KDD Cup 2023 Workshop:

Multilingual Session Recommendation Challenge

Held in conjunction with KDD'23 Aug 6, 2023 - Aug 10, 2023, Long Beach, CA, USA


INTRODUCTION

The objective of this workshop is to discuss the winning submissions of the KDDCup 2023 Challenge on Multilingual Session Recommendation Challenge. In this challenge, we introduce the “Amazon Multilingual Multi-locale Session Dataset (Amazon-M2)”, a collection of anonymized customer sessions containing products from six different locales: English, German, Japanese, French, Italian, and Spanish, published with the aim of encouraging the development of multilingual recommendation systems, which can enhance personalization and understanding of global trends and preferences. For each session, the dataset provides a list of product IDs (Amazon ASIN numbers) interacted by the current user, together with the additional product features. More details of this challenge are available here: [Challenge Link], [Dataset Paper Link] and [OpenReview Link].


SCHEDULE

August 9, 2023, 1:00PM–5:00PM (PST), Long Beach, CA, USA.

204, Long Beach Convention & Entertainment Center

This will be a hybrid session. Zoom link will be provided later.

  Opening
  1:00-1:30 PM

Introduction by organizers.  
Moderator: Dr. Wei Jin, Assistant Professor at Emory University
Speaker: Dr. Xianfeng Tang, Senior Applied Scientist at Amazon

   Invited Speaker
  1:30-2:10 PM

    Description of image Dr. Haixun Wang
    VP Engineering, Algorithms & Distinguished Scientist at Instacart

    Title: LLMs and Generative AI for E-Commerce
    Abstract: In this talk, I will discuss the current status and challenges of product search. In particular, I will highlight the significant effort it takes to create a high-quality product search engine using classical information retrieval methods. Then, I will discuss how recent advances in arge pre-trained language models, may change the status quo. While embedding-based retrieval has the potential to improve classical information retrieval methods, creating a machine learning-based, end-to-end system for general-purpose web search is still extremely difficult. Nevertheless, I will argue that product search for e-commerce may prove to be an area where LLMs can create the first disruption to classical information retrieval systems.

   Invited Speaker
  2:10-2:50 PM

    Description of image Dr. Jian Pei
    Arthur S. Pearse Distinguished Professor at Duke University

    Title: Data and AI Markets for E-Commerce
    Abstract: In the ever-evolving landscape of e-commerce, seamless collaboration between multiple parties is essential to offer comprehensive and end-to-end services transcending supplier, provider, and market sector boundaries. A critical aspect of such cooperation lies in the sharing of data and AI models among these parties. However, data and AI model sharing pose significant challenges. In this enlightening talk, I will unveil insightful strategies and blueprints for establishing data and AI markets tailored to the e-commerce domain. These markets will facilitate efficient data and AI model sharing, discovery, and integration. I will delve into the latest technical advancements while addressing the potential hurdles and opportunities.

   Poster Session
  4:35-5:00 PM


ACCEPTED PAPERS

In total, we have accepted 15 submissions including 6 oral presentations and 8 poster presentationss. The accepted submission are publicly available at OpenReview page: https://openreview.net/group?id=KDD.org/2023/Workshop/Cup
Nvidia: Winning Amazon KDD Cup'23
  Task 1   Task 2   Task 3
Authors: Chris Deotte, Kazuki onodera, Jean-Francois Puget, Benedikt Schifferer, Gilberto Titericz

A Stacking and Transfer Learning with Diverse Similarity For Building Multilingual Session-based Recommendation Systems
  Task 1   Task 2
Authors: Jiangwei Luo, Zhouzhou He, Ye Tang, Wentao Tang, Li Cheng

A Hybrid Approach of Statistics and Embeddings for Multilingual and Multi-Locale Recommendation
  Task 2
Authors: Weijia Zhang, Jin Zhan, Zhongshan Huang, Lu Wang, Qiang Wang

Statistical and Generative Models with Subtitle Extraction for Next Product Title Generation
  Task 3
Authors: Honghee Lee, Youngjae Chang, Kyuri Choi, Jongwuk Lee, Youngjoong Ko.

Practical Content-aware Session-based Recommendation: Deep Retrieve then Shallow Rank
  Task 1
Authors: Yuxuan Lei, Xiaolong Chen, Defu Lian, Peiyan Zhang, Jianxun Lian, Chaozhuo Li, Xing Xie.

