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.

Characterizing and Predicting Early Reviewers for Effective Product Marketing on E-Commerce Websites

“…In this paper, we use the Amazon [7], [22] and Yelp 4 datasets. Amazon dataset originally contains 142.8 million product reviews ranging from May 1996 to July 2014 and Yelp dataset contains 4.7 million product reviews ranging from July 2004 to January 2017.…”

Efficient Tensor Strategy for Recommendation

“…[30] in his Probabilistic polyadic factorization and its application to personalized recommendation demonstrated the effectiveness of directly modelling all the dimensions simultaneously in a unified framework. These among other works presupposes that, tensor decomposition models performed well in terms of prediction efficiency and effectiveness compared to the various matrix factorization algorithms, in particular application to massive data processing [31]- [33]. However, the numerous literature concerning the subject.…”

Large-scale Opinion Relation Extraction with Distantly Supervised Neural Network

“…We extract opinion relations on a subset of Amazon product review corpus provided by (McAuley et al, 2015), which contains 15 domains and 33 million reviews. The statistics of extracted relations are in Table 2.…”