In the fight against poachers and illegal trade, animals can use all the help they can get. Thanks to researchers at the University of Helsinki’s Digital Geography Lab, wildlife may find that aid through a popular tool traffickers use to deal their illegal wares — social media.
“With an estimated two and a half billion users, easy access has turned social media into an important venue for illegal wildlife trade,” Enrico Di Minin, a conservation scientist working on the project, told Digital Trends. “Wildlife dealers active on social media release photos and information about wildlife products to attract and interact with potential customers, while also informing their existing network of contacts about available products. Currently, the lack of tools for efficient monitoring of high volume social media data limits the capability of law enforcement agencies to curb illegal wildlife trade. We plan to develop and use methods from artificial intelligence to efficiently monitor illegal wildlife trade on social media.”
Di Minin and his colleagues are designing a system that they hope will be able to comb through social media posts to identify images, metadata, and phrases associated with illegal wildlife trade, be it products or animals themselves. The task is too big for humans to do alone so they are enlisting software, including image recognition and natural language processing (NLP) algorithms, to filter through all the noise and spot suspicious activity.
“Illegal wildlife trade is booming online, in particular on social media,” Di Minin said. “However, big data derived from social media requires filtering out information irrelevant to illegal wildlife trade. Without automating the process with methods from artificial intelligence, filtering high-volume content for relevant information demands excessive time and resources. As time is running out for many targeted species, algorithms from artificial intelligence provide an innovative way to efficiently monitor the illegal wildlife trade on social media.”
As an example, Di Minin said image-recognition software can be trained to detect specific products, such as a rhinoceros horn or an elephant tusk, while metadata can give clues to an image’s location. Audio-video cues, such a particular bird call, can signal illegal pet trade while NLP algorithms can differentiate between a post that features an animal for sale or in the wild. “Potentially, such algorithms can also identify code words that illegal smugglers use in place of the real names … by processing verbal, visual and audio-visual content simultaneously,” Di Minin said.
A.I. may help spot illegal trade but it can’t alone stop it. For that Di Minin encourages social media platforms to actively crack down on sellers. He admits there is a risk that these dealers will move to other platforms but stressed that partnerships between companies, scientists, and law enforcement are key.
