Tampere University of Technology

TUTCRIS Research Portal

The effect of automated taxa identification errors on biological indices

Research output: Contribution to journalArticleScientificpeer-review

Standard

The effect of automated taxa identification errors on biological indices. / Ärje, Johanna; Kärkkäinen, Salme; Meissner, Kristian; Iosifidis, Alexandros; Ince, Türker; Gabbouj, Moncef; Kiranyaz, Serkan.

In: Expert Systems with Applications, Vol. 72, 13.12.2016, p. 108-120.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

Ärje, J, Kärkkäinen, S, Meissner, K, Iosifidis, A, Ince, T, Gabbouj, M & Kiranyaz, S 2016, 'The effect of automated taxa identification errors on biological indices', Expert Systems with Applications, vol. 72, pp. 108-120. https://doi.org/10.1016/j.eswa.2016.12.015

APA

Ärje, J., Kärkkäinen, S., Meissner, K., Iosifidis, A., Ince, T., Gabbouj, M., & Kiranyaz, S. (2016). The effect of automated taxa identification errors on biological indices. Expert Systems with Applications, 72, 108-120. https://doi.org/10.1016/j.eswa.2016.12.015

Vancouver

Ärje J, Kärkkäinen S, Meissner K, Iosifidis A, Ince T, Gabbouj M et al. The effect of automated taxa identification errors on biological indices. Expert Systems with Applications. 2016 Dec 13;72:108-120. https://doi.org/10.1016/j.eswa.2016.12.015

Author

Ärje, Johanna ; Kärkkäinen, Salme ; Meissner, Kristian ; Iosifidis, Alexandros ; Ince, Türker ; Gabbouj, Moncef ; Kiranyaz, Serkan. / The effect of automated taxa identification errors on biological indices. In: Expert Systems with Applications. 2016 ; Vol. 72. pp. 108-120.

Bibtex - Download

@article{55550b15630349fe8abe4f70a12c5ee2,
title = "The effect of automated taxa identification errors on biological indices",
abstract = "In benthic macroinvertebrate biomonitoring systems, the target is to determine the status of ecosystems based on several biological indices. To increase cost-efficiency, computer-based taxa identification for image data has recently been developed. Taxa identification errors can, however, have strong effects on the indices and thus on the determination of the ecological status. In order to shift the biomonitoring process towards automated expert systems, we need a clear understanding on the bias caused by automation. In this paper, we examine eleven classification methods in the case of macroinvertebrate image data and show how their classification errors propagate into different biological indices. We evaluate 14 richness, diversity, dominance and similarity indices commonly used in biomonitoring. Besides the error rate of the classification method, we discuss the potential effect of different types of identification errors. Finally, we provide recommendations on indices that are least affected by the automatic identification errors and could be used in automated biomonitoring.",
keywords = "Biomonitoring, Classification error, Diversity: Error propagation, Identification, Similarity",
author = "Johanna {\"A}rje and Salme K{\"a}rkk{\"a}inen and Kristian Meissner and Alexandros Iosifidis and T{\"u}rker Ince and Moncef Gabbouj and Serkan Kiranyaz",
year = "2016",
month = "12",
day = "13",
doi = "10.1016/j.eswa.2016.12.015",
language = "English",
volume = "72",
pages = "108--120",
journal = "Expert Systems with Applications",
issn = "0957-4174",
publisher = "Elsevier",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - The effect of automated taxa identification errors on biological indices

AU - Ärje, Johanna

AU - Kärkkäinen, Salme

AU - Meissner, Kristian

AU - Iosifidis, Alexandros

AU - Ince, Türker

AU - Gabbouj, Moncef

AU - Kiranyaz, Serkan

PY - 2016/12/13

Y1 - 2016/12/13

N2 - In benthic macroinvertebrate biomonitoring systems, the target is to determine the status of ecosystems based on several biological indices. To increase cost-efficiency, computer-based taxa identification for image data has recently been developed. Taxa identification errors can, however, have strong effects on the indices and thus on the determination of the ecological status. In order to shift the biomonitoring process towards automated expert systems, we need a clear understanding on the bias caused by automation. In this paper, we examine eleven classification methods in the case of macroinvertebrate image data and show how their classification errors propagate into different biological indices. We evaluate 14 richness, diversity, dominance and similarity indices commonly used in biomonitoring. Besides the error rate of the classification method, we discuss the potential effect of different types of identification errors. Finally, we provide recommendations on indices that are least affected by the automatic identification errors and could be used in automated biomonitoring.

AB - In benthic macroinvertebrate biomonitoring systems, the target is to determine the status of ecosystems based on several biological indices. To increase cost-efficiency, computer-based taxa identification for image data has recently been developed. Taxa identification errors can, however, have strong effects on the indices and thus on the determination of the ecological status. In order to shift the biomonitoring process towards automated expert systems, we need a clear understanding on the bias caused by automation. In this paper, we examine eleven classification methods in the case of macroinvertebrate image data and show how their classification errors propagate into different biological indices. We evaluate 14 richness, diversity, dominance and similarity indices commonly used in biomonitoring. Besides the error rate of the classification method, we discuss the potential effect of different types of identification errors. Finally, we provide recommendations on indices that are least affected by the automatic identification errors and could be used in automated biomonitoring.

KW - Biomonitoring

KW - Classification error

KW - Diversity: Error propagation

KW - Identification

KW - Similarity

U2 - 10.1016/j.eswa.2016.12.015

DO - 10.1016/j.eswa.2016.12.015

M3 - Article

VL - 72

SP - 108

EP - 120

JO - Expert Systems with Applications

JF - Expert Systems with Applications

SN - 0957-4174

ER -