Articles — Universidade de Brasília (UnB)

The usage of software has grown as computers become popular. There have emerged, both in academia and in the market, technological solutions for several areas, among them education. On the other hand, classroom teaching and learning continues to suffer from classical educational problems such as lack of student and teacher motivation and lack of clear educational goals. And although software supports learning across a range of disciplines and ages, children's audiences, especially in mathematics, have been little contemplated with the benefits that technological solutions can bring. Therefore, the use of pedagogical approaches, such as Bloom's Taxonomy and Formative Assessments, together with gamification techniques, such as Octalysis, can be used to develop a technological solution that contemplates this public. The present work aims to propose the development of a software to assist the teaching and learning of mathematics for children in the classroom.

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.