It doesn’t matter what your fintech business does, it will probably either disrupt using machine learning or be disrupted by those that are. If you aren’t thinking about how machine learning fits into your business, it’s time to start. Here’s your beginner’s guide.
Machine learning is easy to understand and tricksy to master. The name tells you what it is; teaching computers to learn without being specifically programmed. For example, developing an algorithm to make predictions from a set of data.
In my first fintech startup, we used machine learning algorithms to build a categorisation engine for transactions, enabling our software to automatically generate an accurate budget from transactional data feeds provided by a third-party. This gave us a massive benefit over other budgeting software that required manual input and meant that our users could get exponentially better spending plans, formulated in the cloud while they did something more intriguing.
Machine learning is where exponential improvement comes from. It’s why level-four self-driving vehicles already do a better job of getting from A to B than any human driver, or why Google Maps is a 10x better way to navigate traffic than a conventional GPS unit.
Machine learning is complex and can be daunting for engineering teams, but the good news is that you don’t need to develop it all yourself. Amazon, Google, Microsoft and startups like Contactable are just some of the machine learning developers that make their tech available via APIs and libraries. These are some of the best places to start incorporating machine learning into your business:
Originally developed by the Google Brain team as part of Google's Machine Intelligence research, TensorFlow was designed for the purposes of conducting deep neural networks research. The system is general enough to be applicable in a wide variety of domains and has been spun out as an independent organisation. TensorFlow is currently used by SnapChat, Twitter and numerous others.
Azure Machine Learning
Microsoft has aimed to make machine learning simple and scalable on its Azure cloud-platform. Besides a solid set of APIs for developers, there’s also a drag-and-drop tool for plotting out analytic flows and put together a specific deployment. It’s also the tech behind Microsoft’s Cortana virtual assistant.
Amazon Machine Learning
It’s highly likely that your business already runs its technology stack on Amazon Web Services. The platform also offers Amazon’s Predictive Analytics technology as a service. This includes visualization tools and wizards that guide you through the process of creating machine learning models without having to learn complex technologies.
H2O is an open source machine learning platform that unlocks analytics on datasets whether they’re hosted on a specialist cloud platform or even just Excel. While still a startup, H2O has taken off with thousands of users from small startups to big organisations.
One of the more interesting machine learning startups, Darktrace bases its security technology on the principles of the human immune system, detecting previously unidentified cyber threats, irrespective of their origin. That might sound like SiFi, but Darktrace is already being used by Toyota, Metro Bank and many other large corporations where security is paramount.