Certain species of bark beetle, such as the mountain pine beetle, have a devastating effect on forests across North America. They have been called “the greatest tree killers known” and their economic impact is measured in the billions. Not all bark beetles are destructive, though, and research is currently being done on using their acoustic signals to identify their species.
Python provides an excellent platform for exploring this costly ecological problem. This talk will look at:
- how libraries such as LibROSA can be used to analyze the audio recordings;
- how numpy, scipy and pandas can be used to extract features;
- and how machine learning algorithms from scikit-learn can then be applied to build a classifier.