Export Publication

Gonçalves, S., Cortez, P. & Moro, S. (2018).  A deep learning approach for sentence classification of scientific abstracts . In V. Kurkova et al. (Ed.), Artificial Neural Networks and Machine Learning – ICANN 2018. (pp. 479-488).  Island of Rhodes, Greece: Springer.

S. Gonçalves et al.,  " A deep learning approach for sentence classification of scientific abstracts ", in Artificial Neural Networks and Machine Learning – ICANN 2018, V. Kurkova et al., Ed.,  Island of Rhodes, Greece, Springer, 2018, pp. 479-488

@inproceedings{gonçalves2018_1765835890208,
	author = "Gonçalves, S. and Cortez, P. and Moro, S.",
	title = " A deep learning approach for sentence classification of scientific abstracts ",
	booktitle = "Artificial Neural Networks and Machine Learning – ICANN 2018",
	year = "2018",
	editor = "V. Kurkova et al.",
	volume = "",
	number = "",
	series = "",
	doi = "10.1007/978-3-030-01424-7_47",
	pages = "479-488",
	publisher = "Springer",
	address = " Island of Rhodes, Greece",
	organization = "European Neural Network Society",
	url = "https://link.springer.com/chapter/10.1007%2F978-3-030-01424-7_47"
}

TY  - CPAPER
TI  -  A deep learning approach for sentence classification of scientific abstracts 
T2  - Artificial Neural Networks and Machine Learning – ICANN 2018
AU  - Gonçalves, S.
AU  - Cortez, P.
AU  - Moro, S.
PY  - 2018
SP  - 479-488
SN  - 0302-9743
DO  - 10.1007/978-3-030-01424-7_47
CY  -  Island of Rhodes, Greece
UR  - https://link.springer.com/chapter/10.1007%2F978-3-030-01424-7_47
AB  - The classification of abstract sentences is a valuable tool to support scientific database querying, to summarize relevant literature works and to assist in the writing of new abstracts. This study proposes a novel deep learning approach based on a convolutional layer and a bi-directional gated recurrent unit to classify sentences of abstracts. The proposed neural network was tested on a sample of 20 thousand abstracts from the biomedical domain. Competitive results were achieved, with weight-averaged precision, recall and F1-score values around 91%, which are higher when compared to a state-of-the-art neural network.
ER  -