Machine Learning Based Seismic Region Classification

Классификация сейсмических регионов на основе машинного обучения
Samuel da Silva Oliveira, Anne M. P. Canuto, Bruno Motta de Carvalho, Márcio Kreutz
2020-07-01

LSTM networksensemble methodsmachine learningseismic region classificationseismic signals
It has become increasingly common in academic and industrial environments the necessity to process huge amounts of seismic signals. Several researchers have been seeking for ways to improve and optimize the processing of these enormous amounts of data that are related to routine demands of geophysicists. One of these demands is the classification of distinct seismic regions captured by the same seismograph, a task that could take up to months of manual data processing. In this paper, we propose the usage of machine learning techniques to the task of the classification of seismic regions, in order to achieve accurate results with better performance and speed. The algorithms K-NN, MLP, Naive Bayes and Decision Tree were used for tests as base classifiers and also combined on ensemble methods. We also employed Deep Learning techniques, namely, a pure RNN network, and a variation of RNN called LSTM. The best results were achieved when using heterogeneous classifiers, showing accuracy rates of up to 98.52%. The results show that one can build an efficient seismic region classification system even when few classified data are already available for a specific seismograph setting.
1
Heterogeneous classifier ensembles achieved the best performance, reaching accuracy rates of up to 98.52%.
2
K-NN, MLP, Naive Bayes, and Decision Tree classifiers were evaluated individually and within ensemble methods, alongside RNN and LSTM deep-learning models.
3
Machine learning was applied to classify distinct seismic regions recorded by the same seismograph, addressing a task that can require months of manual processing.
4
The results indicate that efficient seismic-region classification is feasible even with few labeled data for a specific seismograph setting.

Seismic signals and regions recorded by the same seismograph

Accurate, efficient, and fast classification of distinct seismic regions using machine-learning and deep-learning classifiers, including under limited labeled data

Publication Details
Publication Date
2020-07-01
Journal
Publisher
ISSN
Access Type
Author Information
Authors
Samuel da Silva Oliveira
Anne M. P. Canuto
Bruno Motta de Carvalho
Márcio Kreutz
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%