Subscribe / Unsubscribe Enewsletters | Login | Register

Pencil Banner

Search, compare, and if you find a better decision, go for it!

Ankit Khandelwal, Senior Analytics Consultant, FICO Asia Pacific | Dec. 2, 2015
Ankit Khandelwal of FICO talks about the three major artificial intelligence technologies that could provide value to businesses.

This vendor-written piece has been edited by Executive Networks Media to eliminate product promotion, but readers should note it will likely favour the submitter's approach.

One of Disney's hit film of 2014, "Big Hero 6" showed us a friendly easy-going robot who could identify a person's health and emotions and act accordingly until the patient was fully satisfied with his care. From a technological point of view, this robot was no more than a gigantic combination of databases —Big Data— with a capability for rational thought that mirrored the human mind (artificial intelligence).

We simply have to watch the news to find information related to these two technologies every day. In fact, in July this year, SoftBank, the Japanese internet and telecommunications giant, unveiled Pepper — a humanoid robot that they claim was the first to respond to and exhibit human emotions. More importantly, the robot will be available for businesses in Japan where one of its most attractive functions will be big data collection and analytics. Pepper will be able to count the number of customers entering a shop, estimate their age, gender and emotional state. From there, it is able to analyse how receptive a customer is to a particular product.

We are certainly making great strides in building robots that combine Big Data and artificial intelligence which will help us with our everyday business operations. 

Ongoing innovation in technology that mirrors brain function is helping improve analytics' ability to recognise irregular patterns. Applied to the business and finance universe, we are talking about the chance of anticipating fraudulent transactions and identifying new customers who could provide a greater return in a long-term relationship.

We are now training artificial intelligence predictive models to learn dynamic environments and self-adjust to certain behavioural patterns. In the future, lots of models will experiment and learn unaided and will be able to adapt spontaneously to new scenarios depending on the real-time information they receive. This is the key to turn a huge sum of data into real business value, in other words, to turn Big Data into Big Value.

We are talking about three major artificial intelligence technologies that could provide value to our businesses: neural networks, self-learning based analytics techniques, and neuro-dynamic programming.

Firstly, neural networks are analytics techniques that learn to recognise complex behavioural patterns. This technique emulates the way brain neurons store information connected to other neurons and how the strong points of these synapses change with different sorts of mental activity. This technique has been often used to manage and detect fraud as it allows us to quickly detect irregular data within an operation's large volume of information. In a neural network, connections are made between the data features and a result. These features are super variables that have been determined by analytics experts to turn them into features of a high predictive level. For instance, in fraud detection, one of these features could be the number of transactions made with a credit card in the past hour or the number of transactions of a certain sum made in the past twelve hours compared to the total volume of transactions.

 

1  2  3  Next Page 

Sign up for CIO Asia eNewsletters.