Research And Development Of Financial Oriented Liquidity Data Analysis And Visualization System

The acceleration of financial system reform has driven changes in the interest rate market,posing a huge threat to the stable operation of commercial banks due to deposit loss,asset price changes,and market behavior convergence.The long-term existence of bank funding gaps has forced the emergence of liquidity risks.Liquidity risk occurs when banks are unable to raise sufficient funds in a short period of time to meet the repayment of maturing debts and maintain normal operations.Once liquidity risk occurs,it is easy for banks to fall into operational difficulties.At present,financial managers obtain potential information through simple indicator statistics and stress testing of liquidity data,which poses the problem of incomplete and inaccurate liquidity risk management.It is urgent to use big data technology for in-depth analysis to accurately identify hidden liquidity risks,in order to prevent and control the occurrence of risks.To achieve effective management of commercial bank liquidity data and identify potential liquidity risks,this article evaluates liquidity risk by aggregating liquidity data distributed across multiple business systems and analyzing the correlation between each data.Due to the constant impact of environmental factors on liquidity risk,a weighted multi random decision tree algorithm is introduced to retain the advantages of the random decision tree algorithm while further improving the accuracy of liquidity risk assessment.At the same time,a Spark based parallelization scheme for multiple random decision trees was designed to improve the operational efficiency of the algorithm.Considering the low probability of significant liquidity risks occurring,the corresponding proportion of data in the experiment was reasonably adjusted,and a stratified sampling method was added to the random sampling method to improve the accuracy of predicting different risk levels.This thesis designs a liquidity data analysis and visualization system for finance,which mainly consists of four parts: data collection,data management,data analysis,and visualization display.Among them,the data collection section realizes the collection of liquidity data distributed in multi business systems;The data management section cleans,merges,and categorizes the collected liquidity data;The data analysis section enables efficient analysis and calculation of liquidity data to determine liquidity risk;The data visualization part adopts the front end and rear end separation strategy to achieve the visualization of liquidity data and liquidity risk.The experimental results indicate that the system in this paper can reasonably analyze and manage liquidity data,accurately identify the current situation of liquidity risk,better assist banks in managing liquidity risk,and provide strong support for financial industry managers’ decision-making.