Abstract: Highlights•Propose a hybrid data pipeline integrating online and in-store user behaviors.•Present a scalable approach for multi-source behavior data fusion in e-commerce.•Show that incorporating in-store signals enhances online recommendation quality.•Reveal modeling gains enabled by unified cross-channel customer behavior data.•Identify future directions linking data engineering with algorithmic development.
External IDs:dblp:journals/dke/MaTG0PPX0KKA25
Loading