Train SizeTest SizeLength Number of ClassesNumber of DimensionsType
25000 250000 10200AUDIO
Data Source: Link Here
Donated By: UCR, UEA
Description:

The InsectWingbeat data was generated by the UCR computational entomology group and used [1].

The original data is a reconstruction of the sound of insects passing through a sensor. The data in the archive is the power spectrum of the sound. The 10 classes are male and female mosquitos (Ae. aegypti, Cx. tarsalis, Cx. quinquefasciants, Cx. stigmatosoma), two types of flies (Musca domestica and Drosophila simulans) and other insects. A spectorgram of each 1 second sound segment was created with a window length of 0.061 seconds and an overlap of 70%. Each instance in this multivariate dataset is arranged such that each dimension is a frequency band from the spectrogram. Each of the 10 classes in this dataset consist of 5,000 instances.

  • Chen Y, Why A, Batista G, Mafra-Neto A, Keogh E. Flying Insect Classification with Inexpensive Sensors. Journal of Insect Behaviour, Sep 1;27(5):657-77. 2014.
  • Download this dataset
    Dataset Image

    Best Algorithm:
    Best Accuracy: