品牌微博情绪的分类研究

Branding Emotion Classification in Micro Blogging Environment

品牌微博是企业与用户沟通和交流的平台,也是企业引导消费者品牌情绪的端口。基于三维情感模型,通过人工编码的方法获取品牌微博情绪坐标数据,应用K-means聚类方法将其归结为轻松型、乐观型、感触型和互动型四种类型。研究发现,轻松型品牌微博能吸引较多的点赞,互动型品牌微博更能引导用户参与评论。除转发抽奖等内容外,品牌微博的转发数相对稳定。因此,轻松型和乐观型微博更能巩固新老用户,互动型微博更能增大品牌信息的扩散范围。为不同营销重点的品牌微博运营提供了参考。

Brand micro blogging is not only the communication platform between enterprises and users,but also a brand emotion port where enterprises can guide micro blogging consumers. Based on PAD theory,brand micro blogging emotional coordinates are obtained through artificial coding method,and the emotion is classified into four types by K-means clustering algorithm,that is,relaxing,optimistic,feeling and interactive sentiment. It is found that the brand micro blogging with relaxing or interactive sentiment is easier to attract fans than that with optimistic or feeling sentiment. Among these,relaxing brand micro blogging can attract more likes,and interactive brand micro blogging can guide more users to participate in the comments. Brand micro blogging forwarding number is relatively stable except lottery. According to these results,the relaxing and optimistic micro blogging can consolidate both new and old customers better and the interactive micro blogging can enlarge the transmission range of brand information. The results have some implications for the different aims of brand micro blogging marketing operation.