Co-visitation Meets Token Alignment for Next Product Title Generation
  Task 3
Authors: Chenwang Wu, Leyan Deng, Zhihao Zhu, Defu Lian

Next Product Title Generation in E-commerce: Rule-based Methods and Autoencoder Model
Authors: Junya Miyamoto, Shinnosuke Hirano, Soma Makino, Kazuya Uekado, Xinyi Lao.

Candidate Reranking Solution using Many Variant Features for KDDCup2023, Team YAMALEX Solution
Authors: Hiroki Yamamoto, Takashi Sasaki, Shin Higuchi, Tomonori Fujiwara, Shun Yoshioka.

6th Place Solution to Amazon KDD Cup '23 Task2: Next Product Recommendation for Underrepresented Languages
Authors: Chiaki Ichimura, Daiki Toda.
Language-Enhanced Session-Based Recommendation with Decoupled Contrastive Learning
Authors: Zhipeng Zhang, Piao Tong, YINGWEI MA, Qiao Liu, Xujiang Liu, Xu Luo.

Prediction of next product title by simple deletion of the last word
Authors: Meisaku Suzuki, Ryo Koyama, Shin Ishiguro, Shunya Imada, Yuya Kunimoto.

A completely locale-independent session-based recommender system by leveraging trained model
Authors: Yu Tokutake, Chihiro Yamasaki, Yongzhi Jin, Ayuka Inoue, Kei Harada.

Word Filtering Approach for Next Product Title Generation
Authors: Yusuke Fukushima, Keiichi Ochiai, Yoshitaka Inoue, Masato Hashimoto, Sho Maeoki.

A Two-stage Ranking Framework for Multilingual Recommendation(Team:AIDA)
Authors: Shijie Liu, Chenliang Zhang, Xutao Han, Wen Wu, Wei Zhang.


SUBMISSION GUIDELINES

The objective of this workshop is to discuss the winning submissions of the Submissions to the Amazon KDD Cup 2023 is single-blind (author names and affiliations should be listed). Everyone in the Top-10 leaderboard submissions will have a guaranteed opportunity for an in-person oral/poster presentation. Other submissions will be evaluated by a committee based on their novelty and insights. The deadline for the submissions is July 20, 2023 July 23, 2023 (Anywhere on Earth time). Accepted submissions will be notified latest by August 1, 2023. Please note that the KDD Cup workshop will have no proceedings and the authors retain full rights to submit or post the paper at any other venue.

Link to the submission website: https://openreview.net/group?id=KDD.org/2023/Workshop/Cup

Submissions describing 1 or 2 tasks are limited to a maximum of 4 pages, including all content and references, and must be in PDF format. However, submissions describing 3 tasks can have up to 6 pages. Please use ACM Conference templates (two column format). One recommended setting for Latex file is: \documentclass[sigconf, review]{acmart}. Template guidelines are here: https://www.acm.org/publications/proceedings-template.

In addition, authors can provide an optional one page supplement at the end of their submitted paper (it needs to be in the same PDF file) focused on reproducibility. After the submission deadline, the names and order of authors cannot be changed.

It would be great if you could cite our dataset paper available at ArXiv.
@article{jin2023amazon,
title={Amazon-M2: A Multilingual Multi-locale Shopping Session Dataset for Recommendation and Text Generation},
author={Wei Jin and Haitao Mao and Zheng Li and Haoming Jiang and Chen Luo and Hongzhi Wen and Haoyu Han and Hanqing Lu and Zhengyang Wang and Ruirui Li and Zhen Li and Monica Xiao Cheng and Rahul Goutam and Haiyang Zhang and Karthik Subbian and Suhang Wang and Yizhou Sun and Jiliang Tang and Bing Yin and Xianfeng Tang},
journal={arXiv preprint arXiv:2307.09688},
year={2023},
}
If you have any questions, please concat us at jinwei2@msu.edu and xianft@amazon.com.


DATA

The data and its license is available at the following link.
https://www.aicrowd.com/challenges/amazon-kdd-cup-23-multilingual-recommendation-challenge
If you plan to use this dataset for your own research, please cite this paper.

@article{jin2023amazon,
title={Amazon-M2: A Multilingual Multi-locale Shopping Session Dataset for Recommendation and Text Generation},
author={Wei Jin and Haitao Mao and Zheng Li and Haoming Jiang and Chen Luo and Hongzhi Wen and Haoyu Han and Hanqing Lu and Zhengyang Wang and Ruirui Li and Zhen Li and Monica Xiao Cheng and Rahul Goutam and Haiyang Zhang and Karthik Subbian and Suhang Wang and Yizhou Sun and Jiliang Tang and Bing Yin and Xianfeng Tang},
journal={arXiv preprint arXiv:2307.09688},
year={2023},
}