Occupancy data

Dataset used to run the occupancy models described in <em>Lydia K.D. Katsis, Tessa A. Rhinehart, Elizabeth Dorgay, Emma E. Sanchez, C. Patrick Doncaster, Jake L. Snaddon, Justin Kitzes. A comparison of statistical methods for deriving occupancy estimates from acoustic monitoring data.</em>
<br>
Data consists of
<strong>Ground-truth</strong>
1) ground_truth_data.csv: csv of manually reviewed files for presence or absence of howler monkeys, used as benchmark to compare modelled occupany estimates.
<br>
<strong>Partially annotated datasets</strong>
1) top_ten_annotated.csv : files with top-10 machine learning scores per site manually reviewed
2) random_ten_annotated.csv : 10 randomly selected files per site manually reviewed
3) scheduled_listening_annotated.csv: first file after 5am every 3 days from each site manually reviewed.
<br>
<strong>Machine learning predictions on</strong> <strong>temporally subset dataset</strong>
1) dawn_all.csv : full dataset of predictions, with no subsetting
2) dawn_10_max.csv : temporal interval of 10 minutes, with file with maximum machine learning score sampled
3) dawn_10_random.csv: temporal interval of 10 minutes, first file sampled
4) dawn_30_max.csv: temporal interval of 30 minutes, with file with maximum machine learning score sampled
5) dawn_30_random.csv: temporal interval of 30 minutes, first file sampled
<br>
<br>
Columns in data consist of:

LocationID/site: unique identifier for recorder location
date/timestamp: when the recording was made
annotation: manualy review for presence (1) or absence (0) of howler monkey in clip
logit_present: machine learning score with logit transformation 
softmax_present: machine learning score with softmax transformation 

<br>