基于文本挖掘的我国制造业上市企业数字化水平测度研究
Research on Measuring the Digitalization Level of Listed Manufacturing Companies in China Based on Text Mining

数字化水平作为衡量中国上市企业发展前景的主要依据，已经成为了每个上市企业所关心的问题，尤其是在制造业领域，由于生产流程与产业链的复杂性，数字化水平测度一直是制造业企业面临的重要问题。基于制造业上市公司披露的年报的文本信息，通过Python程序批量抓取年报中与“数字化水平”相关的关键词进行词频统计，据此量化企业数字化水平，并对我国制造业上市企业数字化水平进行横向与纵向对比。研究表明：整体上，中国上市企业数字化整体水平在2013年到2022年逐年提高，但是在2013~2014出现了异常的下降；从行业来看，从2013年到2022年，所有行业的数字化平均水平也在逐年提高，机械电子制造业也一直维持在较高水平；从区域来看，我国上市企业数字化水平发展极不均衡，呈现东部强，中、西部弱的特点。
The level of digitalization, as the main basis for measuring the development prospects of Chinese listed enterprises, has become a concern for every listed enterprise, especially in the manufacturing industry, due to the complexity of production processes and industrial chains, measuring the level of digitization has always been an important issue faced by manufacturing enterprises. Based on the textual information disclosed by listed manufacturing companies in their annual reports, Python programs were used to capture keywords related to “digitalization level” in the annual reports for word frequency statistics and quantify the digitalization level, and the horizontal and vertical comparison was analyzed for digitalization level of Chinese manufacturing listed companies. Research shows that: (a) On the whole, the digitalization level of Chinese listed companies has been increasing year by year from 2013 to 2022, but there has been an abnormal decline from 2013 to 2014; (b) From the perspective of industry, the average level of digitalization in all industries has been increasing during 2013 to 2022, and the mechanical and electronic manufacturing industry has also maintained a relatively high level; (c) From a regional perspective, the digitalization level of listed companies in China is extremely uneven, showing a strong eastern region and weak central and western regions.