Datasets:
tags:
- time series
- time series classification
- monst
- audio
license: other
pretty_name: InsectSound
size_categories:
- 10K<n<100K
Part of MONSTER: https://arxiv.org/abs/2502.15122.
| InsectSound | |
|---|---|
| Category | Audio |
| Num. Examples | 50,000 |
| Num. Channels | 1 |
| Length | 600 |
| Sampling Freq. | 6 kHz |
| Num. Classes | 10 |
| License | Public Domain |
| Citations | [1] [2] |
FruitFlies, taken from the broader UCR archive, consistst of 34,518 (univariate) time series, each of length 5,000, representing acoustic recordings of wingbeats for three species of fruit fly [1, 2]. The recordings are single channel with a sampling rate of 8 kHz (i.e., each recording represents just over half a second of data). The recordings are made using a specialised infrared sensor which detects the vibrations of the wings of the insects. The learning task is to identify the species of fly based on the recordings. This version of the dataset has been split into stratified random cross-validation folds.
[1] Yanping Chen, Adena Why, Gustavo Batista, Agenor Mafra-Neto, and Eamonn Keogh. (2014). Flying insect classification with inexpensive sensors. Journal of Insect Behavior, 27(5):657–677.
[2] Yanping Chen. Flying insect classification with inexpensive sensors. https://sites.google.com/site/insectclassification/ (via Internet Archive), 2014. Public Domain.