EXPRESS: Modeling Categorized Consumer Collections with Interlocked Hypergraph Neural Networks

Consumers curate collections of items for various reasons and categorize them into subsets or categories based on different criteria as their collections grow. The items in a collection reflect a consumer’s preferences and the categories provide insights into the different contexts in which items are consumed. We develop a novel deep generative modeling framework that captures the network structure of consumer collections using multiple interlocked hypergraphs. Our model employs message-passing variational auto-encoders that leverage hypergraph structures and entity-specific covariates to generate probabilistic deep embeddings for consumers, items and item categories. Applying this framework to digital music collections and playlists of music consumers, we demonstrate that our model outperforms several sophisticated benchmarks in predicting linkages within these collections. We then illustrate how our approach enables firms to generate novel personalized product bundles, recommend relevant items and bundles, and dynamically expand existing bundles with new items. Beyond the music application, our method is broadly applicable to other consumer collections, such as food recipes and content collections on social curation platforms like Pinterest.