Network Feature Extraction for Regional Economic Development and Financial Agglomeration Analysis

The feature components extracted by the ISA neural network have good mutual independence and good translation invariance, scale invariance and rotation invariance. It can better acquire hidden internal feature information and is successfully applied in the field of human behavior recognition. Through the optimization process, the ISA neural network can complete the learning phase to train the weight parameters. This paper analyzes regional economic development and financial agglomeration using machine learning algorithms and deep learning, and uses a spatial econometric model to empirically study the differences between financial industry agglomeration and regional economic development. This paper measures the degree offinancial agglomeration and analyzes the degree of financial agglomeration of China’s regional economy and integrates the empirical results. The paper provides a certain reference value for the development of the regional economy.Keywords: Deep Learning; Network Feature Extraction; Deep Network; Financial agglomerationCite AsY. Zhang, Q. Qiang, W. Zhu, "Network Feature Extraction for Regional Economic Development and Financial Agglomeration Analysis", Engineering Intelligent Systems, vol. 27 no. 2, pp. 71-78, 2019.