BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Acoustic classification of New World bats based on phylogenetic an
 d ecological constraints on call design - Veronica Zamora-Gutierrez
DTSTART:20140822T160000Z
DTEND:20140822T163000Z
UID:TALK53903@talks.cam.ac.uk
CONTACT:30967
DESCRIPTION:Acoustic techniques are becoming a popular alternative to moni
 tor bat species. However\, monitoring programs depend upon the effective r
 ecognition of species or any other ‘recognizable unit’ set as the moni
 toring target. In spite the evolution of patterns in call design that make
  some bat species recognizable using acoustic features\, acoustic identifi
 cation also imposes several challenges. Here\, we study the problem of ide
 ntifying bat species from echolocation calls in order to build automated a
 coustic monitoring algorithms. The main purpose of this study was to asses
 s the accuracy and utility of two Machine Learning classification techniqu
 es faced with a number of bat call discrimination tasks that reflect phylo
 genetic and ecological constraints on bat calls design. We constructed a r
 eference search-phase echolocation bat calls library for Mexican bat speci
 es and applied Random Forest and Dynamic Time Warping algorithms to classi
 fy 59 species. We selected 4538 full-spectrum reference calls from a set o
 f 21154 collected by the authors\, donated by colleagues and from Echobank
 . We trained both algorithms with a number of hierarchical tasks. First we
  train them to classify calls to Family\, Guild or Genus then to species a
 nd compared general species classification accuracy to a non-hierarchical 
 classification done directly to species.  Results show good general classi
 fication accuracy\, being this study with the highest number of species cl
 assified so far. This suggests there is potential to develop an automated 
 classification system for bat calls in megadiverse countries even if data 
 is limited. These methods can provide key information on distribution and 
 diversity changes through time and will set the foundations for a future n
 ational acoustic monitoring program.
LOCATION:Tea Room\, Department of Zoology
END:VEVENT
END:VCALENDAR