Machine Learning - it's a NOW thing

Machine learning has often been cited as a thing of the future; think Black Mirror, sci-fi, and robots taking over the world! 

And just five years ago, the statistics would have backed this up - but time flies and five years on, where does machine learning stand in the order of things? Is it still a future thing or is NOW the time to act? 

My good friend and colleague Henry Brown (Head of Machine Learning at Appsbroker) has worked for some of the largest FTSE 100 organisations in their mission to adopt and scale ML. Naturally, he would state that machine learning is not only a NOW thing, but has been for a few years. Nevertheless, being so data-driven, I needed the statistics to back this up. 

Since COVID-19 hit, reports created as recently as February and March this year are no longer representative or relevant to the world we now live in. Most reports also focus on the global landscape, with the UK making up a tiny subset of the data. In response to this, Appsbroker has commissioned its very own report covering topic areas that we know our customers and prospects are interested in - and the results are fascinating!

But what is machine learning? I caught up with fellow #womenintech advocate and Principal Data Scientist, Rachel Lund to gain some key insights into machine learning.

Q. Rachel - In a nutshell, what is Machine Learning?

Machine learning is a specific type of artificial intelligence in which a computer calculates (‘learns’) relationships and patterns in a given set of data, which can be used to make predictions or decisions when shown new data. 

Machine learning can be extremely powerful where there is too much data for human analysts to process in any reasonable timescale. The explosion in data collection and the dramatic fall in the price of computing power over the last decade has driven rapid growth in interest and potential use cases of this technology.

Q. Are Machine Learning and AI the same thing? 

AI and machine learning are not the same thing, although you’ll often hear people use the terms as if they were interchangeable. AI refers to any application in which a computer can react to information in a way that matches or surpasses what human intelligence could do in a similar situation. Its reaction may be determined by extracting patterns from relevant data - and so be an application of machine learning - or it may be determined by a sophisticated set of pre-defined rules and so not be machine learning.

To give an example, you could sit a group of doctors down and ask them to write down which combinations of symptoms lead to which diagnoses. You could take that information and create a program which would diagnose a patient based on the symptoms observed. That program could be described as an application of artificial intelligence, but not of machine learning. 

However, if instead of asking doctors to write down the rules, you collected data on patient symptoms and their eventual diagnosis, and then allowed an algorithm to work out the relationship between the two and used those discovered relationships to diagnose new patients, that would be both an application of AI and of machine learning.

Machine learning enables businesses to make use of the vast amounts of data they are now collecting to: improve business operations, enhance customer experience, and ultimately drive profitability.

Machine learning can help an organisation understand its clients better, allow products and promotions to be shaped to the individual, and focus marketing on the most valuable customers and those in danger of leaving. It can also be used to reduce waste in the supply chain, mitigate risks such as fraud, customer default or equipment failure, and enhance the speed, efficiency and quality of delivery to the customer.

Q. What was the most interesting insight that came out of the report for you? 

The thing that surprised me most was the strength of commitment to invest in machine learning - despite the fact that businesses are facing a severe economic downturn in the wake of the COVID-19 pandemic. Just 11% of businesses have deprioritised investment in machine learning, while nearly 40% have escalated their plans. This only serves to highlight the importance to all businesses across all sectors in placing their trust in machine learning for future success.

Q. What top tip would you give to any business looking to invest in Machine Learning in the next 12 months? 

Pick off the low hanging fruit. It's easy to be dazzled by the excitement surrounding technologies associated with machine learning and by reports of cutting edge use cases and fall into a rabbit hole of complexity from which it's difficult to emerge. Like any new innovation introduced to a business, machine learning projects need to be of tangible business value. The easiest way to achieve this? Focus on using machine learning to tackle your organisation’s biggest challenges, keep it as simple as possible, and keep engaging with your commercial stakeholders.

At Appsbroker, we’ve commissioned a detailed report shedding light on the impact of machine learning in the UK private sector on business growth and profitability. Click the button below to download the full report and discover how you can benefit.

About Appsbroker

Appsbroker is the largest Google Cloud-only Managed Services Provider (MSP) in EMEA.

Our world-class services help multinationals and leading corporates across the globe modernise their applications, data and infrastructure at pace, from implementing applications across hybrid cloud through to deploying complex machine learning to migrating SAP to GCP. 

Our Agile Systems Integrator (ASI) approach means that we collaborate with customers to deliver rapid outcomes with real business value, marrying the best technology with the most modern methodologies.

Appsbroker is the Google Cloud Application Development Specialisation Partner of the Year and holds three Google Cloud Specialisations.

Written by Lisa White