Deep Learning Methods in Sentiment Analysis

Position: Leader
Funding: approximately 49.000,00 €
Duration: 3/2014 – 12/2015
Funding organisation: Subcontract from MSENSIS S.A.
Programme: PAVET 2013, Co-funded by the European Regional Development Fund (ERDF) and the Greek Secretariat of Research and Technology (GSRT)
Project URL:​,​​
Contact person: Konstantinos Diamantaras

The project ​goals are: (a) the analysis and implementation of state-of-the-art Deep Learning methods and Machine Learning methods, in general, for the sentiment analysis/opinion mining from unstructured (free) text. The target languages are Greek and English; (b) the ​optimum parametrization ​of the models, ​the ​vectorial representation of the ​words and the general pre-processing​ of​ the documents to be analyzed.