Pavlović, Branko

  • Email:

Predsednik Udruženja aktuara Srbije
I-mejl: pavlovic.branko@gmail.com

368:371.694:61:659.24:681.3.064:004.4:004.021:679.058:681.322.2:681.3.06 (4-672EEZ) (73)(479.11)
ARTICLES

CHALLENGES IN APPLICATION OF MACHINE LEARNING IN INSURANCE INDUSTRY

The Internet of Intelligent Things, Blockchain technology2 , software robots3 and various aspects of artificial intelligence such as machine learning are often referenced in modern literature as having great potential to improve the processes in insurance business. This paper will focus on machine learning. Machine learning is a subset of artificial intelligence intended for study and recognition of behaviour patterns by using statistical methods for available data processing. In other words, machine learning is a software able to make its own decisions based on previous experience. The key benefit companies can draw from machine learning is prediction of future trends, where software independently discovers patterns in the available data. Insurance companies may use machine learning to optimize their tariffs, settle claims more efficiently, and improve the quality of loss reserving, and as a powerful tool in combating frauds.

Key words: machine learning, insurance

en_GBEN