Privacy must be at forefront of data mining

The fallout from the Cambridge Analytica and Facebook data scandal caused shock waves in the tech financial markets and raised fundamental issues about personal data protection and privacy. The optimism that surrounded social media as a tool to empower citizens has in recent months been quashed by insidious actors, who view the platforms as fertile ground for surveillance and political manipulation.

But before we #DeleteFacebook, let us not overlook the significant role that social media has played in championing free speech and democratising public opinion. Thanks to the proliferation of social media, we have seen the rise of important movements like #MeToo and mass mobilisation of millions in the pro-democracy Arab Spring.

I have also seen first-hand the pitfalls of social media. BrandsEye recently provided data about gender biased tweets to a report by social enterprise group, Atalanta. The report analysed how gender impacted online discourse of female politicians. The findings highlighted the immediacy and anonymity that social media affords abusive trolls with little recourse. Commenting on the findings, Atalanta’s founder, Eva Barboni importantly stressed that “while it may often feel like an uphill battle in the face of harassment and threat, it’s one worth fighting for the sake of our democratic systems and the benefit of the next generation of female political leaders.” And while I share Barboni’s sentiment that we ought to persevere on social platforms, not everyone is as optimistic.

The work of unethical companies, like Cambridge Analytica in particular, has given data mining tools a bad name. But mining of public – and not private – data can be a force that empowers social media users.

Ethical data mining firms do not target or profile users to manipulate them based on their preferences. Instead, they function as polling companies of spontaneous and volunteered opinions. These opinions are carefully and accurately analysed for both sentiment and the underlying issues driving sentiment. This ‘intention data’ from users who chose to share their views publicly, on platforms like Twitter, are then provided to clients who range from subscription services like telco providers and banks to governments agencies and city councils. The objective is to use this data to provide better quality services or products. The social media user is the winner.

At BrandsEye, we mine public opinion data to empower social media users by bringing them closer to decision makers who can effect change with an accurate understanding of how their citizens or customers feel. We are able to amplify the public’s views at scale, and cut out irrelevant data and trolls. We can ensure that those members of the public with something to say, and wish to be listened to, get heard by the people that count.

These benefits were recently highlighted when a major global ride-hailing app introduced a significant change following an opinion mining exercise that looked at what issues were driving sentiment of their customers on social media. The company found that a significant number of their customers, in two developing countries, had raised concerns around rider safety and called for the introduction of an emergency button. This led to the introduction of a new emergency safety feature.

But we cannot turn a blind eye to what we have witnessed with the Cambridge Analytica scandal.

Big data plays a central role in almost every successful business today. And it will continue to play an increasingly important role in improving services from healthcare to e-commerce. For the social platforms, like Facebook, our personal data is central to their business models.

What we see on our feeds – news articles we read and items we purchase online – is informed by our personal data. But where do we draw the line about the use of this data?

We need to continue to the discussion about how to reasonably regulate the way these platforms and their apps use this data so that we can protect individual privacy. In his 2017 book, World Without Mind: The Existential Threat of Big Tech, Franklin Foer offers a pragmatic regulatory solution, a “Data Protection Authority” that “protects privacy in the same way that government protects the environment”. Foer suggests that we be realistic about regulation and not attempt to prevent all collection or even exploitation of data but rather determine a set of constraints about what data can be collected and what can be exploited.

Platforms like Facebook, Instagram and Twitter and the many that will emerge over the ensuing years, must be protected and regulated to allow users to share their opinions. Mining public data from social media has the power to facilitate direct democracy and connect the needs of the people with decision makers inside government and business. It has the power to improve services and even save lives. But, in the wrong hands and without proper restrictions, the same algorithms that mine public data can be used to access and exploit private data, as Cambridge Analytica did in 2016. It is therefore critical that we introduce new regulations and restrictions to ensure that similar abuses of private data do not occur again.