Scientists Pradipto Das and Yiu-Chang Lin of Rakuten Institute of Technology (RIT) Boston attended the 2019 ACM SIGIR ICTIR Conference in the heart of Silicon Valley at Santa Clara University from October 2-5. They presented their paper, “A Dataset and Baselines for e-Commerce Product Categorization” at the News & Entities session.
“A Dataset and Baselines for e-Commerce Product Categorization” was written in collaboration with Andrew Trotman of University of Otago and Surya Kallumadi of The Home Depot. Their research paper analyzes a data collection of a million product titles from 3,008 categories from the Rakuten.com product catalog. Twenty six independent teams from academia and industry participated in an automatic product categorization challenge on this specific dataset. In so doing, they analyze the results of their findings as well as suggest baselines for further research in the area of taxonomy categorization of product listings for e-commerce platforms. This includes better classifiers, modeling over a taxonomy, as well as minimally-supervised attribute extraction from product titles.
The ACM SIGIR International Conference on the Theory of Information Retrieval (ICTIR) is a forum for the “presentation and discussion of research related to the foundational aspects of Information Retrieval (IR), including, for example, human search processes; search and recommendation; learning and optimization for IR, language and representations for IR; sensory information related to IR; artificial intelligence, ethics and responsibility in information retrieval.”
We want to thank ACM SIGIR, the sponsors, and all attendees of the conference.