Retour aux projets

18882.1 PFES-ES

18882.1 PFES-ES

Résumé

Query Expansion: Deep Learning and Crowdsourcing

Textes supplémentaires

Abstract

The relevance of a vertical job search engine's results is key to its success, both in terms of sales and user experience. The challenge is to map the user query intent to the underlying specialized (and often evolving) vocabulary of the documents. In this project we will develop a new Query Expansion engine to create domain-specific word similarity models using deep learning techniques; in addition, we will use crowdsourcing to continuously test and improve these models.

Participants

Maxime Droux

Wirtschaftspartner

maxime@neuvoo.ca

Neuvoo Switzerland Sàrl Maxime Droux EPFL Innovation Park, Bat. C 1015 Lausanne VD CH

Philippe Cudré-Mauroux

Kontaktperson

phil@exascale.info

Université de Fribourg eXascale Infolab Philippe Cudré-Mauroux Bd de Pérolles 90 1700 Fribourg FR CH

Détails du projet

Office de recherche
INNOSUISSE
Statut du projet
Abgeschlossen
Type de projet
Forschungsprojekt
Date de début
2016-08-31
Date de fin
2018-04-30
Coûts totaux approuvés
CHF 230245.00
Section
21 Förderbereich Enabling Sciences
Domaine politique NABS
Industrielle Produktivität und Technologie

Disciplines de recherche

  • T180 Telecommunication engineering 100.0%