A characteristic of an industry-leading company is its forward-thinking, innovative employees. People such as the Fulcrum Group’s David Kenyon.
Kenyon, a systems analyst for Fulcrum Delta, has established himself as a progressive thinker on how to use big data and artificial intelligence to combat insurance fraud – a very real (and expensive) problem the world over.
He presented an academic paper, titled Big Data Science for Predicting Insurance Claims Fraud, at the recent 16th international ISSA Conference in Johannesburg. This is a very big deal, for two principal reasons: firstly, only 30% of papers submitted for the conference make it through a double blind review process, and secondly, his paper has been published by the prestigious US-based Institute of Electrical and Electronics Engineers (IEEE).
ISSA – for Information Security South Africa – is an annual conference held under the auspices of the University of Johannesburg Academy for Computer Science and Software Engineering, the University of South Africa School of Computing and the University of Pretoria School of Information Technology.
The IEEE is the world's largest technical professional organisation for the advancement of technology, with more than 420 000 members in over 160 countries. It is concerned with the educational and technical advancement of electrical and electronic engineering, telecommunications, computer engineering and related disciplines.
For Kenyon, it has been a nerve-wracking – but gratifying – experience. And he regards the publication of his paper by the IEEE as a feather in his cap.
“It’s quite something to have my paper published by the IEEE, which is the leading global organisation of its kind, and be recognised by my peers,” he says.
“I was quite nervous for the presentation, but I think things went well. In fact, I was approached afterwards to prepare a paper for another conference – which is definitely a good sign.”
This is the description of the paper Kenyon presented at ISSA: “There is an increase in the cost of insurance on a global scale. Contributing factors for this increase include new legislation and fraud. These factors are problematic as new legislation is not easily controllable and fraud is very difficult to detect. It can therefore be said that controlling this upward trend is problematic.
“We therefore investigated the use of big data and data science to proactively predict insurance fraud. Big data and machine learning (artificial intelligence) were applied to the short-term insurance industry of a developing world country. This study was performed within constraints such as new privacy legislation which defines standards on how data can be stored and shared.
“We, therefore, suggested and put forward a privacy preservation method of predicting insurance claims fraud to the conference.”
Clodagh da Paixao, Fulcrum’s head of Brand, says the group is proud of Kenyon’s work.
“Fulcrum, as an organisation, strives always to be a frontrunner in the insurance industry. Because we continuously improve our offerings and Paul Schreuder, our CIO, has a strong focus on data, Fulcrum employees are encouraged to be analytical and forward-thinking,” she says.
“David has done us proud with his paper, which is being shared with the leading minds in the information and technology fields the world over. We’re very proud to call him our colleague!”
The paper will be available soon on the IEEE Xplore Digital Library, with the ISBN number 978-1-5386-0544-8 and IEEE Catalog Number CFP1766I-USB.