Big Data-Driven Corporate Financial Forecasting and Decision Support: A Study of CNN-LSTM Machine Learning Models

AiXiang Yang ^*

• Chengdu College of Arts and Sciences, Chengdu, China

With the rapid advancement of information technology, especially the widespread use of big data and machine learning technologies, corporate financial management is undergoing unprecedented transformation. Traditional methods often fall short in providing accurate forecasts, fast decision support, and flexible operations. To address these issues, this study proposes a hybrid model combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks, aiming to improve the prediction accuracy and decision efficiency of corporate financial data. The study utilizes financial data from A-share listed companies in the CSMAR database from 2000 to 2023, covering 54 key financial indicators with a total of 54,389 observations. The data is first preprocessed and reduced using Principal Component Analysis (PCA) to remove redundancy and noise. Then, the CNN-LSTM hybrid model is applied to the reduced data for training and testing. The experimental results show that the proposed model has an MSE of 0.020 and an R² of 0.411, significantly outperforming traditional ARIMA, Random Forest, XGBoost, and standalone LSTM models. A practical enterprise case analysis further validates the model’s effectiveness in improving financial forecasting accuracy, optimizing decision-making efficiency, and reducing financial risks. Overall, the financial forecasting and decision support system based on big data and machine learning can significantly enhance corporate financial management, helping businesses achieve more efficient risk control and sustainable development in an uncertain market environment.