基于Python文本挖掘的消费者对国产彩妆品牌评价的分析
Analysis of Consumers’ Evaluation of Domestic Cosmetics Brand Based on Python Text Mining

通过应用Python Beautifulsoup大数据挖掘方法、Python Wordcloud词频分析方法、以及Python Jieba汉字词库分析方法，本文根据各大平台中消费者的反馈，分析了国产彩妆品牌存在的问题及消费者最看重的消费热点。通过选取定位不同的3个国产彩妆品牌包括橘朵、完美日记、MAOGEPING作为参考。利用大量的评论进行文本挖掘、统计词频、分析消费热点，进而分析每个品牌的特点、目标人群及能够进一步发展的建议。最后，根据词频计算不同消费热点所占的比重，以找出消费者最看重的方向。研究结果可作为参考，以此改进国产彩妆品牌的使用体验，进一步发展本地品牌。
By using the big data mining method of Python beautifulsoup, word frequency analysis method of Python Wordcloud, and the Chinese character analysis method of Python Jieba, this paper an-alyzes the existing problems of domestic cosmetics brands and the most important consumption hot spots for consumers based on the feedback from consumers in various platforms. We select three domestic cosmetics brands with different positioning—Judydoll, Perfect Diary and MAOGEPING as reference. We use a large number of comments for text mining, word frequency statistics, analysis of consumption hot spots, and then analyze the characteristics of each brand, target population and suggestions for further development. Finally, according to word frequen-cy calculation of the proportion of different consumption hot spots, we find out the most im-portant direction of consumers. The research results can be used as a reference to improve the use experience of domestic cosmetics brands and further develop local brands.