Businesses are often data-rich but information-poor. Machine learning is changing that. This application of artificial intelligence (AI) enables computers to learn independently by using algorithms to intepret data. It can help organisations process vast quantities of complex information and analyse to improve business insights, predictive accuracy and decision-making.
Machine learning is already being used in applications from fraud detection to self-driving cars, and sectors from marketing to government. CIO UK looks at how eight leading CIOs are using machine learning in their businesses.
"In a large [legal case] such as Rolls-Royce, which resulted in a £671m settlement, we had 70 investigators working to review over 30 million documents. It's just not possible to manually review that amount of data, so we worked with our technology partners to develop an AI robot to assist with that.
"We were able to prove that this approach is both more accurate and much more efficient than human review alone – in some instances at one-fifth of the cost. The volume of data in our cases is increasing exponentially, and it won’t be long before we have one that is two or three times the size of Rolls. Because of that, using machine learning techniques will be an essential and critical part of every investigation in the future so we can move faster and deliver better value for money."
Ben Denison, Serious Fraud Office CTO
"We are experimenting with machine learning technologies to help with product identification and managing workflows that would have been done by humans in the past."
Neil Pearce, Travis Perkins CIO
"[I'm] in the process of looking at our business model and main business operations to benefit from automated decision making and machine learning. We have managed to use some algorithms to help us match the next five years supply and demand.
"This would have been too complex to be done manually to the detail we can provide and automating the process allows the 'recalculation of scenario planning possible which manually the time/complexity required to do this would be prohibitive. Ultimately this allows us to be more proactive on planning. Our phase two of this is improving data quality to use more analytics for better decision making and price optimisation."
Matthew Butlin, BerryWorld CIO
"We could use machine learning where we currently have manual intervention in aspects of our workflows, and we could even get to the stage where we use a lot of machine learning in our underwriting algorithms," says Capital One Europe CIO Rob Harding. The company has already implemented some aspects of machine learning into its data science team, but Harding believes there is much more to be explored in the area."
Rob Harding, Capital One Europe CIO
"JLL has partnered with Leverton, a Berlin-based 'proptech' start-up company, as we are seeking to increase automation in our lease management operations to ensure improved efficiency and swifter delivery of outputs for our clients. Leverton uses machine learning and deep learning technology to identify, extract and manage key terms and data from corporate documents in multiple languages. Machine learning and deep learning technology searches for key terms such as rental values, dates and figures to radically optimise the way that lease documents are reviewed and analysed
"Our work with Leverton on machine learning technology implementation across our lease administration business is transforming the way we do things. Lease contracts can comprise between three and 15 documents, hence, digitising the process significantly reduces the time spent reading and reviewing each document. It also allows the lease administrator to spend time applying subject matter expertise in recognising patterns, anomalies and opportunities."
Chris Zissis, JLL EMEA CIO
"A number of initiatives in machine learning show great potential to allow our scientists to focus on the science and reduce time to produce predictive models from weeks to hours."
David Smoley, AstraZeneca UK CIO
"Today, every plant around the world brings forecasting, inventory, and production information together in their ERP to help managers meet a demand cycle. With machine learning and AI, we will be connecting the machines in the plant to that ERP, and our ability to determine how, when and where to produce parts to meet a demand forecast will improve dramatically."
Jim Fowler, GE CIO
"We've deployed some AI/ML capability within our sentiment dashboard application, which uses machine learning services in the cloud combined with in-house data to build a picture for the licensee. The BI/analytics initiative we've started centres on SQL Server 2016, which brings high-end analytics and machine learning capability through Cortana and PowerBI."
Mike McMinn, Marston's IT director
"We're running a pilot with Microsoft's backing, plugging building management systems into their machine learning product, to see if we can predict maintenance cycles and energy use. I see that as a really interesting area of development, and it also ties into our IoT work."
Chris Weston, former Bellrock CTO
"At Ovo we have 20% of our base on smart meters already, transmitting consumption data to us up to every 10 seconds. We have developed a smart IoT device that connects to the meter via ZigBee and gives us data in real time. We store everything in NoSQL and time series databases and analyse it using machine learning."
Mariano Albera, former Ovo Energy CTO