Cláudio A. Rocha

Strategies for improving the modeling and interpretability of Bayesian networks

One of the main factors for the knowledge discovery success is related to the comprehensibility of the patterns discovered by applying data mining techniques. Amongst which we can point out the Bayesian networks as one of the most prominent when considering the easiness of knowledge interpretation achieved. Bayesian networks, however, present limitations and disadvantages regarding their use and applicability. This paper presents an extension for the improvement of Bayesian networks, treating aspects such as performance, as well as interpretability and use of their results; incorporating genetic algorithms in the model, multivariate regression for structure learning and temporal aspects using Markov chains.

Decision support in power systems based on load forecasting models and influence analysis of climatic and socio-economic factors

This paper presents a decision support system for power load forecast and the learning of influence patterns of the socio-economic and climatic factors on the power consumption based on mathematical and computational intelligenge methods, with the purpose of defining the future power consumption of a given region, as well as to provide a mean for the analysis of correlations between the power consumption and these factors. Here we use a linear modelo of regression for the forecasting, also presenting a comparative analysis with neural networks, to prove its efectiveness; and also Bayesian networks for the learning of causal relationships from the data.

Epidemiology and Genetic Characterization of Noroviruses among Adults in an Endemic Setting, Peruvian Amazon Basin, 2004-2011.

Background Successful vaccination strategies against norovirus will require understanding the burden of disease and relevant genotypes in populations. However, few data are available from cohort studies of adults living in low- and middle-income countries (LMIC). Materials and Methods We conducted a nested case-control study within a Peruvian military cohort to characterize the burden of norovirus infection, predominant genotypes, and associated symptoms from 2004 through 2011. Randomly selected case and control stools were tested for norovirus, bacteria, and parasites. The odds ratio of the association between norovirus infection and diarrhea was estimated using multiple logistic regression and co-infection adjusted attributable fractions were calculated. Results Of the 3,818 cohort study participants, 624 developed diarrhea. Overall and norovirus-associated diarrhea incidence rates were 42.3 and 6.0 per 100 person-years, respectively. The most prevalent norovirus genogroup was GII (72.5%, 29/40), which was associated with diarrhea (AOR 3.4, 95% CI: 1.3–8.7, P = 0.012). The co-infection adjusted GII attributable fraction was 6.4%. Discussion Norovirus was a frequent cause of diarrhea in an adult population followed longitudinally in an LMIC setting. Vaccine strategies should consider targeting adults in endemic settings and special populations that could serve as community transmission sources.

Social Networks Analysis and Participation in Learning Environments to Digital Inclusion Based on Large-Scale Distance Education

Evaluating and monitoring large-scale distance learning programs require different techniques, systems, and analysis methods. This work presents challenges in evaluating and monitoring digital inclusion training programs, considering the aspects inherent in large-scale distance training, and reports an approach based on network and distance learning. The paper has the following objectives: (i) apply algorithms to extract indicators from interaction networks, in a real scenario and consolidated training based on distance learning; (ii) apply algorithms to correlate interaction indicators with other indicators related to the use and participation in learning environments; and (iii) discuss the relevance of the obtained indicators to promote feedback with information critical to the success of a large-scale distance training program.

Social Network Analysis and Mining to Monitor and Identify Problems with Large-Scale Information and Communication Technology Interventions

