Large-scale data-driven financial risk management & analysis using machine learning strategies

Recently the wave of financial crises that have shaken the economy and financial world have caused severe bank
losses. Some researchers have focused on examining catastrophes to develop an early warning system to handle
financial risks. Financial experts and academics are increasingly interested in developing big data financial risk prevention and control capabilities based on cutting-edge technologies like big data, machine learning (ML), and
neural networks (NN), as well as accelerating the implementation of intelligent risk prevention and control
platforms. This research analyzed and processed the large-scale datasets before training and evaluated using the
three models – cluster based K-nearest neighbor (KNN), cluster based logistic regression (LR), and cluster based
XG Boost for their ability to predict loan defaults and their occurrence of likelihood. The investor’s wealth
proportion measure of the proposed model ranges from 0.02 to 0.09. Applying the value-at-risk strategy, the
optimal consumption stability not exceeded 5% of the total investment wealth. The simulation results of the
proposed model obtained better results of large-scale data-driven financial risks over the state-of-the-art
methods. In this article XG Boost, KNN are the machine learning are proposed for financial risk management
with IOT deployement.