Research on the evolution and driving forces of the manufacturing industry during the "13th five-year plan" period in Jiangsu province of China based on natural language processing.

Shiguang Shen,Chaoyang Zhu,Chenjing Fan,Chengcheng Wu,Xinran Huang,Lin Zhou,Jun Yang

doi:10.1371/journal.pone.0256162

Shiguang Shen, Chaoyang Zhu + Show 5 more

Open Access

https://doi.org/10.1371/journal.pone.0256162

Copy DOI

Abstract

The development of China's manufacturing industry has received global attention. However, research on the distribution pattern, changes, and driving forces of the manufacturing industry has been limited by the accessibility of data. This study proposes a method for classifying based on natural language processing. A case study was conducted employing this method, hotspot detection and driving force analysis, wherein the driving forces industrial development during the "13th Five-Year plan" period in Jiangsu province were determined. The main conclusions of the empirical case study are as follows. 1) Through the acquisition of Amap's point-of-interest (POI, a special point location that commonly used in modern automotive navigation systems.) data, an industry type classification algorithm based on the natural language processing of POI names is proposed, with Jiangsu Province serving as an example. The empirical test shows that the accuracy was 95%, and the kappa coefficient was 0.872. 2) The seven types of manufacturing industries including the pulp and paper (PP) industry, metallurgical chemical (MC) industry, pharmaceutical manufacturing (PM) industry, machinery and electronics (ME) industry, wood furniture (WF) industry, textile clothing (TC) industry, and agricultural and food product processing (AF) industry are drawn through a 1 km× 1km projection grid. The evolution map of the spatial pattern and the density field hotspots are also drawn. 3) After analyzing the driving forces of the changes in the number of manufacturing industries mentioned above, we found that manufacturing base, distance from town, population, GDP per capita, distance from the railway station were the significant driving factors of changes in the manufacturing industries mentioned above. The results of this research can help guide the development of manufacturing industries, maximize the advantages of regional factors and conditions, and provide insight into how the spatial layout of the manufacturing industry could be optimized.

Highlights

China’s economic development, especially in its manufacturing sector, has received global attention [1]
1) Through the acquisition of Amap’s point-of-interest (POI, a special point location that commonly used in modern automotive navigation systems.) data, an industry type classification algorithm based on the natural language processing of POI names is proposed, with Jiangsu Province serving as an example
The natural language processing algorithm proposed in this paper can realize the classification of POI names

Summary

Introduction

China’s economic development, especially in its manufacturing sector, has received global attention [1]. The manufacturing industry is the mainstay of China’s national economy; China’s rapid economic growth has been largely driven by the development of its manufacturing industry. Two main patterns have been noted during the rapid expansion of China’s manufacturing industry: the manufacturing industry has continued to grow both in terms of quality and quantity [4], and manufacturing activities are concentrated in a few regions [5]. The spatial patterns of manufacturing vary in different regions because of differences in human capital, innovation capacity, technology absorptive capacity, and capital policies [6]. The coastal provinces in eastern China have advantageous geographical locations and greater economic capital and technological development and have played a critical role in China’s economic development

Methods

Findings

Discussion

Conclusion