Automatic classification of volcano-seismic signals using ensemble methods.

Mora Stock, Cindy and Bravo, C. (2012) Automatic classification of volcano-seismic signals using ensemble methods. [Poster] In: The Lübeck Retreat, Collaborative Research SFB 574 Volatiles and Fluids in Subduction Zones: Climate Feedback and Trigger Mechanisms for Natural Disasters. , 23.-25.05.2012, Lübeck . The Lübeck Retreat: final colloquium of SFB 574; May 23-25, 2012: program & abstracts. ; p. 20 .

Full text not available from this repository.

Abstract

Volcanoes present different types of seismic activity depending on the source origin, such as tremors,
long-period, and volcano-tectonic events. In Chile, Villarrica and Llaima are two of the most active
volcanoes, constantly presenting seismicity that can be classified into these different types. To obtain
a faster and reliable classification of future activity, pattern recognition ensemble methods were
constructed using neural networks and support vector machines. The method for classification here
presented, will be use to analyze data from two temporary networks in Llaima and Villarrica installed
during Nov. 2009 and Apr. 2011.

Document Type: Conference or Workshop Item (Poster)
Keywords: Geodynamics
Research affiliation: OceanRep > SFB 574
OceanRep > SFB 574 > A2
Kiel University
Related URLs:
Date Deposited: 21 Sep 2012 10:54
Last Modified: 29 May 2013 09:10
URI: http://oceanrep.geomar.de/id/eprint/15072

Actions (login required)

View Item View Item