A Multistep Strategy for Analyzing Consumer Generated Data
Are you already subscribed?
Login to check
whether this content is already included on your personal or institutional subscription.
Abstract
The significant amount of on-line Users Generated Contents (Ugcs) represents an inexhaustible mine of useful information for companies. If written about specific transactions or commercial experiences, such contents constitute the so-called electronic word of mouth (eWom), able to influence the purchase intentions of other potential buyers. When expressed in natural language, eWom can be analyzed for a variety of purposes. In this study, a multistep strategy of analysis is proposed, in order to highlight the multiplicity of useful information for companies that can be drawn from eWom content, as well as the potential of Text mining techniques. To meet these objectives, about 850,000 reviews were collected on best performing 320 products belonging to the 32 macro-categories present on the e-commerce Amazon.com platform. This large corpus was the source for extracting customers’ sentiment, and assessing the levels of their satisfaction regarding products. Furthermore, it was useful to evaluate the effectiveness of the reviews according to their polarity, and identify the salient aspects around which eWom revolve on the e-commerce platform
Keywords
- eWom
- big data
- sentiment analysis
- customer satisfaction
- machine learning