A Study Of The Differences In The Impact Of Personalized Recommendation Types On Consumer Purchase Decisions

With the booming development of Internet technology and e-commerce,online shopping has become an important part of consumers’ daily life,while also making them face an information overload dilemma.At this time,personalized recommendation systems have emerged and are widely used in major e-commerce platforms to provide consumers with personalized product or service recommendations and help them make purchase decisions.However,inappropriate personalized recommendations can reduce consumers’ experience and even make them psychologically resistant.Therefore,how to make personalized recommendations for consumers scientifically and effectively has become a common concern for academia and industry.Previous research on this issue has mostly focused on improving recommendation algorithms from a technical perspective,but few empirical analyses have been conducted from the perspective of consumer behavior,and there is a lack of research comparing the impact of different types of personalized recommendations on consumer behavior,and the mechanism of the role of consumer characteristics and product characteristics in it is yet to be discussed.This study uses the SOR model as the theoretical basis and applies it to the field of personalized recommendation to explore the differences in the influence of two recommendation types,content-based recommendation and collaborative filtering recommendation,on consumers’ purchase decisions,as well as the mediating role of consumers’ perceived trust and perceived value.Also,considering the boundary conditions,consumer knowledge and product type are added as moderating variables in the model to explore whether the differences in the impact of personalized recommendation types on consumers’ purchase decisions change under different contexts.To this end,two virtual scenarios were used to obtain consumer data,and SPSS data analysis software was used to fit the model,test the research hypotheses,and analyze the differences in the influence of different personalized recommendation types on consumers’ purchase decisions and the underlying mechanisms.It was found that there are significant differences in the impact of personalized recommendation types on consumers’ perceived trust,perceived value,and purchase intention,with collaborative filtering recommendation having a greater positive impact than content-based recommendation,and with consumers’ perceived trust and perceived value playing a fully mediating role.Consumer knowledge plays a moderating role in personalized recommendation type and perceived value.Specifically,the higher(lower)the level of consumer knowledge,the smaller(larger)the difference in perceived value between personalized recommendation types,i.e.,the perceived value advantage of collaborative filtering recommendation will be weakened(enhanced).However,this moderating effect does not hold in perceived trust.Product type plays a moderating role in the impact of personalized recommendation types on both perceived value and perceived trust of consumers,with content-based recommendation matched with search products and collaborative filtering recommendation matched with experience products giving consumers higher perceived trust and perceived value.This study enriches the research in the field of personalized recommendation to a certain extent,and also provides implications for ecommerce platforms to optimize personalized recommendation strategies.