Out of the many trends emerging from recent technological developments, few have spread as quickly and exponentially as social media. As of October 2018, Facebook -the titan of this industry- totaled 2.2 billion users worldwide, with its little brother Instagram rising to 1 billion (Statista, 2018). Also worth mentioning are WhatsApp and its Chinese equivalent WeChat, with respectively 1.5 and 1.1 billion users. These platforms have changed the way we interact, with users creating and broadcasting content from their own lives. The amount of data produced on Facebook alone is tremendous. As example, every 60 seconds, 510,000 comments are posted, 293,000 statuses are updated, and 136,000 photos are uploaded (Zephoria.com, 2018).
But what if this data could be used productively? Advancements in the field of AI, and especially machine learning, led to the emergence of deep learning. Thanks to its processing power, it allows the analysis of much bigger data samples than before (Brynjolfsson, 2017). Applications are endless, but a particularly interesting sector is public safety.
In spite of monitoring, surveillance, and predictions, all countries are subject one way or another to catastrophes, natural -as the recent earthquake in Indonesia- or man-made -like terrorist attack in western Europe. It takes time for necessaries parties -the police, firemen, ambulances etc- to be informed of the situation, choose the necessary response, and arrive on site to provide help.
On the other hand, information about such events is spread on social media within minutes. What some companies are suggesting is that we make use of this global sensor network of eyewitnesses (Dataminr.com, 2018). The idea is to use a system, which analyzes and cross-checks real-time social media data. It could revolutionize the way we use emergency services, by detecting the earliest indications of high-impact events and critical information, and by alerting both civilians and support services of hazards.
Of course, this system comes with questions regarding its use. Which data can be used for analysis -especially with GDPR in place? If the use is for public safety, does that mean getting a social media account should be mandatory? How can you be certain of the veracity of the data used?
References:
Brynjolfsson, E., and Mcafee, A. 2017. The business of artificial intelligence: what it can and cannot do for your organization. Harvard Business Review
Most popular social networks worldwide as of October 2018, r. (2018). Global social media ranking 2018 | Statistic. [online] Statista. Available at: https://www.statista.com/statistics/272014/global-social-networks-ranked-by-number-of-users/ [Accessed 19 Oct. 2018].
Dataminr.com. (2018). Dataminr for Corporate Security | Dataminr. [online] Available at: https://www.dataminr.com/corporate-security [Accessed 19 Oct. 2018].
Zephoria.com. (2018). [online] Available at: https://zephoria.com/top-15-valuable-facebook-statistics/ [Accessed 19 Oct. 2018].
