Title: Smart watch unboxing video viewing frequency prediction by support vector machine and Bayesian probit analysis
Authors: Nai-Hua Chen - Chienkuo Technology University (Taiwan) [presenting]
Abstract: The quick changing technology produces make innovation products filled in the marketplace to compete customers intention. The unboxing articles reveal through blogs describing details of new purchased products. Due to the development of media techniques, unboxing process are shared in video type via Youtube. It provides viewers an unboxing-like experience while watching and becomes more popular. These videos help industries to explosure their newly launched products. Some researchers have devoted in modeling the consumer innovation adoption behavior. Viewers behavior towards unboxing videos is associated with their preferences. It can adopt to understand whether customers attitude towards the new products. Several purchase models are developed to understand customer dynamic behavior. The support vector machine (SVM) has gained popularity in visual pattern recognition recently. The negative binomial distribution (NBD) contains two characteristics which are Poisson purchasing and gamma heterogeneity is shown as a robust method in product purchase frequency. We proposed a two-stage innovation adoption model. The first is to classify unboxing video viewers into like and unlike. Next, the Bayesian probit analysis is used to compare effects that influence like and unlike behavior.