English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

1. Species distribution models and maps from large-scale biodiversity dataare necessary for conservation management. One current issue is thatbiodiversity data are prone to taxonomic misclassifications. Methods to accountfor these misclassifications in multispecies distribution models have assumedthat the classification probabilities are constant throughout the study. Inreality, classification probabilities are likely to vary with severalcovariates. Failure to account for such heterogeneity can lead to bias inparameter estimates. 2. Here we present a general multispecies distributionmodel that accounts for heterogeneity in the classification process. Theproposed model assumes a multinomial generalised linear model for theclassification confusion matrix. We compare the performance of theheterogeneous classification model to that of the homogeneous classificationmodel by assessing how well they estimate the parameters in the model and theirpredictive performance on hold-out samples. We applied the model to gull datafrom Norway, Denmark and Finland, obtained from GBIF. 3. Our simulation studyshowed that accounting for heterogeneity in the classification processincreased precision by 30% and reduced accuracy and recall by 6%. Applying themodel framework to the gull dataset did not improve the predictive performancebetween the homogeneous and heterogeneous models due to the smallermisclassified sample sizes. However, when machine learning predictive scoresare used as weights to inform the species distribution models about theclassification process, the precision increases by 70%. 4. We recommendmultiple multinomial regression to be used to model the variation in theclassification process when the data contains relatively larger misclassifiedsamples. Machine prediction scores should be used when the data containsrelatively smaller misclassified samples.

Modelling heterogeneity in the classification process in multi-species distribution models can improve predictive performance