Inferring Networks of Substitutable and Complementary Products

Abstract

In a modern recommender system, it is important to understand how products relate to each other. For example, while a user is looking for mobile phones, it might make sense to recommend other phones, but once they buy a phone, we might instead want to recommend batteries, cases, or chargers. These two types of recommendations are referred to as substitutes and complements: substitutes are products that can be purchased instead of each other, while complements are products that can be purchased in addition to each other. Here we develop a method to infer networks of substitutable and complementary products. We formulate this as a supervised link prediction task, where we learn the semantics of substitutes and complements from data associated with products. The primary source of data we use is the text of product reviews, though our method also makes use of features such as ratings, specifications, prices, and brands. Methodologically, we build topic models that are trained to automatically discover topics from text that are successful at predicting and explaining such relationships. Experimentally, we evaluate our system on the Amazon product catalog, a large dataset consisting of 9 million products, 237 million links, and 144 million reviews.

12 pages, 6 figures

References in corpus (1)

• Supervised Topic Models

Cited by in corpus (14)

• Joint Deep Modeling of Users and Items Using Reviews for Recommendation
• Don't Classify, Translate: Multi-Level E-Commerce Product Categorization Via Machine Translation
• DenseAlert: Incremental Dense-Subtensor Detection in Tensor Streams
• CER: Complementary Entity Recognition via Knowledge Expansion on Large Unlabeled Product Reviews
• Evaluating Social Networks Using Task-Focused Network Inference
• Mining Compatible/Incompatible Entities from Question and Answering via Yes/No Answer Classification using Distant Label Expansion
• Supervised Complementary Entity Recognition with Augmented Key-value Pairs of Knowledge
• KeystoneML: Optimizing Pipelines for Large-Scale Advanced Analytics
• "Are you sure?": Preliminary Insights from Scaling Product Comparisons to Multiple Shops
• Recommending Complementary Products in E-Commerce Push Notifications with a Mixture Model Approach
• Sampling Online Social Networks by Random Walk with Indirect Jumps
• Augmenting the User-Item Graph with Textual Similarity Models
• Dual Attention Network for Product Compatibility and Function Satisfiability Analysis
• A Novel Large-scale Ordinal Regression Model