基于文本挖掘对新旧保险合同准则的分析
Analysis of Old and New Insurance Contract Guidelines Based on Text Mining

新保险合同准则于今年的1月1日开始施行，它标志着我国进一步完善企业会计准则体系，保持与国际财务报告准则的持续趋同。因此，本文利用Python程序语言，通过网页爬虫对2020年发布的新旧保险合同准则进行文本挖掘。另外下载我国财政部于2006年和2009年印发的三份旧保险合同准则，通过本地文本导入，利用jieba库进行文本分词，collection库统计词频，wordcloud库制作词云等，进行对比、联系，发现新旧保险合同都出现如“合同”、“计量”等词汇。总结出新保险准则着重关注在保险合同定义、保险合同计量模型等方面。同时梳理得出新保险合同准则的发布实施有利于凸显“保险姓保”的原则；有利于抑制保险合同美化财务业绩；还有利于提升我国保险行业的国际形象。
The new insurance contract standard came into effect on January 1 this year, marking the further improvement of China’s accounting standards system for enterprises and maintaining continuous convergence with IFRS. Therefore, this paper uses the Python programming language to conduct text mining of the old and new insurance contract guidelines released in 2020 through web crawlers. In addition, download the three old insurance contract guidelines issued by China’s Ministry of Finance in 2006 and 2009, import through local text, use the jieba library for text word segmentation, collection library to count word frequency, wordcloud library to make word clouds, etc. For comparison and contact, and found that new and old insurance contracts appear words such as “contract” and “measurement”. The new insurance standard focuses on the definition of insurance contracts and the measurement model of insurance contracts. At the same time, it is concluded that the issuance and implementation of the new insurance contract standards is conducive to high-lighting the principle of “insurance surname protection”. It is conducive to suppressing insurance contracts and beautifying financial performance; it is also conducive to enhancing the international image of China’s insurance industry.