Presentacionshttp://hdl.handle.net/2117/864552024-03-29T00:13:34Z2024-03-29T00:13:34ZEnglish-to-Hindi system description for WMT 2014: deep source-context features for MosesRuiz Costa-Jussà, MartaGupta, ParthBanchs, Rafael E.Rosso, P.http://hdl.handle.net/2117/1023482022-02-20T10:53:43Z2017-03-10T19:06:41ZEnglish-to-Hindi system description for WMT 2014: deep source-context features for Moses
Ruiz Costa-Jussà, Marta; Gupta, Parth; Banchs, Rafael E.; Rosso, P.
This paper describes the IPN-UPV participation on the English-to-Hindi translation task from WMT 2014 International Evaluation Campaign. The system presented is based on Moses and enhanced with deep learning by meansof asource-context feature function. This feature depends on the input sentence to translate, which makes it more challenging to adapt it into the Moses framework. This work reports the experimental details of the system putting special emphasison: how thefeaturefunction is integrated in Moses and how the deep learning representations are trained and used.
2017-03-10T19:06:41ZRuiz Costa-Jussà, MartaGupta, ParthBanchs, Rafael E.Rosso, P.This paper describes the IPN-UPV participation on the English-to-Hindi translation task from WMT 2014 International Evaluation Campaign. The system presented is based on Moses and enhanced with deep learning by meansof asource-context feature function. This feature depends on the input sentence to translate, which makes it more challenging to adapt it into the Moses framework. This work reports the experimental details of the system putting special emphasison: how thefeaturefunction is integrated in Moses and how the deep learning representations are trained and used.Looking for efficient and accurate ways of computing the global ionospheric electron density distribution from huge amounts of GNSS observationsHernández Pajares, ManuelJuan Zornoza, José MiguelSanz Subirana, JaumeMonte Moreno, EnriqueAragón Ángel, María Ángeleshttp://hdl.handle.net/2117/864532023-02-19T08:09:25Z2016-05-02T08:39:12ZLooking for efficient and accurate ways of computing the global ionospheric electron density distribution from huge amounts of GNSS observations
Hernández Pajares, Manuel; Juan Zornoza, José Miguel; Sanz Subirana, Jaume; Monte Moreno, Enrique; Aragón Ángel, María Ángeles
In this work the authors will explore different potential ways of estimating efficiently and accurately the global
number density of ionospheric free electrons from the most part of nowadays available GNSS measurements, taken from ground based GPS receivers (IGS network) and LEO on-board GPS receivers (such as FORMOSAT-3/COSMIC constellation).
It is basically designed as a bootstrapping approach, from a first determination of VTEC global maps based on
the ground data, to a final electron density extrapolation process aided by simple first-principle conditions, and
passing by an optimal error decorrelation treatment in the VTEC interpolation and corresponding application to
improve the inversion of the GPS occultation measurements.
The performances against external reference data, including dual frequency altimeters and ionosonde measurements, will be also shown to support the conclusions in different Solar Cycle conditions.
2016-05-02T08:39:12ZHernández Pajares, ManuelJuan Zornoza, José MiguelSanz Subirana, JaumeMonte Moreno, EnriqueAragón Ángel, María ÁngelesIn this work the authors will explore different potential ways of estimating efficiently and accurately the global
number density of ionospheric free electrons from the most part of nowadays available GNSS measurements, taken from ground based GPS receivers (IGS network) and LEO on-board GPS receivers (such as FORMOSAT-3/COSMIC constellation).
It is basically designed as a bootstrapping approach, from a first determination of VTEC global maps based on
the ground data, to a final electron density extrapolation process aided by simple first-principle conditions, and
passing by an optimal error decorrelation treatment in the VTEC interpolation and corresponding application to
improve the inversion of the GPS occultation measurements.
The performances against external reference data, including dual frequency altimeters and ionosonde measurements, will be also shown to support the conclusions in different Solar Cycle conditions.