A semi-automated approach to classify and map ecological zones across the dune-beach interface

Guilherme, Pablo Damian Borges; Borzone, Carlos Alberto; Padial, André Andrian; Harris, Linda R.

doi:10.1016/j.ecss.2018.04.030

Cited by 3 publications

(3 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Decision trees and other supervised machine learning approaches (e.g., random forest) attempt to model the relationship between a response and its predictors (Breiman 2001), offer powerful alternatives to traditional ecological modeling approaches (e.g., generalized linear models; De'ath and Fabricius 2000, Olden et al 2008), and have been increasingly applied in investigations of wildlife habitat use (Han et al 2017, Mi et al 2017, Cushman and Wasserman 2018, Carroll et al 2021, Rather et al 2021). We built decision trees by incorporating plot‐level data into a boosted C5.0 algorithm (Quinlan 1993), which we chose because of its nominal sensitivity to multicollinearity and unbalanced data, ability to manage missing values, and relative ease of interpretation (Guilherme et al 2018, Szilassi et al 2019, Moeinaddini et al 2020, Tanyu et al 2021, da Silveira et al 2022). Allowing for missing values was particularly important in selecting our modeling approach, as calculation of some values in our dataset was contingent on another value being non‐zero (e.g., we could not calculate average log diameter in plots that did not contain any logs).…”

Section: Methodsmentioning

confidence: 99%

Structural complexity characterizes fine‐scale forest conditions used by Pacific martens

et al. 2023

View full text Add to dashboard Cite

When wildlife species exhibit unexpected associations with vegetation, replication of studies in different locales can illuminate whether patterns of use are consistent or divergent. Our objective was to describe fine-scale forest conditions used by Pacific martens (Martes caurina) at 2 study sites in northern California that differed in forest composition and past timber harvest. We identified denning and resting locations of radio-marked martens and sampled structure-and plot-level vegetation using standardized forest inventory methods between 2009-2021. Woody structures used by martens were significantly larger than randomly available structures across types (e.g., live tree, snag, log) and at both study sites. Den and rest structures occurred in areas characterized by higher numbers of logs and snags, lower numbers of live trees and stumps, larger diameter live trees and logs, and

show abstract

Section: Methodsmentioning

confidence: 99%

Structural complexity characterizes fine‐scale forest conditions used by Pacific martens

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Ecosystem filed also take a lot of focus to implement the AL learning strategies over in order to classify and process its dataset. such studies we could find in [16][17][18].…”

mentioning

confidence: 95%

Untitled

2019

International Journal of Advanced Research in Computer and Comm

View full text Add to dashboard Cite

The main goal of Active Learning (AL), is to empower the learning process by reducing the cost of obtaining the labels with a limited training samples and by selecting the most informative samples from the unlabelled set. AL was implemented to solve a wide range of problems in all fields [1]. In this paper, we will go forward to utilize a multiclass active learning model over an ecological zones dataset in order to classify the dune-beach interface to a divided sub-environments category. Pool, ranked & stream-based sampling were used as an active learning frames where three query strategies were tested with each frame to achieve the best performance. The performance metrics values would be illustrated in three different comparative statements, in term of different framework implemented by different query strategies, in term of incremental learning process behaviour in a pool-based sampling frame work by selecting 20 different query quires from the unlabelled set to present the incremental learning process behaviour with each query strategy (random, entropy & margin) and finally in term of incremental learning process behaviour above a three different AL frame works and three different query strategies in term of accuracy metric.

show abstract

“…(2017) concluded that a support-vector machine outperformed other algorithms when classifying wetland habitat using UAV-derived imagery. Guilherme et al (2018) found that random forest (RF) algorithms yielded the best classification results when mapping coastal habitats compared to seven other machine learning algorithms. However, algorithms based on neural networks appear to outperform every other classifier when using good training data (Liu et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

Comparing random forests and convoluted neural networks for mapping ghost crab burrows using imagery from an unmanned aerial vehicle

Bycroft

Leon

Schoeman

2019

Estuarine, Coastal and Shelf Science

View full text Add to dashboard Cite

A semi-automated approach to classify and map ecological zones across the dune-beach interface

Cited by 3 publications

References 43 publications

Structural complexity characterizes fine‐scale forest conditions used by Pacific martens

Structural complexity characterizes fine‐scale forest conditions used by Pacific martens

Untitled

Comparing random forests and convoluted neural networks for mapping ghost crab burrows using imagery from an unmanned aerial vehicle

Contact Info

Product

Resources

About