Predicting and Integrating Expected Answer Types into a Simple Recurrent Neural Network Model for Answer Sentence Selection
Document title: Predicting and Integrating Expected Answer Types into a Simple Recurrent Neural Network Model for Answer Sentence Selection
Journal: Computación y sistemas
Database:
System number: 000560448
ISSN: 1405-5546
Authors: 1
1
2
Institutions: 1Université Paris-Saclay, Orsay, Paris. Francia
2Université Paris sud - Paris XI, Ile de France. Francia
Year:
Season: Jul-Sep
Volumen: 23
Number: 3
Pages: 665-673
Country: México
Language: Inglés
Document type: Artículo
English abstract Since end-to-end deep learning models have started to replace traditional pipeline architectures of question answering systems, features such as expected answertypes which are based on the question semantics are seldom used explicitly in the models. In this paper, we propose a convolution neural network model to predict these answer types based on question words and a recurrent neural network model to find sentence similarity scores between question and answer sentences. The proposed model outperforms the current state of the art results on an answer sentence selection task in open domain question answering by 1.88% on MAP and 2.96% on MRR scores.
Disciplines: Ciencias de la computación
Keyword: Inteligencia artificial
Keyword: Question answering,
Deep learning,
Answer sentence selection,
Expected answer types,
Sentence similarity,
Artificial intelligence
Full text: Texto completo (Ver HTML) Texto completo (Ver PDF)