A preliminary text classification of the precursory accelerating seismicity corpus: inference on some theoretical trends in earthquake predictability research from 1988 to 2018

  • A. MignanEmail author
Original Article


Text analytics based on supervised machine learning has shown great promise in a multitude of domains but has yet to be applied to seismology. We describe some common classifiers (Naïve Bayes, k-Nearest Neighbors, Support Vector Machines, and Random Forests) as well as the standard steps of supervised learning (training, validation of model parameter adjustments, and testing). To illustrate text classification on a seismological corpus, we use a hundred articles related to the topic of precursory accelerating seismicity, spanning from 1988 to 2010. This corpus was labelled by Mignan [Tectonophysics, 2011] with the precursor whether explained by critical processes (i.e., cascade triggering) or by other processes (such as signature of main fault loading). We investigate how the classification process can be automatized to help analyze larger corpora in order to better understand trends in earthquake predictability research. We find that the Naïve Bayes model performs best, in agreement with the machine learning literature for the case of small datasets, with cross-validation accuracies showing the model’s predictive ability for both binary classification (“critical process” or else) and a multiclass classification (“non-critical process,” “agnostic,” “critical process assumed,” “critical process demonstrated”). Prediction on a dozen of articles published since 2011 shows however a weak generalization, which can be explained, in part, by the empirical variance of the small training set. This preliminary study demonstrates the potential of supervised learning to reveal textual patterns in the seismological literature. Manual labelling remains essential but is made transparent by an investigation of Naïve Bayes keyword posterior probabilities.


Machine learning Supervised learning Earthquake precursor Critical phenomena 



I thank Pablo Nieto and Marco Broccardo for discussions on the topic of text classification, as well as reviewer Riccardo Zaccarelli for his valuable comments.

Data and resources

All the corpus articles are available on journal websites. The corpus meta-data and labelling are provided in the supplementary material to this article.

Supplementary material

10950_2019_9833_MOESM1_ESM.docx (24 kb)
ESM 1 (DOCX 24 kb)
10950_2019_9833_MOESM2_ESM.json (172 kb)
ESM 2 (JSON 171 kb)
10950_2019_9833_MOESM3_ESM.json (6 kb)
ESM 3 (JSON 5 kb)
10950_2019_9833_MOESM4_ESM.json (6 kb)
ESM 4 (JSON 5 kb)


