Research On Evaluation Of Short-term Debt Ability Of Listed Companies Based On Machine Learning

With the rapid development of modern economy and the improvement of China's financial analysis system,it is becoming more and more important to scientifically evaluate and analyze the short-term solvency of enterprises.Short-term solvency refers to the use of corporate liquid assets to repay short-term debt due.In the course of business operation,an enterprise cannot blindly pursue the profitability of the enterprise and ignore the debt repayment problem of the enterprise,nor can it accumulate excessive liquid assets and cause idle funds,so that the development space of the enterprise is not fully utilized.Enterprises should seek better development on the premise that short-term liquidity is sufficient to repay daily debts.Under such circumstances,it is particularly important to make a correct assessment of the short-term solvency of the enterprise and provide a basis for the correct decision-making of enterprise managers,investors or other relevant parties.This article is mainly based on the existing common indicators of short-term solvency,and further analyzes other financial indicators that affect the short-term solvency of the enterprise,and provides theoretical support for the construction of a comprehensive evaluation system of short-term solvency from the perspective of economic theory.Factor analysis was performed on the sample data of 306 listed companies in multiple industries collected,and the comprehensive score was calculated.The comprehensive score was divided into five categories using the asset liquidity reflected by short-term debt repayment and the K-means clustering method.Analysis of the performance of short-term debt solvency of enterprises under various types shows that: filling enterprises are in the optimal state,while crisis-type enterprises have the weakest short-term solvency,measures should be taken as soon as possible to ensure the normal operation of enterprises.Short-term solvency is somewhere in between,and most(about 41%)companies are in equilibrium.The scientific and rationality of the classification results are verified by the performance of the short-term debt service of special enterprises(ST companies)and the financial performance of the enterprises under different asset flows.Finally,the classifications obtained by factor analysis and clustering analysis are used as labels,and machine learning methods such as BP neural network and support vector machine are used to construct a comprehensive evaluation classification model.The classification comparisons are analyzed and finally found that the BP neural network classifies each classification.The classification accuracy rate is high,the test sample accuracy rate reaches 91.304%,the accuracy rate of each type can reach more than 85%,and the classification effect is remarkable.Therefore,the combination of BP neural network and factor analysis model is selected to form an optimal evaluation model to comprehensively evaluate the short-term debt solvency of listed companies.