Identification of time-evolving product opportunities via social media mining

摘要

This study suggests a social media mining approach for identifying time-evolving product opportunities for product improvement, customer complaint management, and competitive intelligence via event detection and tracking (EDT) and sentiment analysis. This approach applies an aging theory-based EDT algorithm to online product reviews to detect the creation of customer-stated events as groups of similar reviews and track their growth and extinction. Next, the approach uses sentiment analysis and an opportunity algorithm to evaluate time-evolving events and provides quantified clues regarding product-development directions based on events with high opportunity. To show the workings of the proposed approach, a case study involving smart speakers is presented. We expect that the approach will contribute to identifying time-evolving product opportunities from large-scale social-media data in addition to enabling the real-time monitoring of customer needs in rapidly evolving product environments.