The published literature reveals several arguments concerning the strategic importance of information and communication technology (ICT) interventions for developing countries where the digital divide is a challenge. Large-scale ICT interventions can be an option for countries whose regions, both urban and rural, present a high number of digitally excluded people. Our goal was to monitor and identify problems in interventions aimed at certification for a large number of participants in different geographical regions. Our case study is the training at the Telecentros.BR, a program created in Brazil to install telecenters and certify individuals to use ICT resources. We propose an approach that applies social network analysis and mining techniques to data collected from Telecentros.BR dataset and from the socioeconomics and telecommunications infrastructure indicators of the participants’ municipalities. We found that (i) the analysis of interactions in different time periods reflects the objectives of each phase of training, highlighting the increased density in the phase in which participants develop and disseminate their projects; (ii) analysis according to the roles of participants (i.e., tutors or community members) reveals that the interactions were influenced by the center (or region) to which the participant belongs (that is, a community contained mainly members of the same region and always with the presence of tutors, contradicting expectations of the training project, which aimed for intense collaboration of the participants, regardless of the geographic region); (iii) the social network of participants influences the success of the training: that is, given evidence that the degree of the community member is in the highest range, the probability of this individual concluding the training is 0.689; (iv) the North region presented the lowest probability of participant certification, whereas the Northeast, which served municipalities with similar characteristics, presented high probability of certification, associated with the highest degree in social networking platform.

Employing online social networks to monitor and evaluate training of digital inclusion agents

This work presents challenges in evaluating and monitoring digital inclusion training programs, considering the aspects inherent in large-scale training, and report the main challenges in an approach based on network learning. For this, our goals are as follows: develop an architecture to provide all interface features with communication tools, data collection, automatic notification (alerts) of interest to those involved in training and proposal, and survey analyses; employ algorithms to measure the centrality, prestige, and density of interactions in a real case and consolidated training based on learning network; employ algorithms to correlate and measure, in probabilistic terms, the effects of participation in the interaction tools and the use of the resources and activities proposed; and improve a large-scale training program from the implementation of monitoring and tracking services offered in the architecture.

Comparative studies of Statistical and Neural Networks Models for Short and Long Term Load Forecasting: a Case Study in the Brazilian Amazon Power Suppliers

One of the most desired aspects for power suppliers is the acquisition/sell of energy in a future time. This paper presents a study for power supply forecasting of the residential class, based on time series methods and neural networks, considering short and long term forecast, both of great importance for power suppliers in order to define the future power consumption of a given region.

A Novel Probabilistic Approach for Analysis and Planning of Large Capillarity Broadband Networks Based on ADSL2+ Technology

The increasingly spread of information through digital media raises new realities in the worlds present scenario and, thus, new technologies have been emerging in order to streamline the process of disseminating information and providing quality access to such information by the population. The Next Generation Network (NGN) holds tremendous potential, with a promise to merge the transmission of data, voice, video and other media into a single network; unfortunately, several developing countries do not have the necessary infra-structure to implement NGN technology. The main concern in these networks is not the backbone or the transport layer, but in the last mile itself. Last mile has become a popular keyword to indicate the technology which connects the End User to the Network backbone.

Comparative Analyses of Computational Intelligence Models for Load Forecasting: A Case Study in the Brazilian Amazon Power Suppliers

One of the most desired aspects for power suppliers is the acquisition/sale of energy for a future demand. However, power consumption forecast is characterized not only by the variable of the power system itself, but also related to socio-economic and climatic factors. Hence, it is imperative for the power suppliers to design and correlate these parameters. This paper presents a study of power load forecast for power suppliers, comparing application of techniques of wavelets, time series analysis methods and neural networks, considering long term forecasts; thus defining the future power consumption of a given region. The results obtained proved to be much more effective when compared to those projected by the power suppliers based on specialist information, thus contributing to the decision making for acquisition/sale of energy at a future demand.

Strategies for improving the interpretability of Bayesian networks using Markovian time models and genetic algorithms

One of the main factors for the success of the knowledge discovery process is related to the comprehensibility of the patterns discovered by the data mining techniques used. Among the many data mining techniques found in the literature, we can point the Bayesian networks as one of most prominent when considering the easiness of knowledge interpretation achieved in a domain with uncertainty. However, the static Bayesian networks present two basic disadvantages: the incapacity to correlate the variables, considering its behavior throughout the time; and the difficulty of establishing the optimum combination of states for the variables, which would generate and/or achieve a given requirement. This paper presents an extension for the improvement of Bayesian networks, treating the mentioned problems by incorporating a temporal model, using Markov chains, and for intermediary of the combination of genetic algorithms with the networks obtained from the data.