Combining multiple correspondence analysis with association rule mining to conduct user-driven product design of wearable devices

• Product differentiation and product selection of wearable devices are addressed,
• Insights on designing wristband, smart watch, and sport watch are offered,
• MCA (multiple correspondence analysis) is used to achieve product differentiation,
• ARM (association rule mining) is adopted to identify attractive design attributes,
• KNN (K nearest neighbor) is conducted to accomplish efficient product selection.

In recent years, the popularity of smart phones has boomed the emergence of wearable devices like wristband, smart watch, and sport watch since these devices are portable to record human body information, synchronize information with smart phones, and conduct real-time monitoring of physical condition. However, a recent survey indicates that near 70% respondents are not interested in buying Apple's new iWatch although the marketplace is full of competing alternatives like Samsung's Gear fit, LG's G watch, and Sony's SW3. In this study, a novel framework combining multiple correspondence analysis (MCA), association rule mining (ARM), with K nearest neighbor (KNN) is proposed to help brand companies address the following issues: (1) using MCA to explore the latent relationships between users' demographic profiles, user perceptions of design attributes, and user preferences for wearable devices, (2) using ARM to identify key design attributes that can best configure a specific alternative to achieve effective product differentiation (positioning), (3) using KNN to accomplish efficient product selection (recommendation). More importantly, hundreds of consumers are surveyed to justify the validity of the presented framework.