Fixing the Nigerian Credit Score System through ML&AI Models: My NUTM Hackathon Experience

Back in the university, I was so excited about the introduction to computing and programming course that I had an A in the final exam. I had fantasized about turning my ideas into products and I believed that being a good programmer would be a great leverage to achieve that. I was in my second year studying electrical and electronics engineering. I remember how excited I was when I got my first programming textbook: Java How To Program by Paul Deitel and Harvey Deitel.

But then something dramatic happened, a new lecturer was introduced to teach the course in my third year. I just couldn’t stand him and found his teaching style boring. I struggled to grasp the concepts he was trying to teach me. I remember him telling the whole class during his second session with us that his wife was more intelligent than all of us combined. I was so disheartened. I began to ‘hate’ the lecturer and started losing interest in the course. I only did the minimum to get a C.

Fast forward a few years, and I found myself wanting to build a project but my lack of programming skills was holding me back. I realized then that my attitude towards that lecturer had greatly affected my reception of and attitude towards the knowledge.

If you’re an undergraduate, no matter how hard or frustrating your lecturer is, try to separate their personality from the course and find some interesting things around the course that will make you absorb the knowledge. There are lots of resources online (YouTube, free courses from Google and the likes) you can use to supplement whatever gap they’re not filling. Don’t let your dislike for a lecturer affect your ability to learn and grow. You never know when that knowledge will come in handy in the future.

This is the advice I wish I had received back then.

Anyway, I started taking programming more seriously last year. I started with Python and fortunately, I had the opportunity to work on practical projects in the computing and programming course at NUTM with other fascinating colleagues.

With that being said, I am thrilled to announce that my team and I just won first place in the Nigerian University of Technology and Management (NUTM) programming hackathon 3.0 for our project, MIOS Metre.

You might have been a victim to an embarrassing call from a loan application company telling you that your so and so friend is defaulting on their loan payment. It’s a lot of work for financial institutions to know if an individual is creditworthy. In fact, in 2020 the five major banks in Nigeria (also known as Tier-1 banks) reported a total of N683bn in non-performing loans.

In 2020 the five major banks in Nigeria (also known as Tier-1 banks) reported a total of N683bn in non-performing loans.

In developed countries like the USA it’s quite easy. At birth, every child is assigned a Social Security Number (SSN) which is a unique identifier that allows government agencies to easily identify individuals in their records and businesses to keep track of an individual’s financial information. With a Social Security Number (SSN) the government and financial institutions can look up your credit score and determine which kind of loan or mortgage you get.

In countries like Malaysia, Hong Kong, and Singapore, the credit reporting system is more advanced than that in the UK and USA. This is due to better technology and database systems. Additionally, credit reporting for consumers is combined with credit reporting for businesses, which is particularly beneficial for small and medium-sized enterprises looking to obtain credit.

Our team recognized that the Nigerian traditional model of credit scoring, which relies heavily on information supplied by the loan applicant and financial records, is not yielding the desired result. So, we proposed a new solution: by incorporating behavioral data, we created a more reliable credit scoring model using a highly accurate machine learning model and Artificial Intelligence.

With MIOS Metre, financial institutions can make more informed lending decisions by accurately predicting the creditworthiness of individuals applying for loans on their platforms. This means less time and resources spent chasing after loan defaulters.

I would like to thank my team members, Marybeth Ubanwa , Omona D. , Muhammad Aliyu , Shukurat Bello , Mustapha Umar, MBA , and Ibukunoluwa Omotola for their hard work and dedication to making this project a success.

During the breakout sessions, Tochi Ebere gave some insights that helped us refine our model before the presentation. So did the other judges; Kelechi Iwuagwu , Ola O. , Bright Williams ; our guest speaker from Lagos Business School, Pan-Atlantic University Chuka Unozu, and our visiting faculty Prof Thomas Joseph.

My warm gratitude to the Ydevacademy team led by Bernard Okobiemen (O'Bien), Debo Odunlami for being on our neck and making sure we learn, and Gabriel Olatunji for the untiring support.

Thanks to the Nigerian University of Technology and Management and our Program Coordinator Chika Ike, PHRi for making this hackathon possible. And to the NUTM Scholars who did remarkable projects in their various groups, you guys rock! A special shoutout to Temiloluwa Osobisi for capturing these beautiful memories through his photography. I highly recommend him for all your media coverage needs. He'll capture all your moments in a unique and heartfelt way.

I am so grateful for this opportunity and I can’t wait to see where this project will take us.

#MIOSMetre #programming #hackathon #teamwork #creditscore #fintech #machinelearning #ai