Research On Identification Of Big Data Technology In Internet Financial Credit Risk

According to the relevant information released by China Internet Information Center,the number of Internet users in China has reached 772 million in 2017,accounting for 55.8%of the total population of the country.The Internet has penetrated into our daily lives.The Internet has a growing influence on people's lives,and the Internet financial platform is also emerging.According to the data released by the Zero Data,it can be seen that by the end of 2018,the number of P2P network lending platforms has reached 6063,of which the operation is normal.The number of platforms in the state totaled 1,185,accounting for 19.5%of the total number of platforms,a year-on-year decrease of 46.8%,and the number of platforms in an abnormal state totaled 4,6725 accounting for 77.1%of the total number of platforms.The number of problem platforms is large enough to make people worry,and with the generation of credit risks,the loans released by many platforms cannot be recovered,and the non-performing loan ratio has increased.And with the rapid development of Internet teehnology,Internet finance has gradually emerged in the market,and this model has achieved rapid development in a short period of time.From the initial online shopping,to online banking,third-party payment,and then to online banking,it took only a few years.Especially in the P2P industry,the development in recent years has reached an alarming rate,but there have also been a series of credit risk issues.In order to ensure that the credit risk of the Internet financial platform can be effectively reduced,it is necessary to use modern big data technology to rate the risk.This paper mainly studies the large amount of data formed by users on the Internet,and uses data crawler technology to collect and store data.transmission.Then use the big data analysis software RapidMiner big data analysis software to pre-process the crawled data,and use the random forest algorithm to match the best data and rule scheme in the software.Different from the traditional Internet financial credit risk research,based on the random forest algorithm,this paper uses logistic regression analysis to measure the user default probability according to the rules,predicts the user's default and fraud,and analyzes the result.Therefore,it is proposed to effectively reduce the credit risk of Internet financial platform enterprises.