| Description: | The problem is to classify an insect as one of three species of fruit fly, based on the reconstructed audio of its wingbeat.
The time series represent the change in aplitude of a infra-red light as is occcluded by the wings of a fruit fly during flight.
The database contains wav recordings from the same optical sensor inserted in-turn into six insectary boxes containing only one fly species of both sexes. As thy fly through the sensor their wings partially occludes the light from the transmitter to the receiver. The light fluctuation recorded follows the rhythm of the wingbeat.
Recordings are sampled at 8,000Hz.
Correspondence should be addressed to Ilyas Potamitis:
potamitis@staff.teicrete.gr
Instances: 34,518 Time series length: 5,000 Classes:
0 Melanogaster (6,064)
1 suzukii (10,142)
2 zaprionus (18,312)
default train test split created through a random partition. |
