang Ing Ren, George Cavalcanti, Ing Jyh Tsang, Jan Sijbers
No-Reference Image Quality Assessment Using Texture Information Banks
Pedro Garcia Freitas, Mylene Farias, Welington Akamine
Image-Set Matching by Two Dimensional Generalized Mutual Subspace Method
Bernardo Gatto, Eulanda Santos
Image Classification Using Sum-Product Networks for Autonomous Flight of Micro Aerial Vehicles
Bruno Massoni Sguerra, Fábio Cozman
Fault Detection in Hard Disk Drives Based on Mixture of Gaussians
Felipe Brito, Lucas Queiroz, Javam Machado, João Paulo Pordeus, Iago Chaves, Caio Rodrigues
Musical Scales Recognition via Deterministic Walk in a Graph
Andrés Eduardo Coca Salazar, Liang Zhao

Tuesday – 11th October


Bracis - Poster Session (Tuesday – 11th October 8:30 – 10:00)

Chairs: Myriam Delgado and Renata Vieira
Towards fixation prediction: a nonparametric estimation-based approach through key-points
Saulo Oliveira, Ajalmar Rocha Neto, João Paulo Pordeus
Efficient Minimal Learning Machines with Reject Option
Adonias Caetano de Oliveira, João Paulo Pordeus, Ajalmar Rocha Neto, Amauri Holanda
A Concept-based Integer Linear Programming Approach for Single-Document Summarization
Hilário Tomaz de Oliveira, Rinaldo Lima, Rafael Dueire Lins, Fred Freitas, Marcelo Riss, Steven Simske
Physical Topology Design of Optical Networks Aided by Many-Objective Optimization Algorithms
Elliackin Messias Figueiredo, Danilo Araújo, Carmelo Bastos-Filho, Teresa Ludermir
A New Estimation Distributed Algorithm Applied to a Many-Objective Discrete Optimization Problem
Glauber Botelho, André Britto, Leila Silva
A Dynamic Gesture Recognition System to Translate Between Sign Languages in Complex Backgrounds
Davi Neiva, Cleber Zanchettin
BaNHFaP: A Bayesian Network based Failure Prediction Approach for Hard Disk Drives
Iago Chaves, Manoel Rui Pessoa de Paula, Lucas Gonçalves Moura Leite, Lucas Queiroz, João Paulo Pordeus, Javam Machado
Shrinkage k-means: A clustering algorithm based on the James-Stein Estimator
Filipe Damasceno, Marcelo Veras, Diego Mesquita, Joao Gomes, Carlos Eduardo Fish de Brito
Multiagent Circumnavigation by Reactive Descentralized Algorithm
Edson Silva, Luiz Antonio Celiberto Junior, Robson França
Ensembles with clustering-and-selection model using evolutionary algorithms
Leandro Almeida, Pedro Galvão
Building ensembles with classifier selection using self-organising maps
Leandro Almeida, Cleber Zanchettin, Hilton Leite
Automated Left Ventricle Posterior Wall Segmentation Using Kohonen Self-Organizing Map
Salety Baracho, Vinicius Melo, Regina Célia Coelho
Transforming Multi-Agent Planning Into Single-Agent Planning Using Best-cost Strategy
Leonardo Moreira, Celia Ralha
On Validation Setup for Multiclass Imbalanced Data Sets
Evandro Silva, Cleber Zanchettin
ASAClu: Selecting diverse and relevant clusters
Joao Almeida, Tiemi Sakata, Katti Faceli
The Role of Discount Factor in Risk-Sensitive Markov Decision Processes
Valdinei Freire
Multi-class Emotions classification by Sentic Levels as features in Sentiment Analysis
Alex Marino de Almeida, Sylvio Barbon Junior, Emerson Cabrera Paraiso
A hybrid strategy to evolve cellular automata rules with a desired dynamical behavior applied to the task scheduling problem
Tiago Ismailer de Carvalho, Murillo Carneiro, Gina Oliveira
First-Order Logic for Decision Problems with Preference Aggregation
Arnaldo Lima Júnior, Ana Teresa Martins, Davi Vasconcelos

Session Bracis 3A - Evolutionary Computation (Tuesday – 11th October 10:30 - 12:30)

