Abstract:
Since the outbreak of the COVID-19 pandemic in 2020, China has adopted a zero-clearing policy under closed control. It is rather common for residents who are quarantined at home to buy fresh agricultural products online, when COVID-19 spread in big cities. Many e-commerce platforms are trying to develop online shopping channels for fresh agricultural products. However, negative comments and news about those platforms have been increasing because of several reasons, such as the difference in the quality of fresh products, inadequate categories of commodity and inefficient delivery caused by the shortage of personnel and so on. The smooth daily supply of online fresh agricultural products is conducive to soothing the pessimistic emotions and to encouraging their active obedience to epidemic prevention and control policy. Therefore, it is of great importance to explore the preference characteristics of consumers\' online purchase of fresh agricultural products under this critical situation. In this paper, firstly, Pycharm software is used to collect online comment texts of fresh agricultural products on the online platforms with a total of 34,546 pieces of evaluation data. Secondly, the collected data is preformed into the text preprocessing. To be specific, the obtained online comments are processed by Python, including the process of text duplication between sentences, text duplication within sentences and short sentence filtering. After that, processed texts are subjected to Jieba Text Segmentation to form the final word frequency ranking, involving two procedures, part-of-speech tagging and stop-words removal. Lastly, the results of the LDA model indicate the factors that influence consumers\' preferences when they purchase fresh agricultural products online. This study could not only identify the typical features of residents\' online shopping preference in the context of the spread of COVID-19, but also provide pragmatic suggestions for the local government to appease the residents\' negative emotions for the prevention of widespread complaints at the social level.
摘要:
自2020年COVID-19大流行爆发以来,中国采取了封闭控制下的零清零政策。在家隔离的居民在网上购买新鲜农产品是相当普遍的,当COVID-19在大城市传播时。很多电商平台都在尝试开发生鲜农产品的网购渠道。然而,由于几个原因,关于这些平台的负面评论和新闻一直在增加,例如新鲜产品质量的差异,商品类别不足,人员短缺等造成的交货效率低下。线上鲜活农产品每日供应顺畅,有利于舒缓悲观情绪,有利于鼓励其积极服从疫情防控政策。因此,在这种严峻形势下,探索消费者网上购买生鲜农产品的偏好特征具有重要意义。在本文中,首先,使用Pycharm软件在网络平台上收集鲜活农产品的在线评论文本,共34546条评价数据。其次,对采集到的数据进行文本预处理。具体而言,获得的在线评论由Python处理,包括句子之间文本重复的过程,句子和短句子过滤中的文本重复。之后,对处理后的文本进行解巴文本分割,形成最终的词频排名,涉及两个程序,词性标记和停止词删除。最后,LDA模型的结果表明了影响消费者在线购买新鲜农产品时偏好的因素。本研究不仅可以识别COVID-19传播背景下居民网络购物偏好的典型特征,而且可以为当地政府安抚居民的负面情绪提供务实的建议,以防止社会层面的广泛投诉。