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.

Interference competition between ants (Hymenoptera: Formicidae) in Amazonian clearings

The first case of interference competition through soil dumping in South America is documented between Ectatomma quadridens and Pheidole fallax in Amazonian forest clearings. Workers of the diurnally active E. quadridens arrive at nests of P. fallax at dawn, and begin to fill up nest entrances with soil. During the day, E. quadridens workers remain stationary on the closed nest of P. fallax, and fill soil at the first signs of nest openings. Colonies of P. fallax distant from E. quadridens nests are active for 24 hrs; those near E. quadridens nests are limited for foraging nocturnally after opening nest entrances. This pattern was not found between heterospecific colonies at greater distances from the camp midden, according with the prediction that interference competition is more probable as resources become more concentrated. Colonies of P. fallax near E. quadridens nests located near the camp midden had a net forage intake of 60% of those located in areas without E. quadridens. -Author

Interference competition and scavenging by Crematogaster ants (Hymenoptera: Formicidae) associated with the webs of the social spider Anelosimus eximius (Araneae: Theridiidae) in the central Amazon

Crematogaster cf. victima is a common inhabitant in the sheet web nests of the social spider Anelosimus eximius in the central Amazon basin near Manaus. A number of other ant species were found foraging on the non-sticky webs of A. eximius, but none of these reached the web occupation frequency found in C. cf. victima, nor, with the exception of an unidentified species of Pheidole, did they form satellite nests in the web, as did this species. Many prey which escaped the knock-down threads of the sheet web of A. eximius colonies were captured by ants in the lower web portions which they dominated. Furthermore, prey which were rejected by A. eximius, especially large, heavily sclerotized beetles, were also consumed by this ant. Repeated observations and experiments suggest that C. cf. victima is able to deter A. eximius activity through aerial venom release. Resources lost by A. eximius colonies to ants, especially C. cf. victima, in colonial web area and prey, may pose significant costs and may reduce colony growth.

Competition between spin, charge, and bond waves in a Peierls-Hubbard model

We study a one-dimensional extended Peierls-Hubbard model coupled to intracell and intercell phonons for a half-filled band. The calculations are made using the Hartree-Fock and adiabatic approximations for arbitrary temperature. In addition to static spin, charge, and bond density waves, we predict intermediate phases that lack inversion symmetry, and phase transitions that reduce symmetry on increasing temperature.

Larval aggregation and competition for food in experimental populations of Chrysomya putoria (Wied.) and Cochliomyia macellaria (F.) (Dipt., Calliphoridae)

In blowflies, larval aggregation in patches of food can be both intra- and interspecific, depending upon the degree to which competitors are clumped among the patches. In the present study, the implications of spatial aggregation for larval competition was investigated in experimental populations of the introduced blowfly Chrysomya putoria and the native Cochliomyia macellaria, using data from survival to adulthood in a range of single- and double-species larval cultures. The reduction in C. macellaria survival rate in the presence of C. putoria suggests that the former species is the inferior competitor. The results on survival to adulthood for both species in single- and double-species cultures can be explained in the light of the relationship between the level of intra- and interspecific aggregation and the efficiency of the larval feeding process. The possible implications of these results for the population biology of both species in natural environments are discussed.

Toxocariasis: Serological diagnosis by indirect antibody competition elisa

Toxocariasis is caused by infection of man by Toxocara canis and Toxocara, cati larvae, the common roundworm of dogs and cats. Because larvae are difficult to detect in tissues, diagnosis is mostly based on serology. Non specific reactions are observed mainly due to cross-reactivity with Ascaris sp antigens. This investigation aimed at developing and evaluating an indirect antibody competition ELISA (IACE) employing a specific rabbit IgG anti-Toxocara canis excretory-secretory antigens as the competition antibody. in order to improve indirect ELISA specificity performed for toxocariasis diagnosis. For that, the rabbit IgG was previously absorbed by Ascaris suum adult antigens. Sensitivity and specificity of IACE were first evaluated in 28 serum samples of mice experimentally infected with T. canis embryonated eggs. Adopting cut-off value established in this population before infection, sensitivity and specificity were 100% after 20 days post-inoculation. For human population IACE was evaluated using sera from 440 patients with clinical signs of toxocariasis and the cut-off value was established with 60 serum samples from apparently healthy individuals. Using as reference test the indirect ELISA performed by Adolfo Lutz Institute, sensitivity was 60.2%, specificity was 98% and concordance was 77.3%. Repeatability of IACE was evaluated by the inter-reactions variation coefficient (2.4%).

