Título: Aspect Term Extraction Using Deep Learning Structures and Minimal Feature Engineering
Data: 01/11/2019
Horário: 16:30h
Local: Sala de Seminários - Bloco 952
Resumo:
This work proposes a neural network architecture using deep learning structures, and minimal feature engineering, to solve the problem of aspect term extraction in opinionated documents. Aspect term extraction (ATE) is the task of identifying aspects (attributes or characteristics) that have been evaluated in a sentence. The proposed architecture is similar to an Encoder-Decoder used in Neural Machine Translation, that uses an attention mechanism to permit the addition of grammatical relations between words as an additional feature. We also used the Part-of-speech tag (POS tags) as another relevant feature. The proposed architecture obtained state-of-the-art results, with the advantage of using no linguistic rules, only minimal feature engineering.
Banca:
Última atualização (Qua, 30 de Outubro de 2019 14:16)