Categorization of B2B Service Offers: Lessons learnt from the Silex Use case

Molka Dhouib 1 Catherine Faron Zucker 2 Andrea Tettamanzi 2
2 WIMMICS - Web-Instrumented Man-Machine Interactions, Communities and Semantics
CRISAM - Inria Sophia Antipolis - Méditerranée , Laboratoire I3S - SPARKS - Scalable and Pervasive softwARe and Knowledge Systems
Abstract : In the domain of Information Retrieval and Natural Language Processing, text classification has become a crucial task. In this article, we share our experience of text cate-gorization in an industrial context and we present a comparative evaluation of binary and multi-label classification algorithms applied to texts describing service offers, in the SILEX B2B platform. We show that for some use cases like the one we consider, a traditional representation of texts by "bags of words" gives better classification results than the promising representation by "word embeddings".
Document type :
Conference papers
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01830905
Contributor : Sandra Bringay <>
Submitted on : Thursday, July 5, 2018 - 2:29:07 PM
Last modification on : Friday, April 5, 2019 - 9:56:04 AM
Document(s) archivé(s) le : Monday, October 1, 2018 - 5:54:37 PM

File

APIA_2018_paper_15.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01830905, version 1

Collections

Citation

Molka Dhouib, Catherine Faron Zucker, Andrea Tettamanzi. Categorization of B2B Service Offers: Lessons learnt from the Silex Use case. 4ème conférence sur les Applications Pratiques de l'Intelligence Artificielle APIA2018, Jul 2018, Nancy, France. ⟨hal-01830905⟩

Share

Metrics

Record views

136

Files downloads

66