Instantor's Blog

Machine learning has been an incredible breakthrough in the past quarter-century. From diagnostic decisions in hospitals to loan approvals in banking, the conclusions made by machine learning algorithms have key implications on our personal lives, allowing us to process larger quantities of data faster than ever before.

While machine learning and associated artificial intelligence inarguably provide many benefits, unfortunately, they are also prone to error through biases, just like humans. Once we understand how algorithms are designed, we realise that these algorithms that we place so much trust in will sometimes get things wrong.

The financial services landscape has been radically evolving for the past decade, and credit scoring is no exception. Once the domain of a small number of credit bureaus, alternative credit scoring is democratising the credit underwriting process for banks and other lenders. This global trend shows no signs of slowing down, especially with the rise of digital-only banks and as more and more people use banking in unconventional ways, e.g. applying for a mortgage online.

Getting credit or a mortgage declined is a big concern for many of us. In fact, most of us will need to use credit at some point in our lives, so it’s a big deal when we are declined. That’s why credit scores are important – to improve our chances of gaining access to credit and enabling us to navigate quickly through life’s twists and turns. 

At Instantor we want to contribute to a more inclusive and fair financing market. We do this by making people's financial lives easier and helping organisations to understand their customers’ true financial capacity, using transactional data.

Our client, Lendify – Sweden’s largest marketplace for loans – wrote a great opinion article on how traditional credit assessments need to be updated. How new technology, like our AI credit predictor – Insight – can provide a more accurate and detailed risk analysis for lending organisations and be a force for a healthier lending market. Please enjoy a translated version by us that was originally published in Swedish by Realtid.se in October 2018.

AI has been gaining popularity in recent years, particularly as it’s been improving in leaps and bounds for a whole range of tasks: from predicting the risk of child abuse in a given family to beating world champion Go players. 

However, one critical problem with using it for decision making in business is that the machine learning models can't explain the predictions they make: a customer might, reasonably, want to know why their cred it card or home loan application was denied. 

Our latest solution, Instant Access, gives you the power to analyse transactional data through PSD2. This data is used to digitalise credit risk processes – without requiring you to invest in any tech resources or spend any time on system integration. Moneyveo is the first lender in the Ukrainian market to use this solution and our very first partner in Ukraine. 

Just a few years ago, only a few people in the financial industry had access to an individual’s banking data. Data was ‘owned’ by a few major players such as large financial institutions, and all access had to go through these information gatekeepers. Limited sharing of data has resulted in a stagnated offering of banking products to customers that have not significantly changed in decades: people are paying too much for their overdrafts or money sits in current accounts earning little, or sometimes even no interest. To summarise, customers are not happy.

To eliminate extreme poverty, many leading global organisations have made financial inclusion a top priority. Business leaders and policymakers have dedicated themselves to improving the quality of life for the world’s poorest, and they believe economic and social progress starts with an inclusive financial system that meets the needs for all income levels.

The traditional credit scoring system has long been used as the primary method to rate a person’s creditworthiness. The current rating system is led by a select handful of major credit bureaus and agencies within each country, who hold the power to determine an individual’s credit eligibility. Yet new systems, known as alternative credit scoring methods, make a strong case for providing a fairer and more modern scoring system and are already making huge strides in shaking up the industry.

This post is for risk managers who want to explore tackling underwriting challenges through an Artificial Intelligence or Machine Learning (ML) toolbox. Here we examine two of the most popular tools for assessing the accuracy of a credit scoring models; Area Under the Curve (AUC) and Gini coefficient (Gini). We will briefly explain what AUC and Gini mean and how to interpret them in the context of credit scoring.