Exploiting ontologies for deep learning: A case for sentiment mining
Stephan Raaijmakers, Christopher A. Brewster · TNO Repository · 2018
We present a practical method for explaining deep learningbased text mining with ontology-based information.Our approach uses the recently proposed OntoSenticNet ontology for sentiment mining, and consists of a composite deep learning classifier for sentiment mining, endowed with an ontology-driven attention module.The attention module analyzes the attention the neural network pays to semantic labels assigned to bigrams in input texts.