Predicting dangerous seismic activity with Recurrent Neural Networks

Karol Kurach; Krzysztof Pawlowski

doi:10.15439/2016F134

Predicting dangerous seismic activity with Recurrent Neural Networks

Karol Kurach, Krzysztof Pawlowski

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Scopus citations

Abstract

In this paper we present a solution to the AAIA'16 Data Mining Challenge. The goal of the challenge was to predict, from multivariate time series data, periods of increased seismic activity which may cause life-Threatening accidents in underground coal mines. Our solution is based on Recurrent Neural Network with Long Short-Term Memory cells. It requires almost no feature engineering, which makes it easily applicable to other domains with multivariate time series data. The method achieved the 5th place in the AAIA'16 competition, out of 203 teams.

Original language	English (US)
Title of host publication	Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016
Editors	Maria Ganzha, Marcin Paprzycki, Leszek Maciaszek
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	239-243
Number of pages	5
ISBN (Electronic)	9788360810903
DOIs	https://doi.org/10.15439/2016F134
State	Published - Nov 3 2016
Externally published	Yes
Event	2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016 - Gdansk, Poland Duration: Sep 11 2016 → Sep 14 2016

Publication series

Name	Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016

Conference

Conference	2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016
Country	Poland
City	Gdansk
Period	9/11/16 → 9/14/16

ASJC Scopus subject areas

Computer Science Applications
Information Systems
Computer Science (miscellaneous)

Access to Document

10.15439/2016F134

Fingerprint Dive into the research topics of 'Predicting dangerous seismic activity with Recurrent Neural Networks'. Together they form a unique fingerprint.

Cite this

Kurach, K., & Pawlowski, K. (2016). Predicting dangerous seismic activity with Recurrent Neural Networks. In M. Ganzha, M. Paprzycki, & L. Maciaszek (Eds.), Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016 (pp. 239-243). [7733245] (Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.15439/2016F134

Predicting dangerous seismic activity with Recurrent Neural Networks. / Kurach, Karol; Pawlowski, Krzysztof.

Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016. ed. / Maria Ganzha; Marcin Paprzycki; Leszek Maciaszek. Institute of Electrical and Electronics Engineers Inc., 2016. p. 239-243 7733245 (Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Kurach, K & Pawlowski, K 2016, Predicting dangerous seismic activity with Recurrent Neural Networks. in M Ganzha, M Paprzycki & L Maciaszek (eds), Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016., 7733245, Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016, Institute of Electrical and Electronics Engineers Inc., pp. 239-243, 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016, Gdansk, Poland, 9/11/16. https://doi.org/10.15439/2016F134

Kurach K, Pawlowski K. Predicting dangerous seismic activity with Recurrent Neural Networks. In Ganzha M, Paprzycki M, Maciaszek L, editors, Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 239-243. 7733245. (Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016). https://doi.org/10.15439/2016F134

Kurach, Karol ; Pawlowski, Krzysztof. / Predicting dangerous seismic activity with Recurrent Neural Networks. Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016. editor / Maria Ganzha ; Marcin Paprzycki ; Leszek Maciaszek. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 239-243 (Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016).

@inproceedings{b2eb5014d592416d9d60ad02dd5a191f,

title = "Predicting dangerous seismic activity with Recurrent Neural Networks",

abstract = "In this paper we present a solution to the AAIA'16 Data Mining Challenge. The goal of the challenge was to predict, from multivariate time series data, periods of increased seismic activity which may cause life-Threatening accidents in underground coal mines. Our solution is based on Recurrent Neural Network with Long Short-Term Memory cells. It requires almost no feature engineering, which makes it easily applicable to other domains with multivariate time series data. The method achieved the 5th place in the AAIA'16 competition, out of 203 teams.",

author = "Karol Kurach and Krzysztof Pawlowski",

note = "Publisher Copyright: {\textcopyright} 2016 Polish Information Processing Society. Copyright: Copyright 2017 Elsevier B.V., All rights reserved.; 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016 ; Conference date: 11-09-2016 Through 14-09-2016",

year = "2016",

month = nov,

day = "3",

doi = "10.15439/2016F134",

language = "English (US)",

series = "Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "239--243",

editor = "Maria Ganzha and Marcin Paprzycki and Leszek Maciaszek",

booktitle = "Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016",

}

TY - GEN

T1 - Predicting dangerous seismic activity with Recurrent Neural Networks

AU - Kurach, Karol

AU - Pawlowski, Krzysztof

PY - 2016/11/3

Y1 - 2016/11/3

N2 - In this paper we present a solution to the AAIA'16 Data Mining Challenge. The goal of the challenge was to predict, from multivariate time series data, periods of increased seismic activity which may cause life-Threatening accidents in underground coal mines. Our solution is based on Recurrent Neural Network with Long Short-Term Memory cells. It requires almost no feature engineering, which makes it easily applicable to other domains with multivariate time series data. The method achieved the 5th place in the AAIA'16 competition, out of 203 teams.

AB - In this paper we present a solution to the AAIA'16 Data Mining Challenge. The goal of the challenge was to predict, from multivariate time series data, periods of increased seismic activity which may cause life-Threatening accidents in underground coal mines. Our solution is based on Recurrent Neural Network with Long Short-Term Memory cells. It requires almost no feature engineering, which makes it easily applicable to other domains with multivariate time series data. The method achieved the 5th place in the AAIA'16 competition, out of 203 teams.

UR - http://www.scopus.com/inward/record.url?scp=85007197263&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85007197263&partnerID=8YFLogxK

U2 - 10.15439/2016F134

DO - 10.15439/2016F134

M3 - Conference contribution

AN - SCOPUS:85007197263

T3 - Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016

SP - 239

EP - 243

BT - Proceedings of the 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016

A2 - Ganzha, Maria

A2 - Paprzycki, Marcin

A2 - Maciaszek, Leszek

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2016 Federated Conference on Computer Science and Information Systems, FedCSIS 2016

Y2 - 11 September 2016 through 14 September 2016

ER -