With the rapid development of information technology and globalization of economy, financial data are being generated and collected at an unprecedented rate. Consequently, there has been a dire need of automated methods for effective and proficient utilization of a substantial amount of financial data to help in investment planning and decision-making. Data mining methods have been employed to discover hidden patterns and estimate future tendencies in financial markets. In this article, an improved macroeconomic growth prediction algorithm based on data mining and fuzzy correlation analysis is presented. This study analyzes the sequence of economic characteristics, reorganizes the spatial structure of economic characteristics, and integrates the statistical information of economic data. Using the optimized Apriori algorithm, the association rules between macroeconomic data are generated. Distinct features are extracted according to association rules using the joint distribution characteristic quantity of macroeconomic time series. Moreover, the Doppler parameter of macroeconomic time series growth prediction is calculated, and the residual analysis method of the regression model is used to predict the growth of macroeconomic data. Experimental results show that the proposed algorithm has better adaptability, less computation time, and higher prediction accuracy of economic data mining.

中文翻译：

---

基于数据挖掘的宏观经济增长预测算法设计

随着信息技术的飞速发展和经济全球化，金融数据以前所未有的速度生成和收集。因此，迫切需要自动化方法来有效和熟练地利用大量财务数据来帮助进行投资规划和决策。数据挖掘方法已被用于发现隐藏的模式并估计金融市场的未来趋势。本文提出了一种基于数据挖掘和模糊关联分析的改进的宏观经济增长预测算法。本研究分析经济特征序列，重组经济特征空间结构，整合经济数据统计信息。使用优化的 Apriori 算法，宏观经济数据之间的关联规则生成。利用宏观经济时间序列的联合分布特征量，根据关联规则提取特征。并计算宏观经济时间序列增长预测的多普勒参数，利用回归模型的残差分析方法对宏观经济数据的增长进行预测。实验结果表明，该算法具有更好的适应性、更少的计算时间和更高的经济数据挖掘预测精度。并利用回归模型的残差分析方法预测宏观经济数据的增长。实验结果表明，该算法具有更好的适应性、更少的计算时间和更高的经济数据挖掘预测精度。并利用回归模型的残差分析方法预测宏观经济数据的增长。实验结果表明，该算法具有更好的适应性、更少的计算时间和更高的经济数据挖掘预测精度。