English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Text analytics based on supervised machine learning classifiers has showngreat promise in a multitude of domains, but has yet to be applied toSeismology. We test various standard models (Naive Bayes, k-Nearest Neighbors,Support Vector Machines, and Random Forests) on a seismological corpus of 100articles related to the topic of precursory accelerating seismicity, spanningfrom 1988 to 2010. This corpus was labelled in Mignan (2011) with the precursorwhether explained by critical processes (i.e., cascade triggering) or by otherprocesses (such as signature of main fault loading). We investigate rather theclassification process can be automatized to help analyze larger corpora inorder to better understand trends in earthquake predictability research. Wefind that the Naive Bayes model performs best, in agreement with the machinelearning literature for the case of small datasets, with cross-validationaccuracies of 86% for binary classification. For a refined multiclassclassification ('non-critical process' < 'agnostic' < 'critical processassumed' < 'critical process demonstrated'), we obtain up to 78% accuracy.Prediction on a dozen of articles published since 2011 shows however a weakgeneralization with a F1-score of 60%, only slightly better than a randomclassifier, which can be explained by a change of authorship and use ofdifferent terminologies. Yet, the model shows F1-scores greater than 80% forthe two multiclass extremes ('non-critical process' versus 'critical processdemonstrated') while it falls to random classifier results (around 25%) forpapers labelled 'agnostic' or 'critical process assumed'. Those results areencouraging in view of the small size of the corpus and of the high degree ofabstraction of the labelling. Domain knowledge engineering remains essentialbut can be made transparent by an investigation of Naive Bayes keywordposterior probabilities.

journal of seismology

Text Classification of the Precursory Accelerating Seismicity Corpus:  Inference on some Theoretical Trends in Earthquake Predictability Research  from 1988 to 2018