Intelligent Early Warning of Internet Financial Risks Based on Mobile Computing

Intelligent Early Warning of Internet Financial Risks Based on Mobile Computing

Mu Sheng Dong (Collaborative Innovation Center for Green Development in the Wulin San, Yangtze Normal University, China)
DOI: 10.4018/IJMCMC.2020040104
OnDemand PDF Download:
No Current Special Offers


In order to establish the early warning model of internet finance, K-means algorithm improved by quantum evolutionary is used in this paper to divide risk early-warning interval by combining with the given initial value and the value-at-risk measured by China's well-known internet finance company. With the characteristics of parallelism and randomness, quantization algorithm is introduced into K-means algorithm to improve the search efficiency of original algorithm on the basis of maintaining the diversity of population. The sample is conducted with optimal segmentation by using improved algorithm to obtain the accurate early-warning interval and then the risk prediction model for internet financial institutions will be established by using the advantages of GMDH predictive mining and combining with the value-at-risk measured by “Renren Loan” Company. The effectiveness of early-warning model will be illustrated by comparing the actual situation of internet financial companies with more than 40,000 data of “Renren Loan” Company from January 2017 to October 2018.
Article Preview

2. Literature Review

Early warning of internet financial institutions has always been a hot topic at home and abroad, with different views from different perspectives, mainly including the following.

KLR model put forward by Kaminsky is a signal analysis model with monthly or quarterly data, whose prediction features are as follows: The historical data are used to synthesize leading indicators; crisis is defined as exceeding the threshold value, namely, critical value; model accuracy is high, but the design of indicators has a tendency (Kaminsky & Reinhart 1999). The artificial neural network model proposed by Nag breaks through the linear normal form of traditional model, whose advantages lie in the flexible rules and the ability to capture complex relationships among variables (Nag 2012). Zhao, al pointed out that with the development of Internet finance, higher requirements have been put forward for the unification of its supervision, including the unified supervision over cross-industry internet financial institutions, cross-market financial activities and trading procedures across time and space. He believed that it is essential to make a clear distinction between the powers and responsibilities of the central and local financial supervision in a timely and reasonable manner, and to strengthen the supervision and coordination between the central administrative departments and local governments, as well as between local governments, in order to meet the challenges of new finance (Zhao et al., 2018). Mollick argued that public financing refers to that individual entrepreneurs or business groups with cultural, social or commercial purposes attract a considerable number of individuals to invest relatively small amounts of money through the internet without resorting to the intermediary role of traditional financial institutions, in order to achieve their creative behaviors (Mollick 2014). Samreen found that borrowers’ credit and transparency of information play an important role in the internet financial P2P lending model. They pointed out that the lower a borrower's credit, the higher his or her loan and default rates; and the higher the openness and transparency of information, the lower the corresponding default rate of bad debt. Lin also found that borrowers’ credit and transparency of information play an important role in the internet financial P2P lending model (Samreen 2018).

Complete Article List

Search this Journal:
Volume 14: 1 Issue (2023)
Volume 13: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing