Particle Competition and Cooperation in Networks for Semi-Supervised Learning

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Particle Competition and Cooperation in Networks for Semi-Supervised Learning with Concept Drift

Concept drift is a problem of increasing importance in machine learning and data mining. Data sets under analysis are no longer only static databases, but also data streams in which concepts and data distributions may not be stable over time. However, most learning algorithms produced so far are based on the assumption that data comes from a fixed distribution, so they are not suitable to handle concept drifts. Moreover, some concept drifts applications requires fast response, which means an algorithm must always be (re) trained with the latest available data. But the process of labeling data is usually expensive and/or time consuming when compared to unlabeled data acquisition, thus only a small fraction of the incoming data may be effectively labeled. Semi-supervised learning methods may help in this scenario, as they use both labeled and unlabeled data in the training process. However, most of them are also based on the assumption that the data is static. Therefore, semi-supervised learning with concept drifts is still an open challenge in machine learning. Recently, a particle competition and cooperation approach was used to realize graph-based semi-supervised learning from static data. In this paper, we extend that approach to handle data streams and concept drift. The result is a passive algorithm using a single classifier, which naturally adapts to concept changes, without any explicit drift detection mechanism. Its built-in mechanisms provide a natural way of learning from new data, gradually forgetting older knowledge as older labeled data items became less influent on the classification of newer data items. Some computer simulation are presented, showing the effectiveness of the proposed method.

Particle competition and cooperation to prevent error propagation from mislabeled data in semi-supervised learning

Semi-supervised learning is applied to classification problems where only a small portion of the data items is labeled. In these cases, the reliability of the labels is a crucial factor, because mislabeled items may propagate wrong labels to a large portion or even the entire data set. This paper aims to address this problem by presenting a graph-based (network-based) semi-supervised learning method, specifically designed to handle data sets with mislabeled samples. The method uses teams of walking particles, with competitive and cooperative behavior, for label propagation in the network constructed from the input data set. The proposed model is nature-inspired and it incorporates some features to make it robust to a considerable amount of mislabeled data items. Computer simulations show the performance of the method in the presence of different percentage of mislabeled data, in networks of different sizes and average node degree. Importantly, these simulations reveals the existence of the critical points of the mislabeled subset size, below which the network is free of wrong label contamination, but above which the mislabeled samples start to propagate their labels to the rest of the network. Moreover, numerical comparisons have been made among the proposed method and other representative graph-based semi-supervised learning methods using both artificial and real-world data sets. Interestingly, the proposed method has increasing better performance than the others as the percentage of mislabeled samples is getting larger. © 2012 IEEE.

Fuzzy community structure detection by particle competition and cooperation

Identification and classification of overlapping nodes in networks are important topics in data mining. In this paper, a network-based (graph-based) semi-supervised learning method is proposed. It is based on competition and cooperation among walking particles in a network to uncover overlapping nodes by generating continuous-valued outputs (soft labels), corresponding to the levels of membership from the nodes to each of the communities. Moreover, the proposed method can be applied to detect overlapping data items in a data set of general form, such as a vector-based data set, once it is transformed to a network. Usually, label propagation involves risks of error amplification. In order to avoid this problem, the proposed method offers a mechanism to identify outliers among the labeled data items, and consequently prevents error propagation from such outliers. Computer simulations carried out for synthetic and real-world data sets provide a numeric quantification of the performance of the method. © 2012 Springer-Verlag.

Particle competition and cooperation for semi-supervised learning with label noise

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Segurança e defesa no Cone Sul: Transições com transformações?

The new conditions of international security don't only attach importance to sovereignty and interstate relations. This new framework has led to changes in the scope and requirements for Politics of Defense. Countries of the Southern Cone seek to establish new parameters in the definition of their policies, but need to define between dissuasive policies, as in the past, or establish mechanisms for cooperation with their neighbors. Furthermore, these policies, by their nature and size, are not exclusively militaries. The purpose of this article is to examine the definitions constraints on strategic deterrence and cooperation in international security, and identify the degree of autonomy and the nature of the concepts derived from the armed forces in shaping Politics of Defense. Finally, we try to check the level of civil policy direction in defining those policies.

Aquaculture Research and Development as an Entry-Point and Contributor to Natural Resources and Coastal Management

Recent, fervent international dialogue concerning the existence and magnitude of impacts associated with aquaculture has had both positive and negative outcomes. Aquaculture stakeholders have become sensitized to requirements for improved environmental management of aquaculture. on the other hand, in some cases aquaculture development has been negatively affected by some of the unwarranted and unproved allegations to the detriment of the stakeholders most in need of aquaculture development (i.e., resource users, particularly the poor, who are dependent on natural resources). These resource users are targeted by, and directly influence biodiversity and conservation agendas; hence the need to understand how to gain their active participation. This discussion focuses on examples of how aquaculture research and development can be a useful tool or strategy for resource management initiatives and provide tangible positive including increased stakeholder participation and cooperation, offering alternatives to resource extraction and use in otherwise difficult or intransigent resource management conflicts.