Chair: Gina Oliveira
Metaheuristic Design Pattern: Visitor for Genetic Operators
Giovani Guizzo, Silvia Vergilio
Revisiting the Sequential Symbolic Regression Genetic Programming
Luiz Otávio Oliveira, Fernando Otero, Luis Fernando Miranda, Gisele Pappa
Time series imputation using genetic programming and Lagrange interpolation
Fabio Lobato, Ádamo Santana, Damares Oliveira de Resende
Discovery Biological Motifs using Heuristics Approaches
Jader Caldonazzo Garbelini, Andre Kashiwabara, Danilo Sanches
Adaptive Operator Selection in NSGA-III
Richard Gonçalves, Carolina Paula de Almeida, Lucas Pavelski, Sandra Venske, Josiel Kuk, Aurora Pozo
Many-objective Evolutionary Algorithms for Multicast Routing with Quality of Service
Thiago Lafetá, Marcos Luiz de Paula Bueno, Christiane Brasil, Gina Oliveira

Session Bracis 3B - Agents and Multiagent Systems (Tuesday – 11th October 10:30 - 12:30)

Chair: Patricia Tedesco
A Trust and Reputation Framework for Game Agents: Providing a Social bias to Computer Players
Joao Carlos da Silva, Carla Delgado, Fábio Couto
Allocating Social Goals Using the Contract Net Protocol in Online Multi-Agent Planning
Rafael Cardoso, Rafael Bordini
Evaluating the SBR Algorithm using Automatically Generated Plan Libraries
Giovani Farias, Lucas Welter Hilgert, Felipe Meneguzzi, Rafael Bordini
Evaluating the Impact of Reputation-based Agents in Social Coalition Formation
Cristina Verçosa Barrios de Souza, Fabricio Enembreck
Reasoning about Trust and Belief in Possibilistic Answer Set Programming
João Alcântara, Gabriel Amaral
A proposal for tuning the alpha parameter in a copula function applied in fuzzy rule-based classification systems
Giancarlo Lucca, Gracaliz Dimuro, Benjamin Bedregal, Humberto Bustince, José Antonio Sanz

Session Industrial Panel (Tuesday – 11th October 13:45 – 15:45)

Abstract: Industrial meeting on artificial intelligence and data science in industry.

Session Bracis 4A - Pattern Recognition and Cluster Analysis (Tuesday – 11th October 16:00 – 18:00)

Chair: Liang Zhao
Allocation of Volunteers in Non-Governmental Organizations Aided by Non-Supervised Learning
Carlos M. M. Bezerra, Danilo Araújo, Valmir Macário
Estimating the Number of Clusters Based on Sequential Clustering Algorithms
Eduardo Real
Random Walk in Feature-Sample Networks for Semi-Supervised Classification
Filipe Verri, Liang Zhao
Preserving Context between Classification Evaluation and Data Exploration with RadViz
Igor Corrêa, André Ponce de Leon F de Carvalho
Anomaly Detection Under Cost Constraint
Bruno Barbarioli, Renato Assuncao
A Coral Reefs Optimization Algorithm for Cluster Ensembles Optimization
Huliane Medeiros, Anne Canuto, João C. Xavier-Júnior, Inacio Medeiros

Session Bracis 4B - Natural Language Processing (Tuesday – 11th October 16:00 – 18:00)

Chair: Evandro Costa
A method for automatic determination of the feature vector size for text categorization
Rogério Fragoso, Roberto Pinheiro, George Cavalcanti
Discriminating beetwen Brazilian and European Portuguese National Varieties on Twitter Texts
Dayvid Castro, Ellen Polliana Ramos Souza, Adriano Oliveira
Automatic Identification of Security Risks in Edicts for Software Procurement
Rodrigo Peclat, Guilherme Ramos
Topic Modeling for Short Texts with Co-occurrence Frequency-based Expansion
Marcelo Pita, Gabriel Miranda Pedrosa, Paulo Viana Bicalho, Anisio Lacerda, Gisele Pappa
Exploring unsupervised features in Conditional Random Fields for Spanish Named Entity Recognition
Jenny Copara Zea, José Eduardo Ochoa Luna, Camilo Thorne, Goran Glava
Investigating How “Good” Characteristics’ Presence are Related with Questions’ Performance
Cleyton Souza, Evandro Costa, Joseana Fechine, Franck Aragão, José Remígio

