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.
The IST-PORTFOLIO Report TEMPLATE, version v2.1-2017, is used in to produce the Activities Report and the Learnings Report of INSTITUTO SUPERIOR TECNICO, UNIVERSIDADE DE LISBOA, Independent Studies Courses.
The Template is "smart", allowing to select the language of writing (English or Portuguese), as well as the "type" of Report to be written. Please view instructions in the README.TXT file.
Intended to help me creating tons os articles out of Markdown files, including the article meta-information (such as Title and Author) in a semi-transparent white box in a small cover background.
This version (1.2) has the "mobile" option (just uncomment it in main.tex to see it) to have an output more suitable for smartphones.
In "mobile" way the image will cover the entire page, toc will also get a full page and sections (and subsections and subsubsections), except if it's the first children, will also clear the page. I did this based on a stackexchange answer but I forgot to copy the URL to reference it. And URLs will be inline.
(I'd like to thank Lian Tze Lim! I'm glad I found her blog post about markdown usage on overleaft)
Sorry for the Portuguese instructions and the poor tex structure.