Uncovering illegal wildlife trade on social media

Automatic data collection, deep learning filters, identification

Forskningsoutput: KonferensbidragSammanfattningForskning

Sammanfattning

A significant share of the illegal market for wildlife trade has moved to social media and the dark web, with recent contributions emphasising the role of social media both in cross-border trade and in local distribution on the destinations’ markets. So far, the use of social media data in conservation science has been limited, including manual, labour-intensive, efforts to classify information. Deep learning is a machine-learning approach used for diverse tasks, such as automated image recognition, natural language translation, and speech recognition. It has been applied within conservation science for identifying wild animals in camera-trap images. Here, we propose a framework using deep learning to filter and identify data pertaining illegal wildlife trade from social media platforms. We show how such an approach can be used to identify species and wildlife products – e.g. rhino horn and elephant ivory –, their origins, destinations, routes, and involved actors more efficiently from social media data.
The proposed presentation is a methodological contribution directed towards researchers and practitioners in the battle against illegal wildlife trade. It will discuss, in a hands-on manner, the necessary steps to i) collect data from social media, ii) train a neural network, and iii) detect content which is relevant in the context of illegal wildlife trade. The aim of the presentation is to give illegal wildlife trade enforcers access to the tools necessary to tap the tremendous amounts of data available on social media in their fight against illegal killing of endangered species, such as rhinoceros or elephants.
Originalspråkengelska
StatusPublicerad - 6 mar 2018
MoE-publikationstypEj behörig
EvenemangSavanna Science Network Meeting - Nombolo Mdluli Conference Centre, Skukuza, Kruger National Park, Sydafrika
Varaktighet: 4 mar 20189 mar 2018
Konferensnummer: 2018

Konferens

KonferensSavanna Science Network Meeting
Förkortad titelSSNM
LandSydafrika
OrtSkukuza, Kruger National Park
Period04/03/201809/03/2018

Vetenskapsgrenar

  • 1172 Miljövetenskap
  • 518 Medie- och kommunikationsvetenskap

Citera det här

Fink, C. A., & Di Minin, E. (2018). Uncovering illegal wildlife trade on social media: Automatic data collection, deep learning filters, identification. Abstract från Savanna Science Network Meeting, Skukuza, Kruger National Park, Sydafrika.
@conference{eb6a24ce6e524e7f84d222e8c4fb22d6,
title = "Uncovering illegal wildlife trade on social media: Automatic data collection, deep learning filters, identification",
abstract = "A significant share of the illegal market for wildlife trade has moved to social media and the dark web, with recent contributions emphasising the role of social media both in cross-border trade and in local distribution on the destinations’ markets. So far, the use of social media data in conservation science has been limited, including manual, labour-intensive, efforts to classify information. Deep learning is a machine-learning approach used for diverse tasks, such as automated image recognition, natural language translation, and speech recognition. It has been applied within conservation science for identifying wild animals in camera-trap images. Here, we propose a framework using deep learning to filter and identify data pertaining illegal wildlife trade from social media platforms. We show how such an approach can be used to identify species and wildlife products – e.g. rhino horn and elephant ivory –, their origins, destinations, routes, and involved actors more efficiently from social media data. The proposed presentation is a methodological contribution directed towards researchers and practitioners in the battle against illegal wildlife trade. It will discuss, in a hands-on manner, the necessary steps to i) collect data from social media, ii) train a neural network, and iii) detect content which is relevant in the context of illegal wildlife trade. The aim of the presentation is to give illegal wildlife trade enforcers access to the tools necessary to tap the tremendous amounts of data available on social media in their fight against illegal killing of endangered species, such as rhinoceros or elephants.",
keywords = "1172 Environmental sciences, 518 Media and communications",
author = "Fink, {Christoph Alexander} and {Di Minin}, Enrico",
year = "2018",
month = "3",
day = "6",
language = "English",
note = "Savanna Science Network Meeting, SSNM ; Conference date: 04-03-2018 Through 09-03-2018",

}

Fink, CA & Di Minin, E 2018, 'Uncovering illegal wildlife trade on social media: Automatic data collection, deep learning filters, identification' Savanna Science Network Meeting, Skukuza, Kruger National Park, Sydafrika, 04/03/2018 - 09/03/2018, .

Uncovering illegal wildlife trade on social media : Automatic data collection, deep learning filters, identification. / Fink, Christoph Alexander; Di Minin, Enrico.

2018. Abstract från Savanna Science Network Meeting, Skukuza, Kruger National Park, Sydafrika.

Forskningsoutput: KonferensbidragSammanfattningForskning

TY - CONF

T1 - Uncovering illegal wildlife trade on social media

T2 - Automatic data collection, deep learning filters, identification

AU - Fink, Christoph Alexander

AU - Di Minin, Enrico

PY - 2018/3/6

Y1 - 2018/3/6

N2 - A significant share of the illegal market for wildlife trade has moved to social media and the dark web, with recent contributions emphasising the role of social media both in cross-border trade and in local distribution on the destinations’ markets. So far, the use of social media data in conservation science has been limited, including manual, labour-intensive, efforts to classify information. Deep learning is a machine-learning approach used for diverse tasks, such as automated image recognition, natural language translation, and speech recognition. It has been applied within conservation science for identifying wild animals in camera-trap images. Here, we propose a framework using deep learning to filter and identify data pertaining illegal wildlife trade from social media platforms. We show how such an approach can be used to identify species and wildlife products – e.g. rhino horn and elephant ivory –, their origins, destinations, routes, and involved actors more efficiently from social media data. The proposed presentation is a methodological contribution directed towards researchers and practitioners in the battle against illegal wildlife trade. It will discuss, in a hands-on manner, the necessary steps to i) collect data from social media, ii) train a neural network, and iii) detect content which is relevant in the context of illegal wildlife trade. The aim of the presentation is to give illegal wildlife trade enforcers access to the tools necessary to tap the tremendous amounts of data available on social media in their fight against illegal killing of endangered species, such as rhinoceros or elephants.

AB - A significant share of the illegal market for wildlife trade has moved to social media and the dark web, with recent contributions emphasising the role of social media both in cross-border trade and in local distribution on the destinations’ markets. So far, the use of social media data in conservation science has been limited, including manual, labour-intensive, efforts to classify information. Deep learning is a machine-learning approach used for diverse tasks, such as automated image recognition, natural language translation, and speech recognition. It has been applied within conservation science for identifying wild animals in camera-trap images. Here, we propose a framework using deep learning to filter and identify data pertaining illegal wildlife trade from social media platforms. We show how such an approach can be used to identify species and wildlife products – e.g. rhino horn and elephant ivory –, their origins, destinations, routes, and involved actors more efficiently from social media data. The proposed presentation is a methodological contribution directed towards researchers and practitioners in the battle against illegal wildlife trade. It will discuss, in a hands-on manner, the necessary steps to i) collect data from social media, ii) train a neural network, and iii) detect content which is relevant in the context of illegal wildlife trade. The aim of the presentation is to give illegal wildlife trade enforcers access to the tools necessary to tap the tremendous amounts of data available on social media in their fight against illegal killing of endangered species, such as rhinoceros or elephants.

KW - 1172 Environmental sciences

KW - 518 Media and communications

M3 - Abstract

ER -

Fink CA, Di Minin E. Uncovering illegal wildlife trade on social media: Automatic data collection, deep learning filters, identification. 2018. Abstract från Savanna Science Network Meeting, Skukuza, Kruger National Park, Sydafrika.