Wednesday – 12th October


Plenary Talk – Roberto Santana (Wednesday – 12th October 8:30 – 10:00)

Title: Estimation of distribution algorithms: Competent evolutionary algorithms, knowledge extraction techniques, or an arena for testing problem structural hypotheses and machine learning methods?
Abstract:

Estimation of distribution algorithms (EDAs) were originally conceived as a more efficient class of evolutionary approaches based on building probabilistic models of the search space. Traditionally, machine learning methods are used to learn these models from the solutions evaluated along the evolution, and to sample new promising solutions from the models. EDAs have been successfully applied to a variety of practical problems in Bioinformatics, energy production, industrial scheduling, etc. This talk discusses recent developments, extensions, and relevant results on the theory and application of EDAs. In particular, results on the application of these algorithms to unveil previously unknown information about the structure of the addressed optimization problem are reviewed. Examples of EDAs applications to real-world optimization problems are discussed. The link between EDAs, other model-based evolutionary algorithms, and research areas such as knowledge-discovery, data mining, and transfer learning are covered in the presentation. Finally, some open and challenging problems on EDAs are briefly presented.

Session Bracis 5A - Search and Swarm Intelligence (Wednesday – 12th October 10:30 – 12:30)

Chair: Aurora Pozo
A Novel Context-Free Grammar to Guide the Construction of Particle Swarm Optimization Algorithms
Péricles Miranda, Ricardo Prudêncio
Data Clustering Using Group Search Optimization with Alternative Fitness Functions
Luciano Pacífico, Teresa Ludermir
An Unsupervised Particle Swarm Optimization Approach for Opinion Clustering
Ellen Polliana Ramos Souza, Adriano Oliveira, Diego Santos, Gustavo Oliveira, Alisson Silva
Tabu Search with Ejection Chain for the Biobjective Adjacent-only Quadratic Spanning Tree
Sílvia Maia, Elizabeth Goldbarg, Lucas Pinheiro, Marco Goldbarg
Novelty Detection Based on Genuine Normal and Artificially Generated Novelty Examples
George Cabral, Adriano Oliveira
A Pareto-based Symbiotic Relationships Model for Unconstrained Continuous Optimization
Rafael Parpinelli, Leandeson André

Session Bracis 5B - Knowledge Representation and Reasoning (Wednesday – 12th October 10:30 – 12:30)

Chair: Fred Freitas
An Ontology for Property Crime based on Events from UFO-B Foundational Ontology
Cleyton Rodrigues, Fred Freitas, Ryan Ribeiro de Azevedo
Markov Decision Processes Specified by Probabilistic Logic Programming: Representation and Solution
Thiago Bueno, Denis Mauá, Leliane Nunes de Barros, Fábio Cozman
A module-based approach for evaluating differential genome-wide expression profiles
José Flávio de Souza Dias Júnior, Ronnie Alves, Thérèse Commes
Finding Inference Rules using Graph Mining in Ontological Knowledge Bases
Lucas Navarro, Estevam Hruschka Jr., Ana Paula Appel

Session Bracis 6 - Machine Learning and Data Mining 2 (Wednesday – 12th October 16:00 – 18:00)

Chair: Cleber Zanchettin
Authorship attribution via network motifs identification
Vanessa Marinho, Diego Amancio, Graeme Hirst
SSDP: A Simple Evolutionary Approach for Top-K Discriminative Patterns in High Dimensional Databases
Tarcísio Lucas, Renato Vimieiro,Teresa Ludermir
A Comparative Study of Group Profiling Techniques in Co-Authorship Networks
João Emanoel Ambrósio Gomes, Ricardo Prudêncio, Andre Camara
Fusion Approaches of Feature Selection Algorithms for Classification Problems
Jhoseph Jesus, Daniel Araújo, Anne Canuto
A Meta-Learning Method to Select Under-Sampling Algorithms for Imbalanced Data Sets
Péricles Miranda, Romero Morais, Ricardo de Abreu Silva

Promotion