Articles — Project / Lab Report

In most corruption scandals, the use of front companies for money laundering is almost ubiquitous. This work proposes to apply image classification to detect such organizations, through the use of Convolutional Neural Networks (CNN), namely the AlexNet architecture. The images are obtained by address search in Google Street View API, and the resulting classification will be further used along with other features to detect front com- panies in order to help the auditors from the Ministry of Transparency and Office of the Comptroller General (CGU, in Portuguese). To this moment, we applied classification to almost 15 thousand suppliers scenes with active contracts with the Brazilian Government until September 2016, obtained through data matching between the Government Purchases database and the Brazilian Federal Revenue Office database (more recent scenes should be added as this work progresses). Preliminary results with a pre-trained AlexNet CNN show the need for developing new scene classes more suited to the Brazilian context. In order to do this, we propose to apply clustering algorithms in features extracted from the last fully-connected layer of this net. The classes obtained will be used to fine-tune the AlexNet CNN for future classification, through the use of training from scratch or fine tuning techniques.

AME Fall '17 Senior Design

Relatório de estágio desenvolvido para a obtenção do título de Bacharel em Ciência da Computação no Instituto de Computação na Universidade Federal de Mato Grosso.

informe de investigacion

With the aim to develop more efficient aircraft configurations, the Blended-Wing-Body (BWB) unmanned aerial vehicles have grown attention in recent years. Compare to conventional aircraft configurations, the BWB structure has several advantages in aerody- namics and fuel efficiency. Topology optimization (TO) is also a relatively new structure optimization approach which has applied successfully in automotive industry for a consid- erable time. In this paper, topology optimization method will be applied on a special BWB structure UAV called BITU on both 2D and 3D models in ABAQUS. The optimization goal is to minimize compliance energy under specified loading and boundary conditions which will be computed in modeling and simulation section. Finally, optimized result compared to initial design will demonstrate TO is a rational and efficient design tool for structure optimization, especially in Aircraft industry.

Motivation: The Neurosky Mindwave device allows monitoring the electrical signals generated by the brains neural activities. The easy access of this devices opens a new area of researching fields not only in the gamification and disable people but also to understand the cognitive behavior of human beings. The purpose of this study is to investigate the consistency and effectiveness level of a non-invasive consumer product BCI. We investigate the output of the headset data both in quality and quantity of that gathered data and determine how it could be used for human involved research settings. A sample of two participants in terms of an interview and an interviewer interchanging questionnaires. The cognitive tasks and EEG output signals captured by the participants both attentive and meditation values. Results: The data we collected produces mixed results. The meditation or relaxing level provide a consisting output while the attention level need more to explore due to the nature of every individuals and also the surrounds such as noise, mode distraction levels etc. Keywords: BCI, Brain activity,EEG,NeuroSky, Attention, Meditation

Understanding dating behavior is intrinsically interesting and practically important for every individual in the society. People want to find someone who they want to share stories and emotions, understand and sympathize, commit the rest of life together. Accordingly, people spend a huge part of life finding "the one" or "soul-mate" who they believe potentially maximize their happiness and satisfaction in life. Ironically, they often end up breaking up and saying "He/She was not the right person". Researchers in many areas have studied dating behavior in varying ways to understand why people repeat the vicious circle and still get in there to find the right person. Here, we investigated dating behavior by analyzing relations between multi-aspect variables that include physical and psychological features of individuals and the probability of match in a speed-dating situation. We used theoretical approach and machine learning approach to investigate the pattern of dating behavior and to find the best predictor of match in dataing. For theory driven approach, we used multilevel linear model and multilevel logistic regression. For machine-learning approach, we used learning vector quantization and extreme gradient boosting.

The knowing of politics is mostly an unknown problem on a large scale in Colombia. Here is proposed a web visualization project, in which the historical information of the votes and the elected representatives are presented in an entertaining and inclusive way, in order to generate a feeling of empathy or politician relevance in the spectator creating the assumption that there's a familiar relationship

As of this writing, the algorithm employed for difficulty adjustment in the CryptoNote reference code is known by the Monero Research Lab to be flawed. We describe and illustrate the nature of the flaw and recommend a solution. By dishonestly reporting timestamps, attackers can gain disproportionate control over network difficulty. We verify this route of attack by auditing the CryptoNote reference difficulty adjustment code, which, we reimplement in the Python programming language. We use a stochastic model of blockchain growth to test the CryptoNote reference difficulty formula against the more traditional Bitcoin difficulty formula. This allows us to test our difficulty formula against various hash rate scenarios. This research bulletin has not undergone peer review, and reflects only the results of internal investigation.