🧼

docs/_tex/index.tex

@@ -219,7 +219,7 @@
 1%
 }%
 }
-\date{2024-05-17}
+\date{2024-06-04}
 
 \usepackage{setspace}
 \usepackage[left]{lineno}
@@ -330,11 +330,11 @@
 will aid them in being able to select the correct model to help them to
 reach their goal. In order to be able to make informed decisions it is
 important that one has a strong grasp of exactly what it means to
-`code'/define a food web Section~\ref{sec-network-anatomy}, a clear
+`code'/define a food web (Section~\ref{sec-network-anatomy}), a clear
 understanding of why one wants to predict a food web
-Section~\ref{sec-network-why}, and ultimately one needs to be able to
+(Section~\ref{sec-network-why}), and ultimately one needs to be able to
 asses and evaluate which model family is going to best match up with the
-goal of network prediction Section~\ref{sec-network-build}. Here we
+goal of network prediction (Section~\ref{sec-network-build}). Here we
 specifically aim to look at not look at only the performance of the
 different models but also initiate a (thus far lacking) discussion
 around how the interplay between the language used to define networks
@@ -451,7 +451,7 @@ \subsection{Putting the parts together; what does it
 research programmes (or even practical needs) that have been driving the
 construction of them.
 
-\begin{tcolorbox}[enhanced jigsaw, toprule=.15mm, leftrule=.75mm, breakable, rightrule=.15mm, opacitybacktitle=0.6, arc=.35mm, title=\textcolor{quarto-callout-note-color}{\faInfo}\hspace{0.5em}{Box 1 - Mechanisms that determine feeding links}, colback=white, titlerule=0mm, opacityback=0, colframe=quarto-callout-note-color-frame, left=2mm, bottomtitle=1mm, coltitle=black, toptitle=1mm, bottomrule=.15mm, colbacktitle=quarto-callout-note-color!10!white]
+\begin{tcolorbox}[enhanced jigsaw, bottomrule=.15mm, toptitle=1mm, colback=white, titlerule=0mm, colbacktitle=quarto-callout-note-color!10!white, breakable, arc=.35mm, leftrule=.75mm, bottomtitle=1mm, opacityback=0, colframe=quarto-callout-note-color-frame, title=\textcolor{quarto-callout-note-color}{\faInfo}\hspace{0.5em}{Box 1 - Mechanisms that determine feeding links}, left=2mm, rightrule=.15mm, toprule=.15mm, opacitybacktitle=0.6, coltitle=black]
 
 There are many ideas as to what are the underlying mechanisms that
 determine the links between species. The way one chooses to encode a
@@ -726,7 +726,7 @@ \subsection{Model families}\label{model-families}
 
 \end{figure}%
 
-\begin{tcolorbox}[enhanced jigsaw, toprule=.15mm, leftrule=.75mm, breakable, rightrule=.15mm, opacitybacktitle=0.6, arc=.35mm, title=\textcolor{quarto-callout-note-color}{\faInfo}\hspace{0.5em}{Box 2 - Assessing model outputs}, colback=white, titlerule=0mm, opacityback=0, colframe=quarto-callout-note-color-frame, left=2mm, bottomtitle=1mm, coltitle=black, toptitle=1mm, bottomrule=.15mm, colbacktitle=quarto-callout-note-color!10!white]
+\begin{tcolorbox}[enhanced jigsaw, bottomrule=.15mm, toptitle=1mm, colback=white, titlerule=0mm, colbacktitle=quarto-callout-note-color!10!white, breakable, arc=.35mm, leftrule=.75mm, bottomtitle=1mm, opacityback=0, colframe=quarto-callout-note-color-frame, title=\textcolor{quarto-callout-note-color}{\faInfo}\hspace{0.5em}{Box 2 - Assessing model outputs}, left=2mm, rightrule=.15mm, toprule=.15mm, opacitybacktitle=0.6, coltitle=black]
 
