Merchandise Recommendation for Retail Events with Word Embedding Weighted Tf-idf and Dynamic Query Expansion

TL;DR

This paper proposes a merchandise recommendation method for retail events that uses word embedding similarity, an enhanced tf-idf formula, and dynamic query expansion to improve item retrieval accuracy.

Contribution

It introduces a novel approach combining word embedding-based keyword expansion with a weighted tf-idf scheme for better retail merchandise recommendations during seasonal events.

Findings

Improved retrieval accuracy demonstrated in experiments.

Effective keyword expansion using word embeddings.

Enhanced tf-idf formula boosts search relevance.

Abstract

To recommend relevant merchandises for seasonal retail events, we rely on item retrieval from marketplace inventory. With feedback to expand query scope, we discuss keyword expansion candidate selection using word embedding similarity, and an enhanced tf-idf formula for expanded words in search ranking.