ABSTRACT The current product design not only takes into account the function and reliability, but also concerns about the affective aspects in order to meet the consumers’ emotional needs. However, there is always a gap between affective intentions of manufacturers and affective responses of consumers. Traditional methods rely on manual surveys to understand the gap, which is costly, time-consuming and in a small scale. In this paper, we propose a text mining method to extract affective intentions and affective responses from the online product description and consumer reviews. We build an affective profile for each product to visualise the gap between affective responses and affective intentions of the product. To evaluate the effectiveness of the proposed method, a case study is conducted based on the public data from Amazon.com. We construct affective profiles for selected products and analyze affective gaps. We also evaluate the usefulness of the extracted affective information in product recommendations. The results showed that the gap between consumer's affective responses and manufacturer's affective intentions can be identified and visualised, which may help manufacturers to improve their products and services. Affective information is also useful for product recommendations.

中文翻译：

---

从非结构化文本中挖掘产品的情感反应和情感意图

摘要 当前的产品设计不仅兼顾功能和可靠性，更注重情感方面，以满足消费者的情感需求。然而，制造商的情感意图与消费者的情感反应之间总是存在差距。传统方法依靠人工调查来了解差距，成本高、耗时长且规模小。在本文中，我们提出了一种文本挖掘方法来从在线产品描述和消费者评论中提取情感意图和情感反应。我们为每个产品建立一个情感档案，以可视化产品的情感反应和情感意图之间的差距。为了评估所提出方法的有效性，基于 Amazon.com 的公共数据进行了案例研究。我们为选定的产品构建情感档案并分析情感差距。我们还评估了提取的情感信息在产品推荐中的有用性。结果表明，消费者的情感反应和制造商的情感意图之间的差距可以被识别和可视化，这可能有助于制造商改进他们的产品和服务。情感信息对于产品推荐也很有用。