UniMatch: A Unified User-Item Matching Framework for the Multi-purpose Merchant Marketing

TL;DR

UniMatch introduces a unified framework for user-item matching that efficiently combines recommendation and targeting tasks, reducing costs and maintaining high performance through a novel bias-corrected loss.

Contribution

It proposes a model-agnostic, joint modeling approach with a new loss function that corrects biases, enabling multi-purpose marketing with a single model.

Findings

Significant performance improvements over state-of-the-art methods

Reduced computational costs and maintenance effort

Effective joint modeling of recommendation and targeting

Abstract

When doing private domain marketing with cloud services, the merchants usually have to purchase different machine learning models for the multiple marketing purposes, leading to a very high cost. We present a unified user-item matching framework to simultaneously conduct item recommendation and user targeting with just one model. We empirically demonstrate that the above concurrent modeling is viable via modeling the user-item interaction matrix with the multinomial distribution, and propose a bidirectional bias-corrected NCE loss for the implementation. The proposed loss function guides the model to learn the user-item joint probability $p(u,i)$ instead of the conditional probability $p(i|u)$ or $p(u|i)$ through correcting both the users and items' biases caused by the in-batch negative sampling. In addition, our framework is model-agnostic enabling a flexible adaptation of different model architectures. Extensive experiments demonstrate that our framework results in significant performance gains in comparison with the state-of-the-art methods, with greatly reduced cost on computing resources and daily maintenance.