%0 Journal Article
%T Enhancing Domain-Specific Supervised Natural Language Intent Classification with a Top-Down Selective Ensemble Model
%A Barbara McGillivray
%A Gard B. Jenset
%J -
%D 2019
%R https://doi.org/10.3390/make1020037
%X Abstract Natural Language Understanding (NLU) systems are essential components in many industry conversational artificial intelligence applications. There are strong incentives to develop a good NLU capability in such systems, both to improve the user experience and in the case of regulated industries for compliance reasons. We report on a series of experiments comparing the effects of optimizing word embeddings versus implementing a multi-classifier ensemble approach and conclude that in our case, only the latter approach leads to significant improvements. The study provides a high-level primer for developing NLU systems in regulated domains, as well as providing a specific baseline accuracy for evaluating NLU systems for financial guidance. View Full-Tex
%K natural language understanding
%K dialogue systems
%K multi-classifier
%K word embeddings
%K domain adaptation
%K conversational artificial intelligence
%K financial domain
%K long short-term memory
%U https://www.mdpi.com/2504-4990/1/2/37