基于长短期记忆网络与聚类分析的县域经济发展差异研究——以晋江数据为参照的福建省实证分析  

Research on the difference of county economic development based on long short-term memory network and cluster analysis——Empirical analysis of Fujian Province with Jinjiang data as reference

作　　者：潘郡 PAN Jun(College of Digital Intelligence and Financial Management,Minnan University of Science and Technology,Quanzhou,Fujian 327000,China)

机构地区：[1]闽南理工学院数智财务管理学院,福建泉州327000

出　　处：《中国科技论文在线精品论文》2024年第4期554-560,共7页Highlights of Sciencepaper Online

基　　金：2023年度福建省教育系统哲学社会科学研究省中青年教师教育科研项目(JAS23169)。

摘　　要：本文通过分析福建省各县市相对于晋江的经济、社会和环境指标,揭示了区域发展的差异。采用长短期记忆网络(long short-term memory,LSTM)降维,并结合K均值聚类算法,将福建省各县市划分为三类。聚类0的城市经济中等,具发展潜力;聚类1的城市经济较发达,产业多元化,表现最接近晋江;聚类2的城市相对落后,需更多政策支持和资源投入。研究结果为区域经济发展提供了新的视角,并为政策制定者制定扶持措施、推广晋江经验提供了依据。This study analyzes the economic,social and environmental indicators of counties in Fujian Province relative to Jinjiang,revealing regional development disparities.Using long short-term memory(LSTM)for dimensionality reduction and K-means clustering,counties are categorized into three groups.Group 0 includes cities with moderate economic potential;Group 1 comprises cities with diverse industries and strong economies,closely aligned with Jinjiang;Group 2 consists of underdeveloped cities requiring more policy support.The findings offer new insights into regional economic development and serve as a basis for policymakers to implement targeted measures and promote Jinjiang’s experience.

关 键 词：农村经济学 晋江经验 区域经济 LSTM 聚类分析 

分 类 号：F202[经济管理—国民经济]