Do we need to mine social media data to detect exotic vertebrate-pest introductions?

Abstract

Invasive alien species are responsible for considerable biodiversity loss and environmental damage. Timely detection of new incursions is critical in preventing novel populations establishing. Citizen reports currently account for the majority of alien species detections, arising from the massive observation effort that the physical and digital ‘eyes and ears’ of citizens provide, in combination with crowd-sourced species identification. Because the reporting of alien species sightings is generally not mandatory, there is interest in whether mining social media data via image recognition and/or natural language processing can improve on existing passive citizen surveillance in a cost-effective manner. Here, we illustrate, using examples from Australia, how citizen surveillance for most vertebrate groups appears to currently be effective using existing voluntary reporting mechanisms. Where citizen surveillance is currently ineffective, for reasons of inadequate sampling, data mining of social media feeds will be similarly affected. We argue that mining citizens’ social media data for evidence of invasive alien species needs to demonstrate not only that it will be an improvement on the business as usual case, but also that any gains achieved cannot be achieved by alternative approaches. We highlight the potential role of education in increasing the surveillance effectiveness of citizens for detecting and reporting sightings of alien species. Should data mining of social media platforms be pursued, we note that the scale of the task in terms of the potential number of exotic vertebrate species to be classified is very large. The expected number of false positive classifications would present a considerable workload to process, possibly undermining the efficiency rationale for the use of data mining. Hence, prioritisation is needed, and we illustrate how the number of species to be classified can be reduced considerably. If we are to deploy data mining and analysis of social media data to help with detecting introductions of invasive alien species, we need to conduct it in a manner where it adds value and is trusted.