A novel user-generated content-driven and Kano model focused framework to explore the impact mechanism of continuance intention to use mobile APPs

User-generated content (UGC), which generates vast amounts of content in real-time through social networks, offers a significant opportunity for mining new knowledge. The survival of information technology products such as mobile APPs (mAPPs) depends on continuance. The exploration of the impact mechanism underlying continuance intention is crucial given the low continued usage of many mAPPs. Mining real-world UGC provides an efficient approach for user experience and product evaluation compared to traditional interviews or surveys. This study proposes a novel UGC-driven, Kano model focused, pipelined framework to automatically identify impact factors and determine the impact mechanism underlying continuance intention. The above method framework involves unsupervised clustering text analysis to identify user needs and product functions from UGC, followed by the construction of a statistical model to explore the relationship between these factors and user satisfaction and dissatisfaction (Kano model). Additionally, a model to distinguish the attributes of factors that significantly affect satisfaction or dissatisfaction is proposed. Finally, user satisfaction and dissatisfaction are used as mediators to build models of continuance and discontinuance intention. Empirical validation of the proposed method is conducted with a case study of mobile health apps, involving the mining of 86,423 user reviews and structural equation modeling based on 1,025 user responses. The results indicate that the UGC-driven method effectively explores the impact mechanism of continuance and discontinuance intention.