The democratisation of analytics is apparent when you consider who is involved these days. SAS has been doing statistical analysis for 40 years (without the aid of big data in the early days) and had algorithms for machine learning in its portfolio long before machine learning was a buzzword. In the early years, we spoke with specialists (mathematicians, statisticians, and, now, data scientists). That has changed in the last five to 10 years. Lately, we have been working directly with the departments for which the data is relevant.
When algorithms compose music
Today we not only have vast quantities of data, we also have numerous possibilities for analysing it all. Algorithms are capable of learning to recognise people as people on images and even to recognise them as specific individuals. The formula behind it can be put simply as: data + algorithms + machine learning = artificial intelligence (AI).
People are often surprised at what machine learning is already able to do. AIVA (artificial intelligence virtual artist) generates works of classical music by way of algorithms. The app Replika copies the behavioural patterns of a friend. The CaliBurger robot cooks up hamburger patties, recognises when they are done, and even develops new recipes.
In the B2B segment, applications are found across industries and fields: customer segmentation, fraud detection, risk analysis, predicting employee attrition and cybersecurity are just a few examples. Hasn’t that sort of thing been around for a while? Yes, but the new methods deliver better results.
The new possibilities make for new uses: proactive health management instead of health insurance (or rather, sickness insurance), or automatic handling of insurance claims using image analysis. That all sounds pretty cool, but the question remains: Is there any money to be made in machine learning?
Companies have to integrate analytics into their business processes instead of just plugging the results of analyses into PowerPoint presentations. Despite many investments, both in innovative and legacy AI environments, most AI projects outside the high-tech industry are stuck in the prototype phase and seldom deliver the expected business value. The primary challenge is to implement AI applications, embedding them into enterprise business processes.
Furthermore, business lines struggle to process analytics insights in a timely manner, and the data and analytics infrastructure is often inefficient and costly.
To cope with the high business demand for valuable insights, the lack of time and resources from analytics practitioners, and the need for IT to get a platform it can trust and share, SAS offers an open and industrial platform to put analytics into action. The SAS Platform is best suited to deliver enterprise analytics across the organisation and support the complete life cycle, from discovery to build and deployment to production, in a diverse business context. This, and great customer stories (one from SciSports for the football lovers) is on the agenda of the SAS Platform Roadshow. Find out more and register now!
Is there any money to be made in #MachineLearning? #ParadiseFound #AI Click To Tweet