Analyzing Social Network Images with Deep Learning Models to Fight Zika Virus

Barros, PH; Lima, BGC; Crispim, FC; Vieira, T; Missier, P; Fonseca, B

doi:10.1007/978-3-319-93000-8_69

Analyzing Social Network Images with Deep Learning Models to Fight Zika Virus

Lookup NU author(s): Professor Paolo Missier ORCiD

Downloads

Accepted version [.pdf]

Licence

This is the authors' accepted manuscript of a conference proceedings (inc. abstract) that has been published in its final definitive form by Springer Verlag, 2018.

For re-use rights please refer to the publisher's terms and conditions.

Abstract

© 2018, Springer International Publishing AG, part of Springer Nature. Zika and Dengue are viral diseases transmitted by infected mosquitoes (Aedes aegypti) found in warm, humid environments. Mining data from social networks helps to find locations with highest density of reported cases. Differently from approaches that process text from social networks, we present a new strategy that analyzes Instagram images. We use two customized Deep Neural Networks. The first detects objects commonly used for mosquito reproduction with 85% precision. The second differentiates mosquitoes as Culex or Aedes aegypti with 82.5% accuracy. Results indicate that both networks can improve the effectiveness of current social network mining strategies such as the VazaZika project.