  1. Adamaki AK, Roberts RG (2017) Precursory activity before larger events in Greece revealed by aggregated seismicity data. Pure Appl Geophys 174:1331–1343. CrossRefGoogle Scholar
  2. Aggarwal CC (2018) Machine learning for text. Springer Nature, 493 pp.
  3. Bak P, Tang C (1989) Earthquakes as a self-organized critical phenomenon. J Geophys Res 94:15,635–15,637CrossRefGoogle Scholar
  4. Bennet KP, Campbell C (2000) Support vector machines: hype or hallelujah? SIGKDD Explor 2:1–13CrossRefGoogle Scholar
  5. Benoit K (2018) Quantitative analysis of textual data, package 'quanteda', available at (last assessed August 2018)
  6. Blei DM, Ng AY, Jordan MI (2003) Latent Dirichlet allocation. J Mach Learn Res 3:993–1022Google Scholar
  7. Bouchon M, Marsan D (2015) Reply to 'Artificial seismic acceleration'. Nat Geosci 8:83CrossRefGoogle Scholar
  8. Bouchon M, Durand V, Marsan D, Karabulut H, Schmittbuhl J (2013) The long precursory phase of most large interplate earthquakes. Nat Geosci 6:299–302CrossRefGoogle Scholar
  9. Breiman L (2001) Random forests. Mach Learn 45:5–32CrossRefGoogle Scholar
  10. Breiman L, Friedman JH, Olshen RA, Stone CJ (1984) Classification and regression trees. Chapman & Hall/CRC, Taylor & Francis Group 358 ppGoogle Scholar
  11. Bufe CG, Varnes DJ (1993) Predictive modeling of the seismic cycle of the greater San Francisco Bay region. J Geophys Res 98:9,871–9,883CrossRefGoogle Scholar
  12. Christou EV, Karakaisis G, Scordilis E (2016) Time dependent seismicity along the western coast of Canada. Res Geophys 5:5730CrossRefGoogle Scholar
  13. Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20:273–297Google Scholar
  14. Cover TM, Hart PE (1967) Nearest neighbor pattern classification. IEEE Trans Inf Theory IT-13:21–27CrossRefGoogle Scholar
  15. De Santis A, Cianchini G, Di Giovambattista R (2015) Accelerating moment release revisited: examples of application to Italian seismic sequences. Tectonophysics 639:82–98. CrossRefGoogle Scholar
  16. Domingos P, Pazzani M (1997) On the optimality of the simple Bayesian classifier under zero-one loss. Mach Learn 29:103–130CrossRefGoogle Scholar
  17. Felzer KR, Page MT, Michael AJ (2015) Artificial seismic acceleration. Nat Geosci 8:82–83CrossRefGoogle Scholar
  18. Forman G (2008) BNS feature scaling: an improved representation over TF-IDF for SVM text classification, ACM 17th Conf. Info. and Knowl. Management 263-270Google Scholar
  19. Freund Y, Schapire RE (1999) A short introduction to boosting. J Japanese Soc AI 14:771–780Google Scholar
  20. Geller RJ (1997) Earthquake prediction: a critical review. Geophys J Int 131:425–450CrossRefGoogle Scholar
  21. Glez-Peña D, Laurenco A, Lopez-Fernandez H, Reboiro-Jato M, Fdez-Riverola F (2013) Web scraping technologies in an API world. Brief Bioinform 15:788–797CrossRefGoogle Scholar
  22. Grimmer J, Stewart BM (2013) Text as data: the promise and pitfalls of automatic content analysis methods for political texts. Polit Anal 21:267–297. CrossRefGoogle Scholar
  23. Grün B, Hornik K (2017). Topic models, package 'topicmodels', available at (last assessed August 2018)
  24. Guilhem A, Bürgmann R, Freed AM, Tabrez Ali S (2013) Testing the accelerating moment release (AMR) hypothesis in areas of high stress. Geophys J Int 195:785–798. CrossRefGoogle Scholar
  25. Hardebeck JL, Felzer KR, Michael AJ (2008) Improved tests reveal that the accelerating moment release hypothesis is statistically insignificant. J Geophys Res 113:B08310. CrossRefGoogle Scholar
  26. Hechenbichler, K., and K. P. Schliep (2004). Weighted k-nearest-neighbor techniques and ordinal classification. Discussion paper 399, SFB 386, Ludwig-Maximilians University, MunichGoogle Scholar
  27. Hough S (2010) Predicting the unpredictable: the tumultuous science of earthquake prediction. Princeton University Press 272 ppGoogle Scholar
  28. Huang H, Meng L (2018) Slow unlocking processes preceding the 2015 Mw 8.4 Illapel, Chile, earthquake. Geophys Res Lett 45:3914–3922. CrossRefGoogle Scholar
  29. Jain AK (2010) Data clustering: 50 years beyond K-means. Pattern Recogn Lett 31:651–666. CrossRefGoogle Scholar
  30. Jiang C, Wu Z (2012) Insights into the long-to-intermediate-term pre-shock accelerating moment release (AMR) from the March 11, 2011, off the Pacific coast of Tohoku, Japan, M9 earthquake. Earth Planets Space 64:765–769CrossRefGoogle Scholar
  31. Jiang C, Wu Z (2013) Intermediate-term medium-range precursory accelerating seismicity prior to the 12 May 2008, Wenchuan earthquake. Pure Appl Geophys 170:209–219. CrossRefGoogle Scholar
  32. Joachims T (1998) Text categorization with support vector machines: learning with many relevant features. Mach Learn ECML-98:137–142Google Scholar
  33. Karakaisis GF, Parazachos CB, Scordilis EM (2013) Recent reliable observations and improved tests on synthetic catalogs with spatiotemporal clustering verify precursory decelerating-accelerating seismicity. J Seismol 17:1063–1072. CrossRefGoogle Scholar
  34. Karatzoglou A, Smola A, Hornik K, Zeileis A (2004) kernlab—an S4 package for Kernelt methods in R. J Stat Softw 11:1–20CrossRefGoogle Scholar
