Just the other day while browsing the web, a pop up advert on breast pumps covered the bottom right hand corner of my screen. It was quite annoying at first, given that I’m not a lactating mother with a new born.
You do get a little frustrated when such product recommendations have little or no relevance to your life, especially if it’s something you wouldn’t buy in a million years. On second thought, I found the suggestive advert quite amusing, as it was clearly a misguided effort of a ‘not-so-smart-machine’.
Actually, these days such misplaced product suggestions are not so common. Normally when browsing the web or using social media like Facebook, or Netflix for that matter, you’d find these recommendations or pop up ads quite relevant. Something you’ve been interested in lately and have spent some time researching about it, or something that’s related to your past online experienceThat’s the power of ‘Machine Learning’, as it's enabling brands to learn from customers’ past experiences, and helps create relevant and effective content that reaches out to offer a more personalised, engaging and valued experience for the customers!
What then is AI or Machine Learning?
Though Artificial Intelligence (AI) and Machine Learning (ML) is often used interchangeably, and are very much related, they don’t mean the same.
A definition from Customerthink says:
Artificial Intelligence (AI) is an umbrella term that describes the part of computer science that deals with making computers emulate human intelligence.
Machine learning is the part of AI that makes a computer appear intelligent
The illustration below clarifies the relation between the terms AI, Machine Learning, Deep learning and Bots:
Stanford University describes Machine Learning as“The science of getting computers to act without being explicitly programmed.”
From effective web search, speech recognition, facial recognition to self-driving cars, Machine Learning (ML) seems to be seamlessly entering into our everyday lives. The core principle lies with the machines taking in data and ‘learning’ for themselves as they go. The more they are fed with the data set, the better they adapt, learn and improve their accuracy at a given task.
A study from Demandbase claims that 80% of marketing executives predict Artificial Intelligence will revolutionise marketing by 2020. However, only 10 percent are currently using it. Another report from Allied Market Research predicts the Global Computing Market (Machine Learning, in other words) is expected to garner $13.7 billion by 2020.
The future does seem to hold immense value for Artificial Intelligence (AI) and Machine Learning, but for now let’s look into 3 ways how they're already helping us redefine our customers’ experience:
Have you realised the page you see when you log on to Amazon.com is different than that of your friend who enters the same URL? That’s because Amazon is using machine learning to track, interpret and analyse your unique profile signature, based on your browsing and purchase history.
It tailors more personalised offers, deals, adverts and recommendations taking into consideration your buying behaviour and preferences, creating a memorable and valued customer experience for you.
The machines employ advanced algorithms in an autonomous manner to learn and adapt from customer’s experiences to mimic the thought process behind human decision making.
Any new data added to the existing model is rapidly processed and analysed to bring in new results, which in turn is used to delight and satisfy customers on a personal level by anticipating their needs and preferences.
As mentioned earlier, any product suggestions, adverts or special deals offered to you on sites like Amazon, Flipkart, Facebook, Instagram, etc., is uniquely targeted to your interests and is aimed at creating a satisfying personalised ‘buyer journey’ for you.
Personalisation through Machine Learning can also be found even in a simplified process as when you’re searching in Google. For instance if you’re searching for ‘Element 7 Digital’ but accidently make a typo ‘Element 7 Degree’, then Google will immediately give a message saying, “did you mean…digital?”. That’s the working of Google’s machine learning algorithms; which remembered you correcting the same typo in your earlier searches – and now it has learned to correct it for you.
For marketers and businesses, Machine Learning can arm your Marketing department with such a customer-centric predictive analysis that you can engage your customers with effective marketing campaigns, improve interactions and establish stronger relationships.
2. Automation – Self Driving Cars!
Another area where AI and Machine Learning is being rapidly implemented, is in the automation of works. Take for example the self-driving cars of Google, Uber or Tesla. If you get into one of these cars, you’ll be amazed how they operate autonomousl, changing lanes, stopping at traffic lights, monitoring cyclists and pedestrians, turn left or make an exit.
And this has been possible because of the AI and the Machine Learning capability of the car to adapt and learn to navigation guidelines, discern the routes and road networks and to make informed decisions by interacting to its surroundings.
The most common of the algorithm used in such autonomous vehicles are based on Object Tracking. These algorithms are targeted to accurately pinpoint and differentiate between two objects. It helps to inform the car whether it’s another vehicle, a pedestrian, an animal or a cyclist that’s in its proximity and thus it maneuvers accordingly.
To get it right it needs to undergo a sophisticated pattern recognition algorithms, where it's fed with many images. The algorithm inspects the images and guesses the objects, most early guesses will be wrong. The process continues with changed parameters, where the changes that decrease the algorithm’s accuracy are discarded and ones increasing the accuracy are retained, until the images are correctly identified.
Later, when new images are presented, the algorithm will classify them with high accuracy. That’s when it is declared to have ‘learned’.
Google Self driving cars are already operational in the Silicon Valley and other brands like Uber, and Tesla have also joined this mission of delivering a unique driving experience for its customers.
3. Fraud Detection
As businesses are increasingly adopting online paying systems, the risks for fraudulent transactions is also on the rise. With the volume of real time payments, banks are facing more pressure to find effective solutions for fraud detection, as they only have a window of mere milliseconds to figure out if a transaction is genuine or not.
However, big data analytics and Machine Learning systems have come to the rescue, and have been able to tackle this fraud detection challenge for the banks and financial institutions. The algorithms are capable of analysing a wide variety of metrics per transaction, such as trustworthiness of a vendor, to ascertain whether the time and location of the activity matches the individual’s earlier behaviour.
The analytics tools are designed to learn and constantly evolve by identifying the patterns of new information and comparing it to past data, to make decisions about the authenticity of the transactions.
Online brands like PayPal uses similar Machine Learning Technology to understand the users’ purchase history and finds patterns to help it devise new rules to deter scams and misappropriations. It's not as simple as studying customers buying behaviour and flagging any suspicious purchase disturbing a usual pattern. Other factors/indicators are also needed to build a complete picture or legitimacy of any transaction.
A tool like Stripe Radar is also helping businesses find and prevent online fraud. It runs A Machine Learning system at its core, that scans every card payment across its 100,000 plus businesses, after which it then synthesises the payment data into behavioural signals to identify and predict ones that resemble as fraud, thus blocking the payments having maximum indicators for fraudulent transactions.
To Sum It Up…
As customers are becoming more sophisticated, with an expectation for brands to provide more personal, secure and enjoyable experiences, Machine Learning and AI technology will become pivotal for businesses in finding that competitive edge!
Call Andy Fox (me) on (03) 5249 5570 or email firstname.lastname@example.org
Our Website is element7digital.com.au