2024-03-29T07:06:27Zhttps://repositori.uji.es/oai/requestoai:repositori.uji.es:10234/988632019-11-18T13:05:09Zcom_10234_7038com_10234_9col_10234_98487
00925njm 22002777a 4500
dc
Soleymani, Ali
author
Van Loon, E. Emiel
author
Robert, Weibel
author
2014-06
Recent advances in tracking technologies provide an unprecedented opportunity for a better understanding of animal movement. Data from multiple sensors can be used to capture crucial factors deriving the behaviors of the animal. Typically, accelerometer data is used to describe and classify fine-grained behaviors, while GPS data are rather used to identify more large-scale mobility patterns. In this study, however, the main research question was to what extent fine-grained foraging behaviors of wading birds can be classified from GPS tracking data alone. The species used in this study was the Eurasian Oystercatcher, Haematopus ostralegus. First, a supervised classification approach is employed based on parameters extracted from accelerometer data to identify and label different behavioral categories. Then, we seek to establish how movement parameters, computed from GPS trajectories, can identify the previously labeled behaviors. A decision tree was developed to see which movement features specifically contribute to predicting foraging. The methods used in this study suggest that it is possible to extract, with high accuracy, fine-grained behaviors based on high-resolution GPS data, providing an opportunity to build a prediction model in cases where no additional sensor or observational data on behavior is available. The key to success, however, is a careful selection of the movement features used in the classification process, including cross-scale analysis.
9789081696043
http://hdl.handle.net/10234/98863
Association of Geographic Information Laboratories for
Europe ( AGILE) Conference
Geographic Information Science
Información geográfica
Movement analysis
GPS
accelerometer
foraging behavior
movement parameters
classification
Capability of movementfeatures extracted fromGPS trajectoriesforthe classification of fine‐grained behaviors