Machine Learning and Big Data in Finance Services

Abstract

Today, the combination of machine learning (ML) and big data is gaining much importance in the academic and industry fields. ML (a branch of artificial intelligence) can help to make predictions and to extract intelligent decisions. In the finance domain, ML can help to detect fraud, forecast trading, reach new customers and provide smart decisions. ML relies on historical data collected from different data sources. Nowadays, data can be generated at a very high rate from different data sources (e.g., social media, stock trends, Internet of Things); big data technology can manage data with respect to the three Vs characteristics: volume, variety and velocity. The volume refers to the huge amount of data collected while the variety means that data can be of different natures (e.g., structured, non-structured, semi-structured). The velocity means that data are generated at a very high speed such as in social media. Given this nature of data, ML algorithms need to be expanded in order to deal with different types of data. In this chapter, we will discuss this interesting combination of big data and ML for financial services. We will highlight the main advantages of this integration showing how this can be applied in the financial services. Furthermore, we will present the current challenges and opportunities in this area.