Language-Agnostic Representation Learning for Product Search on E-Commerce Platforms

Ahuja, Aman; Rao, Nikhil; Katariya, Sumeet; Subbian, Karthik; Reddy, Chandan K.

doi:10.1145/3336191.3371852

Cited by 22 publications

(12 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first experiment is on a public multilingual Amazon review dataset (Prettenhofer and Stein, 2010). In addition, we conduct experiments on a real-world industrial multilingual search relevance dataset (Ahuja et al, 2020) used for E-commerce product search.…”

Section: Methodsmentioning

confidence: 99%

“…Baselines Additionally, we compare with the start-of-the-art baseline LAPS (Ahuja et al, 2020), which relies on external task-specific cross-lingual parallel data (Ahuja et al, 2020), i.e., productto-product and query-to-query correspondences among all 5 languages.…”

Section: Cross-lingual Relevance Classificationmentioning

confidence: 99%

“…On the contrary, MGL resorts to learning how to adapt across languages from each CLT task. Empirically, extensive experiments on both the public multilingual Amazon review dataset (Prettenhofer and Stein, 2010) and the real-world industrial multilingual search relevance dataset (Ahuja et al, 2020) demonstrate the effectiveness of the MGL method.…”

Section: Introductionmentioning

confidence: 97%

See 2 more Smart Citations

Learn to Cross-lingual Transfer with Meta Graph Learning Across Heterogeneous Languages

Kumar

Headden

et al. 2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

The recent emergence of multilingual pretraining language model (mPLM) has enabled breakthroughs on various downstream crosslingual transfer (CLT) tasks. However, mPLMbased methods usually involve two problems:(1) simply fine-tuning may not adapt generalpurpose multilingual representations to be task-aware on low-resource languages; (2) ignore how cross-lingual adaptation happens for downstream tasks. To address the issues, we propose a meta graph learning (MGL) method. Unlike prior works that transfer from scratch, MGL can learn to cross-lingual transfer by extracting meta-knowledge from historical CLT experiences (tasks), making mPLM insensitive to low-resource languages. Besides, for each CLT task, MGL formulates its transfer process as information propagation over a dynamic graph, where the geometric structure can automatically capture intrinsic language relationships to guide cross-lingual transfer explicitly. Empirically, extensive experiments on both public and real-world datasets demonstrate the effectiveness of the MGL method.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Cross-lingual Relevance Classificationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 97%

See 1 more Smart Citation

Learn to Cross-lingual Transfer with Meta Graph Learning Across Heterogeneous Languages

Kumar

Headden

et al. 2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

show abstract

“…SBERT is a modification of the pretrained BERT network based on a twin network to derive semantically meaningful sentence embeddings that can be compared via cosine similarly. In e-commerce, product retrieval plays a key role and is a well-researched topic (Lu et al;Ahuja et al, 2020). E-commerce systems typically perform two steps: a retrieval and a ranking step.…”

Section: Related Workmentioning

confidence: 99%

Combining semantic search and twin product classification for recognition of purchasable items in voice shopping

Le¹,

Weber²,

Bradford³

2021

Proceedings of the 4th Workshop on E-Commerce and NLP

View full text Add to dashboard Cite

The accuracy of an online shopping system via voice commands is particularly important and may have a great impact on customer trust. This paper focuses on the problem of detecting if an utterance contains actual and purchasable products, thus referring to a shopping-related intent in a typical Spoken Language Understanding architecture consisting of an intent classifier and a slot detector. Searching through billions of products to check if a detected slot is a purchasable item is prohibitively expensive. To overcome this problem, we present a framework that (1) uses a retrieval module that returns the most relevant products with respect to the detected slot, and (2) combines it with a twin network that decides if the detected slot is indeed a purchasable item or not. Through various experiments, we show that this architecture outperforms a typical slot detector approach, with a gain of +81% in accuracy and +41% in F1 score.

show abstract

“…Recently, semantic matching models (Huang et al, 2013;Pang et al, 2016) have been adopted to improve retrieval performance (Mitra et al, 2018). These models are trained using click/purchase logs and typically separated by countries (Ahuja et al, 2020). However, such per-country specific training schema exposes three major drawbacks.…”

Section: Introductionmentioning

confidence: 99%

Graph-based Multilingual Product Retrieval in E-Commerce Search

Lu¹,

Hu²,

Zhao³

et al. 2021

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

View full text Add to dashboard Cite

Nowadays, with many e-commerce platforms conducting global business, e-commerce search systems are required to handle product retrieval under multilingual scenarios. Moreover, comparing with maintaining per-country specific e-commerce search systems, having a universal system across countries can further reduce the operational and computational costs, and facilitate business expansion to new countries. In this paper, we introduce a universal end-to-end multilingual retrieval system, and discuss our learnings and technical details when training and deploying the system to serve billion-scale product retrieval for e-commerce search. In particular, we propose a multilingual graph attention based retrieval network by leveraging recent advances in transformer-based multilingual language models and graph neural network architectures to capture the interactions between search queries and items in e-commerce search. Offline experiments on five countries data show that our algorithm outperforms the state-of-the-art baselines by 35% recall and 25% mAP on average. Moreover, the proposed model shows significant increase of conversion/revenue in online A/B experiments and has been deployed in production for multiple countries.

show abstract

Language-Agnostic Representation Learning for Product Search on E-Commerce Platforms

Cited by 22 publications

References 22 publications

Learn to Cross-lingual Transfer with Meta Graph Learning Across Heterogeneous Languages

Learn to Cross-lingual Transfer with Meta Graph Learning Across Heterogeneous Languages

Combining semantic search and twin product classification for recognition of purchasable items in voice shopping

Graph-based Multilingual Product Retrieval in E-Commerce Search

Contact Info

Product

Resources

About