Abstract
Temporal data is ubiquitous in ecology and ecologists often face the challenge of accurately differentiating these data into predefined classes, such as biological entities or ecological states. The usual approach consists of transforming the time series into user-defined features and then using these features as predictors in conventional statistical or machine learning models. Here we suggest the use of deep learning models as an alternative to this approach. Recent deep learning techniques can perform the classification directly from the time series, eliminating subjective and resource-consuming data transformation steps, and potentially improving classification results. We describe some of the deep learning architectures relevant for time series classification and show how these architectures and their hyper-parameters can be tested and used for the classification problems at hand. We illustrate the approach using three case studies from distinct ecological subdisciplines: i) insect species identification from wingbeat spectrograms; ii) species distribution modelling from climate time series and iii) the classification of phenological phases from continuous meteorological data. The deep learning approach delivered ecologically sensible and accurate classifications demonstrating its potential for wide applicability across subfields of ecology.
Highlights
The recent increase in affordability, capacity, and autonomy of sensor-based technologies (Bush et al, 2017; Peters et al, 2014), as well as an increasing number of contributions from citizen scientists and the establishment of international research networks (Bush et al, 2017; Hurlbert and Liang, 2012), is allowing an unprecedented access to time series of interest for ecological research
We here introduce the use of deep learning models as a generic approach for the classification of temporal data and demonstrate how these models can be implemented and evaluated for distinct tasks across subfields of ecology
In the first case study, an Incep tionTime model performed well in distinguishing insect species based on spectrograms of their wingbeats
Summary
The recent increase in affordability, capacity, and autonomy of sensor-based technologies (Bush et al, 2017; Peters et al, 2014), as well as an increasing number of contributions from citizen scientists and the establishment of international research networks (Bush et al, 2017; Hurlbert and Liang, 2012), is allowing an unprecedented access to time series of interest for ecological research. A common aim of ecologists using these data concerns assigning them into predefined classes, such as ecological states or biological entities. Typical examples include the recognition of bird species from sound recordings (e.g. Priyadarshani et al, 2020), the distinction between phases in the annual life cycle of plants (i.e., ‘phenophases’) from spectral time series (Melaas et al, 2013), or the recognition of behavioral states from animal movement data (Shamoun-Baranes et al, 2016). The assignment of time series into one of two or more predefined classes (hereafter referred to as ‘time series classification’; Keogh and Kasetty, 2003) can be performed using a variety of different approaches, ranging from manual, expert-based, classification (Priyadarshani et al, 2020) to fully automated procedures (see Bagnall et al, 2017 for ex amples). This approach is known as ‘feature-based’, where the ‘features’ are the variables that are extracted to summarize the time series
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