  35. Kazemian J, Hatami MR (2017) Temporal variations of seismic parameters in Tehran region. Pure Appl Geophys 174:3841–3852. CrossRefGoogle Scholar
  36. Kharde VA, Sonawane SS (2016) Sentiment analysis of Twitter data: a survey of techniques. Int J Comput Appl 139:5–15Google Scholar
  37. King GCP (1983) The accommodation of large strains in the upper lithosphere of the earth and other solids by self-similar fault systems: the geometrical origin of b-value. Pure Appl Geophys 121:761–815CrossRefGoogle Scholar
  38. Kohavi R (1995) A study of cross-validation and bootstrap for accuracy estimation and model selection. IJCAI'95 Proceed 14th Int Joint Conf AI 2:1137–1143Google Scholar
  39. Kuhn T (1970) The structure of scientific revolutions, enlarged. In: International encyclopedia of unified science, 2nd edn. The University of Chicago Press 210 ppGoogle Scholar
  40. Lagios E, Papadimitriou P, Novali F, Sakkas V, Fumagalli A, Vlachou K, Del Conte S (2012) Combined seismicity pattern analysis, DGPS and PSInSAR studies in the broader area of Cephalonia (Greece). Tectonophysics 524-525:43–58. CrossRefGoogle Scholar
  41. Liaw A, Wiener M (2018). Breiman and Cutler's random forests for classification and regression, package 'randomForest', available at (last assessed August 2018)
  42. Mignan A (2011) Retrospective on the accelerating seismic release (ASR) hypothesis: controversy and new horizons. Tectonophysics 505:1–16. CrossRefGoogle Scholar
  43. Mignan A (2012) Seismicity precursors to large earthquakes unified in a stress accumulation framework. Geophys Res Lett 39:L21308.
  44. Mignan A (2014) The debate on the prognostic value of earthquake foreshocks: a meta-analysis. Sci Rep 4:4099. CrossRefGoogle Scholar
  45. Mignan A (2015) Modeling aftershocks as a stretched exponential relaxation. Geophys Res Lett 42:9726–9732. CrossRefGoogle Scholar
  46. Mignan A, King GCP, Bowman D (2007) A mathematical formulation of accelerating moment release based on the stress accumulation model. J Geophys Res 112:B07308. CrossRefGoogle Scholar
  47. Mouselimis L (2018). Kernel k nearest neighbors, package 'KernelKnn', available at (last assessed August 2018)
  48. Ng AY, Jordan MI (2001) On discriminative vs. generative classifiers: a comparison of logistic regression and naive Bayes. Adv Neural Inf Proces Syst 14:605–610Google Scholar
  49. Ng S-K, Wong M (1999) Toward routine automatic pathway discovery from on-line scientific text abstracts. Genome Inform 10:104–112Google Scholar
  50. Ogata Y (1988) Statistical models for earthquake occurrences and residual analysis for point processes. J Am Stat Assoc 83:9–27CrossRefGoogle Scholar
  51. Papadopoulos GA (1988) Long-term accelerating foreshock activity may indicate the occurrence time of a strong shock in the Western Hellenic Arc. Tectonophysics 152:179–192CrossRefGoogle Scholar
  52. Papazachos BC, Karakaisis GF, Papazachos CB, Scordilis EM (2007) Evaluation of the results for an intermediate-term prediction of the 8 January 2006 Mw 6.9 Cythera earthquake in the southwestern Aegean. Bull Seismol Soc Am 97:347–352. CrossRefGoogle Scholar
  53. Pearce D, Rantala V (1983) New foundations for metascience. Synthese 56:1–26Google Scholar
  54. Pliakis D, Papakostas T, Vallianatos F (2012) A first principles approach to understand the physics of precursory accelerating seismicity. Ann Geophys 55:165–170. Google Scholar
  55. Rokach L (2010) Ensemble-based classifiers. Artif Intell Rev 33:1–39.
  56. Rumelhart DE, Hinton GE, Williams RJ (1986) Learning representations by back-propagation errors. Nature 323:533–536CrossRefGoogle Scholar
  57. Salton G, McGill M (eds) (1983) Introduction to modern information retrieval. McGraw-HillGoogle Scholar
  58. Sammis CG, Sornette D (2002) Positive feedback, memory, and the predictability of earthquakes. PNAS 99:2501–2508. CrossRefGoogle Scholar
  59. Sebastiani F (2002) Machine learning in automated text categorization. ACM Comput Surv 34:1–47CrossRefGoogle Scholar
  60. Seif S, Mignan A, Zechar JD, Werner MJ, Wiemer S (2017) Estimating ETAS: the effects of truncation, missing data, and model assumptions. J Geophys Res Solid Earth 122:449–469. CrossRefGoogle Scholar
  61. Seif S, Zechar JD, Mignan A, Nandan S, Wiemer S (2018) Foreshocks and their potential deviation from general seismicity. Bull Seismol Soc Am 109:1–18. CrossRefGoogle Scholar
  62. Sornette D (2000) Critical phenomena in natural sciences, chaos, fractal, self-organization and disorder: concepts and tools. Springer 434 ppGoogle Scholar
  63. Steinwart I, Christmann A (2008) Support vector machines, information science and statistics. Springer 601 ppGoogle Scholar
  64. Tsytsarau M, Palpanas T (2012) Survey on mining subjective data on the web. Data Lin Knowl Disc 24:478–514. CrossRefGoogle Scholar
  65. Welbers K, Van Atteveldt W, Benoit K (2017) Text analysis in R. Commun Methods Meas 11:245–265. CrossRefGoogle Scholar

Copyright information

© Springer Nature B.V. 2019

Authors and Affiliations

  1. 1.Swiss Federal Institute of Technology Zurich, ETHZZurichSwitzerland

Personalised recommendations