Martin Rezk and Laurent Ach from Rakuten Institute of Technology Paris will give a talk on Thursday, October 6, at Inria Turing Building, Salle Thomas Flowers.
Rakuten is a Japanese company and one of the world’s leading Internet service companies providing e-commerce, financial and digital content services. PriceMinister, Viber, Kobo, for instance, are part of Rakuten group.
Laurent Ach: Improving User Experience by Understanding Product Data
Rakuten’s E-commerce market places gather hundreds of millions of products, and their associated online content includes a big amount of raw texts and images. It is a great challenge to improve the quality of the catalogs by better understanding this content, extract structured data, and develop similarity measures. Automatically understanding these data also allows for better recommender systems and simplified user interfaces. At Rakuten Institute of Technology in Paris, we work on several projects using natural language processing and computer vision to improve both the user experience and the background organization of products. We will give an overview of them in this talk.
Martin Rezk: Extracting Semantic Information for e-Commerce
Rakuten e-comerce site, Ichiba, uses a large taxonomy to organize the items it sells. Currently, the taxonomy classes that are relevant in terms of profit generation and difficulty of exploration are being manually extended with data properties. In this talk we will present a scalable approach that aims to automate this process, automatically selecting the relevant and semantically homogenous subtrees in the taxonomy, extracting from semi-structured text in items descriptions a core set of properties and a popular subset of their ranges, then extending the covered range using relational similarities in free text, and finally tagging the items with the new semantic information and exposing them as RDF triples (via OBDA). This work has been accepted for publication in ISWC’16.