O agente comunitário de saúde frente ao processo de trabalho em equipe: facilidades e dificuldades

O estudo propõe-se a analisar as dificuldades e facilidades dos Agentes Comunitários de Saúde (ACS) frente ao trabalho em equipe. A análise pautou-se na perspectiva hermenêutica-dialética, tendo como referência os princípios do método de interpretação dos sentidos. As dificuldades e facilidades apontadas por eles revelam que trabalhar em equipe demanda relações efetivas, com ênfase na comunicação, respeito e cooperação, sendo as reuniões de equipe estratégia importante para isso. Depreende-se a necessidade de constantes investimentos nas relações entre os membros da equipe.

Espaço geográfico, produção e movimento: uma reflexão sobre o conceito de circuito espacial produtivo

O propósito deste artigo é demonstrar a importância e a operacionalidade dos conceitos de circuito espacial produtivo e círculos de cooperação no espaço, num momento histórico em que as esferas da produção e da troca tornam-se geograficamente mais dispersas, fazendo da circulação uma prioridade e um campo de atuação estratégica de Estados e empresas. Procura-se alcançar esse objetivo em três passos: 1) elaboração de uma discussão teórica com base nos autores que propuseram e procuraram desenvolver os conceitos; 2) proposição de uma distinção entre circuito espacial produtivo e cadeia produtiva; 3) uma breve análise da produção agrícola moderna em áreas de fronteira agrícola no território brasileiro à luz dessa teoria.

O papel da escola e do professor numa situação de não aprendizagem: o que dizem as crianças

O artigo apresenta dados parciais de uma pesquisa sobre a construção do conhecimento social a partir da perspectiva piagetiana, mais especificamente as ideias das crianças a respeito da escola e do professor. Os participantes do estudo foram 52 crianças entre 7 e 8 anos inseridas em ambientes educacionais diferenciados: um considerado como ambiente tradicional de ensino e o outro considerado como ambiente sociomoral construtivista. O instrumento apresentado aqui, utilizado para coleta de dados, é uma história envolvendo uma situação de não aprendizagem. Os participantes eram convidados a pensar sobre as questões inerentes à história, bem como o papel da escola e do professor na situação proposta. Os dados indicaram não haver diferença entre os dois ambientes no que se refere à construção desse conhecimento social. No entanto, houve diferença muito significativa na maneira utilizada pelos alunos para resolverem os problemas da história: no ambiente tradicional a coerção e a expiação foram mais mencionadas e no ambiente sociomoral construtivista, o diálogo e a cooperação. Os dados apontam ainda para a necessidade do trabalho com esse tipo de conhecimento em sala de aula, visto que as respostas dos sujeitos caracterizaram-se por uma compreensão parcial da realidade, centrada em aspectos mais visíveis e aparentes dos fatos e na não consideração de processos ocultos.

Programa de controle de surto de escorpião Tityus serrulatus, Lutz e Mello 1922, no município de Aparecida, SP (Scorpiones, Buthidae).

A scorpion control program was proposed for the town of Aparecida (SP), an endemic region of Tityus serrulatus. Clusters of scorpions in urban and rural areas, environmental degradation of the town's outskirts and new scorpion procreation and dispersal habitats were studied. In addition, infrastructure problems such as the disposal and collection of residential and municipal refuse, sanitation (sewage and storm sewer), condition of vacant lots and constructions in the urban area were evaluated. After an epidemiological study, educational measures such as the distribution of pamphlets, cleaning group work, visits to residences and cooperation from High School teachers and students were also suggested. Chemical control was indicated in high-risk sites, especially those of near-school buildings. Furthermore, the use of natural predators was also mentioned within the present sanitation regulations for urban areas. The authors assert that these procedures must be integrated and continued uninterruptedly for several years. They also suggest a collaborative work with those responsible for the dengue eradication program, as well as the institution of the scorpion study week, which would greatly contribute to the education of the population, to preventive programs and to scorpion control.

Analysing green supply chain management practices in Brazil's electrical/electronics industry using interpretive structural modelling

Industries need to adopt the environmental management concepts in the traditional supply chain management. The green supply chain management (GSCM) is an established concept to ensure environment-friendly activities in industry. This paper identifies the relationship of driving and dependence that exists between GSCM practices with regard to their adoption within Brazilian electrical/electronic industry with the help of interpretive structural modelling (ISM). From the results, we infer that cooperation with customers for eco-design practice is driving other practices, and this practice acts as a vital role among other practices. Commitment to GSCM from senior managers and cooperation with customers for cleaner production occupy the highest level. © 2013 Copyright Taylor and Francis Group, LLC.

Experiência em arte-educação: importâncias, relações e sentidos

Pós-graduação em Artes - IA

Geografia de redes e da logística no transporte rodoviário de cargas: fluxos e mobilidade geográfica do capital

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)