Abstract
With the widespread application of IoT technology in the world, the new industry of IoT finance has emerged. Under this new business model, commercial banks and other financial institutions can realize safer and more convenient financial services such as payment, financing and asset management through the application of IoT technology and communication network technology. In the cloud computing model, the local terminal device of IOT will transmit the collected data to the cloud server through the network, and the cloud server will complete the data operation. Cloud computing model can well solve the problem of poor performance of IoT devices, but with the increasing number of IoT terminal devices and huge number of devices accessing the network, cloud computing model is constrained by network bandwidth and performance bottleneck, which brings a series of problems such as high latency, poor real-time and low security. In this paper, based on the new industry of IoT finance which is developing rapidly, we construct a POT (Peaks Over Threshold) over threshold model to empirically analyze the operational risk of commercial banks by using the risk loss data of commercial banks, and estimate the corresponding ES values by using the control variables method to measure the operational risk of traditional commercial banks and IoT finance respectively, and compare the total ES values of the two. This paper adopts the control variable method to reduce the frequency of each type of loss events of operational risk of commercial banks in China respectively.
Highlights
Related WorkDue to the strict regulation of foreign financial environment, the transformation of IoT technology for finance is restricted, which leads to no scholars in foreign countries to conduct segmented research on IoT finance [13]
Image processing can be combined with a variety of application scenarios, and has huge application space in various fields
We collect the loss data of commercial banks from 2010 to 2017 from public sources for each type of operational risk, and by constructing the extreme value theory model POT over threshold model and combining with the control variables method, we measure and empirically analyze the operational risk of traditional commercial banks and commercial banks in the IoT financial industry, and calculate the corresponding VaR and ES values respectively to compare the operational risk of traditional commercial banks and IoT financial industry
Summary
Due to the strict regulation of foreign financial environment, the transformation of IoT technology for finance is restricted, which leads to no scholars in foreign countries to conduct segmented research on IoT finance [13]. In response to the poor real-time computing performance in the cloud, as well as the high requirements for power consumption, storage and network bandwidth, Ancuti et al designed an edge computing system based on artificial intelligence image processing, in which crowd information data will be processed at the edge to accelerate the computing of crowd information analysis and save network bandwidth and power consumption [18]. For cloud computing requires real-time image uploading from the edge to the central server for unified analysis and processing, which has poor real-time performance, and requires high power consumption, storage and network bandwidth for a series of problems, this project proposes a convolutional neural network-based edge computing model for real-time crowd information detection [19]. The cloud server can perform further in-depth analysis, visualization processing and intelligent decision on the data, and make backups of the historical data
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
