By: Martín Rago
The current generation has the rare privilege of living through the biggest change in human history in the world of work, here and now: the industrial revolution based on Machine Learning technology. This is the ability of computers to learn to perform really complex tasks, through algorithms and the enormous processing power available.
The exponential advance that is occurring is because computers are not programmed to perform tasks, but to learn how to perform them, usually using historical information. That is the paradigm shift.
The adoption of this technology has led to huge improvements in a number of areas: in diagnostic imaging for the detection of diseases or the identification of heart problems; in transport with self-driving cars and trucks; in law firms where algorithms learn based on historical information composed of all laws, all jurisprudence of all cases and court rulings, together with all the results of the expert opinions carried out in order to design the best defence strategy; in the retail industry, algorithms propose products of interest to the consumer based on the purchases of other people with similar tastes; in leisure activities and social networks to capture the individual's attention most of the time; among many others.
Machine learning technologies also have a use in fraud prevention with a huge economic value contribution; and analysts face the challenge of becoming good stewards of it to have the best hybrid team of humans and machines. Incorporating machine learning is like having Lionel Messi on the fraud prevention team.
Algorithms can learn to discern what small details and characteristics make a transaction fraudulent. They use historical information from millions of transactions and thousands of variables to achieve the best results in the level of detection and false positive ratio.
On the other hand, this world of enormous advances has started a race among financial institutions for who becomes the leader in the age of digitisation. Open institutions, digital linking, and omni-channeling have opened a door to new opportunities for fraudsters, who nefariously take advantage of technological advances.
Monitor Plus® uses machine learning based on deep learning and the XGBoost and LightGBM algorithms to jointly meet the necessary security and customer experience requirements. This solution meets today's needs, has the flexibility needed for the channels and products of the future, and contains rules-based technology to incorporate new fraud typologies not foreseen during training. In addition, it makes use of communications, incorporating the customer as part of risk management through two-way messaging.
