DEFTri: A Few-Shot Label Fused Contextual Representation Learning For Product Defect Triage in e-Commerce

TL;DR

This paper introduces DEFTri, a novel few-shot learning framework that leverages fine-tuned BERT with label-fused text embeddings to improve automated defect triage in e-commerce, addressing data scarcity and domain-specific challenges.

Contribution

It proposes a new framework combining label-fused embeddings and few-shot learning for defect triage, along with a proprietary dataset for multi-label classification.

Findings

Enhanced contextual defect representations with BERT.

Effective few-shot defect classification on proprietary dataset.

Improved accuracy over baseline methods.

Abstract

Defect Triage is a time-sensitive and critical process in a large-scale agile software development lifecycle for e-commerce. Inefficiencies arising from human and process dependencies in this domain have motivated research in automated approaches using machine learning to accurately assign defects to qualified teams. This work proposes a novel framework for automated defect triage (DEFTri) using fine-tuned state-of-the-art pre-trained BERT on labels fused text embeddings to improve contextual representations from human-generated product defects. For our multi-label text classification defect triage task, we also introduce a Walmart proprietary dataset of product defects using weak supervision and adversarial learning, in a few-shot setting.