Application of Musical Pattern Recognition and Machine Learning to Geophysical Data Analysis and Seismic Facies Classification – free download
Free download paper.
We developed a novel interpretation approach combining audio-video display of geophysical data with musical pattern recognition techniques and musical information retrieval methods.
This novel approach allows multisensory perception of seismic data (and of other geophysical data) and, at the same time, efficient automatic classification.
Method and tests on real data
After accurate conversion from geophysical formats (example: SEG-Y) to musical formats (example: Musical Instrument Digital Interface, or briefly MIDI), we extract musical features from the converted data. These can be single-valued attributes, such as pitch and sound intensity, or multi-valued attributes, such as pitch histograms, melodic, harmonic and rhythmic patterns.
Using a real data set, we show that these musical features can be diagnostic for seismic facies classification in a complex exploration area. They can be complementary with respect to “conventional” seismic attributes. Using a supervised machine learning approach based on the k-Nearest Neighbors algorithm and on Automatic Neural Networks, we classify three gas-bearing channels. The good performance of our classification approach is confirmed by borehole data available in the same area.
Applications
Our approach is addressed to multiple tasks, including:
1) Seismic facies clustering and classification
2) Seismic data interpretation supported by Machine Learning approaches
3) Multimodal analysis of geophysical data (in time-frequency domain and by audio-visual display)
4) Multimodal analysis of well log data (interactive as well as automatic)
5) Big data mining
6) Earthquake data analysis …
Additional applications
The same approach can be exported and applied to every type of data series and for many purposes outside the geoscience domain, including:
a) Medical data and diagnosis
b) Analysis of financial trends
c) Business data analysis
d) Study of social trends …
Free download paper
Authors: Paolo Dell’Aversana, Gianluca Gabbriellini, Alfonso Iunio Marini, Alfonso Amendola
Open Access Journal: AIMS Geosciences, 2 (4): 401–413 - DOI: 10.3934/geosci.2016.4.401.
