Abstract:

Seismic networks around the world produce and process seismographs in large volume and at high velocity for various applications including earthquake early warning, nuclear explosion monitoring, and oil and gas exploration. Intelligent and automated seismic data processing is the key to the success of a seismic network. Challenging data science problems are formed in such a data processing pipeline including seismic signal detection, phase identification, depth estimation, aftershock detection and event discrimination. In this tutorial, we will describe the data flow of a traditional processing pipeline from a data mining perspective. We will describe various open seismic data sources, explain data cleaning and preprocessing techniques, describe offline and online machine learned models for the above-mentioned problems, provide code-snippets and demonstrate software packages for seismic data processing. We will end the tutorial with a discussion on a few open problems and related data processing challenges.

Presenter Bio:

Abdullah Mueen is an Associate Professor in Computer Science at the University of New Mexico. His major interest is in temporal data mining with a focus on compliance, security, and trust applications. He performs NSF and AFRL sponsored research in developing seismic data mining algorithms for applications of national interest. He has been actively publishing in the data mining conferences including KDD, ICDM and SDM, and journals including DMKD and KAIS. He has received runner-up award in the Doctoral Dissertation Contest in KDD 2012. He has won the best paper award in SIGKDD 2012 and the KDD Test-of-Time award in 2022.