Human experts vs. machines in taxa recognition

TL;DR

This study compares human experts and machine learning models in taxa recognition, showing that deep learning approaches can approach expert accuracy and challenging previous assumptions about hierarchical classification.

Contribution

We introduce a systematic deep learning approach for taxa recognition, evaluate it against human experts, and provide a new benchmark dataset for future research.

Findings

Deep learning models achieve near-human accuracy in taxa recognition.

Flat classification approaches outperform hierarchical methods in machine learning.

A new multi-pose taxonomic dataset is publicly available for benchmarking.

Abstract

The step of expert taxa recognition currently slows down the response time of many bioassessments. Shifting to quicker and cheaper state-of-the-art machine learning approaches is still met with expert scepticism towards the ability and logic of machines. In our study, we investigate both the differences in accuracy and in the identification logic of taxonomic experts and machines. We propose a systematic approach utilizing deep Convolutional Neural Nets with the transfer learning paradigm and extensively evaluate it over a multi-pose taxonomic dataset with hierarchical labels specifically created for this comparison. We also study the prediction accuracy on different ranks of taxonomic hierarchy in detail. We used support vector machine classifier as a benchmark. Our results revealed that human experts using actual specimens yield the lowest classification error ($\overline{CE}=6.1\%$). However, a much faster, automated approach using deep Convolutional Neural Nets comes close to human accuracy ($\overline{CE}=11.4\%$) when a typical flat classification approach is used. Contrary to previous findings in the literature, we find that for machines following a typical flat classification approach commonly used in machine learning performs better than forcing machines to adopt a hierarchical, local per parent node approach used by human taxonomic experts ($\overline{CE}=13.8\%$). Finally, we publicly share our unique dataset to serve as a public benchmark dataset in this field.