Machine learning is the next frontier in big data innovation
Machine learning enables organisations to do more, learn more and ultimately get more value from their data, argues Postcode Anywhere CEO Guy Mucklow
Amazon recently shook up the big data space by launching Amazon Machine Learning, opening up its machine learning technology to all developers.
Amazon is a leading light in the big data sector, with big data technology underpinning the product recommendations that the site makes to customers. Such a big player committing to machine learning demonstrates that it really is the next frontier in big data innovation.
The announcement follows hot on the heels of IBM's launch of Watson Analytics and Microsoft's Azure Machine Learning, with a host of options now available to help developers bring machine learning intelligence into their apps.
This latest move by Amazon helps organisations to use all of the data that they are collecting and then make predictions at scale.
Within just a few years, the role of data within organisations has completely changed, and machine learning will mark another step change in how data is being managed and mined.
With organisations collecting more and more data, what they then do with it is becoming the pressing issue. This is where machine learning comes in, enabling organisations to do more, learn more and ultimately get more value from their data.
At present, competitive advantage is lost during the time that it takes for data to be mined and turned into actionable insights. Descriptive and predictive analytics operate by looking backwards at what has happened to make assumptions about the future.
What this latest move by Amazon indicates is that big data's potential will be harnessed through machine learning techniques and prescriptive analytics, rather than traditional data strategies.
Looking at one sector in particular, a big challenge faced by retailers is how to streamline the customer journey on their website and turn more visitors into customers. This challenge is what our big data tool, Triggar, will tackle by implementing interventions in real time, and across all sectors the emphasis is increasingly going to shift to real time rather than retrospective action.
Rather than looking into historical customer data and how customers have behaved in the past, the future of big data is increasingly going to be about actions being suggested and implemented in real time.
That's where machine learning comes in - because for an online retailer by the time the information about a customer's visit has been logged and analysed it is too late and a potential customer has drifted away.
Much has been written and said about the potential that big data has to take business to the next level, get much closer to their customers and operate more efficiently.
But big data is far from simple and has gained a reputation for being great in theory but difficult in practice, with Gartner finding that more than 85 per cent of Fortune 500 organisations will not be able to exploit big data effectively in 2015.
Amazon's new machine learning feature is designed to address this problem, using the power of the internet giant's own techniques and technology to comb through any company's data to build and deploy models.
Start-ups have struggled to benefit from big data, lacking expertise and the cash required, however services such as this new platform will help bridge the gap and allow small businesses to become much more data savvy from day one.
This move by Amazon, and similar plays by rivals like Microsoft, will help disrupt the status quo, where high-tech solutions have often been reserved for big companies.
Competition is the bedrock of economic progress and growth, and this move is reflective of a broader trend whereby sectors are constantly being disrupted.
Big data is no different, and the emphasis in the big data space will continue to shift towards machine learning and real-time analytics as more organisations come to realise that this is how an organisation's data can be harnessed to add the most value.
Moves by big players that open up access to technology and lower the barriers to entry will help innovative companies to take on the establishment and usher in a new era of big data, so watch this space.
Guy Mucklow is CEO of software company Postcode Anywhere and co-founder of big data start-up Triggar