I am Daniel Torregrosa, PhD by Universidad de Alicante (www.ua.es), under the direction of Prof. Juan Antonio Pérez-Ortiz and Full Prof. Mikel L. Forcada. In 2018, I defended my thesis Black-box interactive translation prediction, a novel approach for Interactive translation prediction that does not require a closely coupled corpus-based machine translation system; instead, it can use any bilingual resource (including, but not limited to machine translation) to generate the suggestions. A simple heuristic selects the most appropriate suggestions on a per-keystroke basis, but better results can be obtained by training a neural network model with a few thousand parallel sentences. This approach was released both as an OmegaT plugin (github.com/transducens/forecat-omegat) and a standalone Java/GWT server that offers both a web interface and web services (github.com/transducens/forecat).
Occupied a 6 month position at the International Maritime Organization, where I adapted, deployed and trained users in the usage of CAT tools.
At insight@NUIG, I will take part on the NURS project as a postdoctoral researcher.
Research topics: Interactive Translation Prediction, Computer Assisted Translation, Machine Translation, Natural Language processing.
Interests: Machine Translation for under-resourced languages, Computer Assisted Translation, Multi-modality.