It used to take months to be able to say whether a particular treatment for cancer was working – wasting precious time which might otherwise have been used to save a patient’s life. Now using analytics, we can predict that treatment’s effectiveness within days.
When addressing the question of what to expect in the tech space in 2018, there is no limit. AI is already doing things we never before would have dreamed possible. From writing music to creating videos, we are achieving milestones which we previously would have considered strictly human.
And yes, it is even helping to save lives.
One of the major forces driving the world of tech and AI is the increasing volume and availability of data. Think of devices like the Fitbit, which provides a wealth of data concerning your health, such as heart rate and sleeping patterns.
At the same time, we’ve developed technology that allows us to analyse more data than ever before. And thanks to a massive improvement in compute power, analytical solutions can now analyse these massive volumes of data at blistering speed. Data scientists can develop machine learning models in minutes, which can enable businesses to deliver results quickly.
A great example of the technology that allows this is SAS VIYA, which is an end-to-end analytical platform. The platform fuels the analytics life cycle from data preparation to model development and finally deployment. This is all done in a single interface
One feature of the SAS platform that I’m particularly excited about is the ability to analyse images. This capability is already helping when it comes to wildlife conservation. In the past game rangers had to manually take pictures of particular species of animals and tag them. While this wasn’t part of their core focus, it absorbed a great deal of time. But using SAS’s new technology they can simply take the picture and allow the AI to classify, not only the species of the animal, but other helpful traits such as the sex as well. At the end of the day this frees up the rangers to tackle more important tasks.
More accurate predictions
While the algorithms used in machine learning have been relatively unchanged for decades, we are now seeing the emergence of new algorithms, such as extreme gradient boosting, which have proven to be very successful in data mining competitions like Kaggle. Extreme gradient boosting is a significant development in analytics because it generalises well, enabling more accurate predictions.
While we’ve been drawing on structured data sources like transactional data for some time, no-one has really been tapping into unstructured data sources. For example, customer complaints, reviews and other text data sources.
But these two sources when combined together can be extremely powerful. Say, for instance, you wanted to develop a customer churn prediction model. By including data sources like customer complaints, as opposed to just structured and traditional data sources, you can develop a model that is more accurate at predicting churn.
Deep learning has created a lot of hype, and for good reason.
It is a type of machine learning, based on a set of algorithms that model high-level abstractions in data, by using multiple processing layers with complex structures.Instead of organising data to run through predefined equations, deep learning sets up basic parameters about the data and trains the computer to learn on its own by recognising patterns using many layers of processing. This means it can train computers to perform human-like tasks, such as recognising speech, identifying images or making predictions.
Deep learning is already being used to make significant inroads into areas such as image recognition, fraud detection and the highly regulated credit risk modelling. In fact, SAS is currently working with credit bureau, Equifax, using deep learning techniques in credit risk modelling. The results are promising as the accuracy of the models has improved traditional techniques.
Bots that understand emotion
Another exciting space in AI is bot technology. Chatbots are programmes that use natural language processing and AI to create conversations between machines and humans.
Instead of having a human respond to complaints or queries, this can now be done by a chatbot to save time and money on mundane and repetitive tasks. For example, responses to queries on bank accounts. Some banks are using bots to advise customers on financial advice and investments.
Until now, AI has generally been designed to do specific things like fraud detection. The human ability to perform tasks has always been greater than machines as we can generalise and perform a much wider set of functions.
But incredibly we’re starting to see AI train itself to learn.
In 2016 Google created a programme called AlphaGo. It was capable of beating even the most skilled human players at the ancient Chinese strategy game, Go – considered to be one of the most complicated games on earth.
But this was taken a step further through the creation of AlphaGo Zero, a programme provided with a very limited amount of training data. The idea was that it would learn by playing against itself. Over a period of time, AlphaGo Zero beat AlphaGo.
Essentially it had taught itself to think.
On the threshold of a future in which machines can think and learn: as we step into 2018, one could say nothing is impossible.