Competition between humic substances and α-amino acids by metal species

The aim of the present work was to carry out experimental comparison between humic substances (HS) and representative α-amino acids (methionine, methionine sulfoxide and cysteine hydrochloride) in relation to the complexation of biologically active trace elements (Al, Cu, Pb, Mn, Zn, Cd and Ni). A mobile time-controlled tangential-flow UF technique was applied to differentiate between HS-metal and α-aminoacids-metal complexes. Metal determinations were conventionally carried out using a ICP-OES. The results showed that HS may be considered as a selective complexing agents with higher metal bonding capability in relation to Al, Cu and Pb, the fact that may be clinically important.

Adapted Judo training on bone-variables in postmenopausal women in pharmacological treatment

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

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.

Indices of competition and bio-agroeconomic efficiency of lettuce and tomato intercrops in greenhouses

Intercropping is a cropping system for the production of greenhouse vegetables. It uses space more efficiently, thus reducing the cost of production. Intercropping tomato and lettuce has not been studied, but knowledge of the competitive and agroeconomic indices of these vegetables can help in the management of the intercropping system. The objectives of this study were to assess, through biological and agroeconomic indices, the competition between species and the profitability of intercropping tomato and lettuce at different times of transplantation over two growing seasons (autumn-winter and summer-winter) in greenhouse conditions. In autumn-winter, two experiments were conducted with a randomised complete-block design and five replicates. Tomato and lettuce were the main crops in the individual experiments. Treatments were arranged in a factorial of two cropping systems (intercropping and individual crops) with four transplants of the secondary crop (0, 10, 20 and 30 days after) plus an additional treatment (individual main crop). These two experiments were repeated in summer-winter. Tomato was the dominant crop regardless of transplant order. Intercropping systems established with transplants of both species on the same day had higher values of indices of competition and bio-agroeconomic efficiency than systems with longer periods of transplants between main and secondary crops. The intercropping of lettuce and tomato in greenhouses, regardless of transplant time or order, had bio-agroeconomic advantages over individual crops. The transplantation of tomato after lettuce is recommended for greater profitability.

Combining adapted Judo training and pharmacological treatment to improve bone mineral density on postmenopausal women: A two years study

Background & Study Aim: Physical activity has been an important factor to increase bone mineral density (BMD) and, consequently, to prevent and treat osteoporosis. The study aimed the effects of adapted Judo training on BMD in postmenopausal women, during pharmacological treatment. Material & Methods: Eighteen female volunteers participated in this study. They were separated into two groups: Adapted Judo training (AJT) (n=11; 52.2±5.3 years) and control group (CG) (n=7; 53.8±4.4 years). Lunar GE Dual Energy X-Ray Absorptiometry (DXA) measured BMD at lumbar L2-L4, femoral neck and trochanter sites. The training period for AJT was two years, comprised 12 mesocycles with different intensities. ANOVA compared 2 groups in 3 moments of testing and Scheffé Test allowed multiple comparisons between groups for the L2-L4 and femoral neck sites, but at trochanter was Fisher LSD. Results: ANOVA showed significant differences in the AJT group (F(2, 32)=15.187, p=0.000023). Scheffé Test showed significant increase on lumbar BMD after one year of AJT (Δ%=+8.9%, p=0.000017) and after two years this improvement stand still (p=0.33). The CG after one year presented significant decrease in BMD of femoral neck (Δ%=-6.9%, p=0.03) and trochanter (Δ%=-3.7%, p=0.0084). However, the CG recovered the loss of BMD of femoral neck (Δ%=+7.6%, p=0.02) and trochanter (Δ%=+3.8%, p=0.0079) after two years of study. Conclusions: Drug therapy, without the physical activity practice, can aid the maintenance of BMD. AJT may be considered as an efficient physical activity for postmenopausal women with low BMD in pharmacological treatment. © ARCHIVES OF BUDO | SCIENCE OF MARTIAL ARTS.

A iniciação competitiva precoce e a formação do técnico desportivo de judô: um estudo de caso dos resultados dos campeonatos paulistas de 1999 até 2011

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Nonlinear dynamics of within-host parasite competition

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Ecological competition and the incidence of Acinetobacter baumannii bloodstream infections in a teaching hospital in Southeastern Brazil

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