Italian Event Detection Goes Deep Learning

Tommaso Caselli

Italian Event Detection Goes Deep Learning

Tommaso Caselli^*

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

1 Citation (Scopus)

339 Downloads (Pure)

Abstract

This paper reports on a set of experiments with different word embed- dings to initialize a state-of-the-art Bi- LSTM-CRF network for event detection and classification in Italian, following the EVENTI evaluation exercise. The net- work obtains a new state-of-the-art result by improving the F1 score for detection of 1.3 points, and of 6.5 points for classifica- tion, by using a single step approach. The results also provide further evidence that embeddings have a major impact on the performance of such architectures.

Original language	English
Title of host publication	Proceedings of the Fifth Italian Conference on Computational Linguistics (CLiC-it 2018)
Publisher	CEUR Workshop Proceedings (CEUR-WS.org)
Number of pages	6
Volume	2253
ISBN (Electronic)	1613-0073
Publication status	Published - 2018

Keywords

event extraction
deep learning

Access to Document

Italian Event Detection Goes Deep LearningFinal publisher's version, 287 KBLicence: Unspecified

Handle.net

Persistent link

Cite this

@inproceedings{24fbb4209475415c91cd8e31426e40f0,

title = "Italian Event Detection Goes Deep Learning",

abstract = "This paper reports on a set of experiments with different word embed- dings to initialize a state-of-the-art Bi- LSTM-CRF network for event detection and classification in Italian, following the EVENTI evaluation exercise. The net- work obtains a new state-of-the-art result by improving the F1 score for detection of 1.3 points, and of 6.5 points for classifica- tion, by using a single step approach. The results also provide further evidence that embeddings have a major impact on the performance of such architectures.",

keywords = "event extraction, deep learning",

author = "Tommaso Caselli",

year = "2018",

language = "English",

volume = "2253",

booktitle = "Proceedings of the Fifth Italian Conference on Computational Linguistics (CLiC-it 2018)",

publisher = "CEUR Workshop Proceedings (CEUR-WS.org)",

}

TY - GEN

T1 - Italian Event Detection Goes Deep Learning

AU - Caselli, Tommaso

PY - 2018

Y1 - 2018

N2 - This paper reports on a set of experiments with different word embed- dings to initialize a state-of-the-art Bi- LSTM-CRF network for event detection and classification in Italian, following the EVENTI evaluation exercise. The net- work obtains a new state-of-the-art result by improving the F1 score for detection of 1.3 points, and of 6.5 points for classifica- tion, by using a single step approach. The results also provide further evidence that embeddings have a major impact on the performance of such architectures.

AB - This paper reports on a set of experiments with different word embed- dings to initialize a state-of-the-art Bi- LSTM-CRF network for event detection and classification in Italian, following the EVENTI evaluation exercise. The net- work obtains a new state-of-the-art result by improving the F1 score for detection of 1.3 points, and of 6.5 points for classifica- tion, by using a single step approach. The results also provide further evidence that embeddings have a major impact on the performance of such architectures.

KW - event extraction

KW - deep learning

M3 - Conference contribution

VL - 2253

BT - Proceedings of the Fifth Italian Conference on Computational Linguistics (CLiC-it 2018)

PB - CEUR Workshop Proceedings (CEUR-WS.org)

ER -

Italian Event Detection Goes Deep Learning

Abstract

Keywords

Access to Document

Handle.net

Fingerprint

Cite this