Learning Representations from Product Titles for Modeling Shopping Transactions

TL;DR

This paper introduces BASTEXT, a text-based model for shopping transaction analysis that effectively learns product representations from titles, addressing cold-start issues and improving next product recommendation and keyword search.

Contribution

BASTEXT is a novel, efficient model that leverages product titles to learn representations, outperforming existing methods in recommendation and search tasks.

Findings

BASTEXT outperforms state-of-the-art recommendation methods.

The model efficiently handles millions of baskets.

It serves as a strong baseline for keyword-based product search.

Abstract

Shopping transaction analysis is important for understanding the shopping behaviors of customers. Existing models such as association rules are poor at modeling products that have short purchase histories and cannot be applied to new products (the cold-start problem). In this paper, we propose BASTEXT, an efficient model of shopping baskets and the texts associated with the products (e.g., product titles). The model's goal is to learn the product representations from the textual contents to capture the relationships between the products in the baskets. Given the products already in a basket, a classifier identifies whether a potential product is relevant to the basket based on their vector representations. This relevancy enables us to learn high-quality representations of the products. The experiments demonstrate that BASTEXT can efficiently model millions of baskets and that it outperforms the state-of-the-art methods in the next product recommendation task. We also show that BASTEXT is a strong baseline for keyword-based product search.