隨著科技的發展以及網路購物的普及，線上評論對於電商領域可謂說是越 來越重要，有學者認為評論對於消費者的影響甚至會比賣家提供的資訊還大。 評論不僅可以讓消費者了解到產品的品質，也可以提升消費者對於平台的黏著 度。 過去有文獻探討評論內容的廣度、情緒對於評論幫助性的影響，本研究更 深入探討當消費者閱讀經驗性商品與搜尋性商品評論時，覺得有幫助的評論是 否會不一樣。本篇研究使用文字探勘的手法分析評論的情緒以及評論的廣度， 以及分析消費者閱讀不同產品類別時吸收資訊的方式，進而提供平台營運上的 建議。;Online reviews have become more and more important in Ecommerce industry and are thought to be more helpful than seller generated information. Moreover, reviews have been shown to improve customers’ perception of the product quality, and can also increase customers’ stickiness to the website. With these benefits of review, understanding the important attributes of a helpful review let the Ecommerce be able to identify and promote more informative content, to develop encouraging content, and ultimately increase customers’ stickiness and user satisfaction to the website. Previous researches have analyzed how review breadth and review sentiment affects perceived review helpfulness; moreover, there are also some researches mention the differences of information processing method between search goods and experience when reading reviews. Therefore, the study utilizes text mining technique to analyze how product types affect the perceived review helpfulness. Positive and negative emotion is calculated using VADER, and this study also conduct BERTopics to calculate the topic distribution for each review. Finally, this paper discover the different information processing method between search goods and experience goods, and hence give the suggestion to Ecommerce platform base on the finding.
