This paper introduces a novel method aimed at enhancing online-to-offline (O2O) services recommendations by utilizing two-layer knowledge networks. The primary objective of this method is to assist consumers in efficiently navigating the myriad of options available when choosing O2O services. Using co-occurrence relationships, we construct a two-layer knowledge network system, comprising a service knowledge network based on service usage information as the first layer and a consumer knowledge network, built on “co-used” behaviors as the second layer. The former is established upon service use data, while the latter is founded on “co-used” behaviors among consumers. The features and information of these two knowledge networks can complement each other to produce precise and effective recommendations. Empirical findings gained from our experiments demonstrate that: (1) the proposed recommendation method outperforms widely-used and state-of-the-art recommendation methods; (2) both the service knowledge network and consumer knowledge network play an equally significant role in O2O service recommendations; (3) the location of O2O services is an essential factor in consumers’ choices for services. Notably, this research also identifies the optimal parameter settings for the proposed recommendation method.
Read full abstract