 Although understanding the underlying philosophy of the different model
 families is beneficial it is also important to understand in what

docs/_tex/references.bib

@@ -1,3 +1,19 @@
+@article{adhuryaNovelMethodPredicting,
+  title = {A Novel Method for Predicting Ecological Interactions with an Unsupervised Machine Learning Algorithm},
+  author = {Adhurya, Sagar and Park, Young-Seuk},
+  journal = {Methods in Ecology and Evolution},
+  volume = {n/a},
+  number = {n/a},
+  issn = {2041-210X},
+  doi = {10.1111/2041-210X.14358},
+  urldate = {2024-06-03},
+  abstract = {This gap in knowledge regarding ecological interactions has prompted the development of various predictive approaches. Traditionally, ecological interactions have been inferred using traits. However, the lack of trait information for numerous organisms necessitates using phylogenetic data and statistical insights from interaction matrices for prediction. Previous studies have overlooked the validation of model-predicted interactions. This study used a novel method in predicting ecological interactions using a self-organizing map (SOM), an unsupervised machine learning algorithm. The SOM learns from the input interaction matrix by grouping the nodes into output layers based on their interactions. Subsequently, the trained model predicts the interactions as scores. To distinguish between interactions and non-interactions, we employed F1 score maximization, setting scores above a specific threshold as interactions and the remainder as non-interactions. We applied this method to three unipartite metawebs and one bipartite metaweb and subsequently validated the predicted interactions using two innovative approaches: taxonomic and interaction recovery validation. Our method exhibited outstanding predictive performance, particularly for large networks. Various binary classification performance indicators, including F1 score (0.84--0.97) and accuracy (0.97--0.99), indicated high performance. Moreover, the method generated minimal predicted interactions, signifying low noise in the predictions, particularly for large networks. Taxonomic validation excels in metawebs with a connectance {$>$}0.1 but performs poorly in metawebs with very low connectance. In contrast, interaction recovery was most effective in larger metawebs. Our proposed method excels at making highly accurate predictions of ecological interactions with minimal noise, solely utilizing input interaction data without relying on traits or phylogenetic information regarding interacting nodes. These predictions are particularly precise for large networks, underscoring their potential to address knowledge gaps in emerging extensive metawebs. Notably, taxonomic validation and interaction recovery methods are sensitive to connectance and network size, respectively, suggesting prospects for developing robust interaction validation methods.},
+  copyright = {{\copyright} 2024 The Author(s). Methods in Ecology and Evolution published by John Wiley \& Sons Ltd on behalf of British Ecological Society.},
+  langid = {english},
+  keywords = {ecological interaction,ecological network,Eltonian shortfall,interaction prediction,interaction validation,metaweb,network prediction,self-organizing map (SOM)},
+  file = {/Users/tanyastrydom/Zotero/storage/6MDYZGGG/Adhurya and Park - A novel method for predicting ecological interacti.pdf;/Users/tanyastrydom/Zotero/storage/EGHM9LFC/2041-210X.html}
+}
+
 @article{allesinaFoodWebModels2009,
   title = {Food Web Models: A Plea for Groups},
   shorttitle = {Food Web Models},
@@ -486,6 +502,27 @@ @article{cohenStochasticTheoryCommunity1985
   file = {/Users/tanyastrydom/Zotero/storage/ZGDQ9C63/Cohen et al. - 1997 - A stochastic theory of community food webs I. Mode.pdf}
 }
 
+@article{cooperDeepDiveEstimatingGlobal2024,
+  title = {{{DeepDive}}: Estimating Global Biodiversity Patterns through Time Using Deep Learning},
+  shorttitle = {{{DeepDive}}},
+  author = {Cooper, Rebecca B. and {Flannery-Sutherland}, Joseph T. and Silvestro, Daniele},
+  year = {2024},
+  month = may,
+  journal = {Nature Communications},
+  volume = {15},
+  number = {1},
+  pages = {4199},
+  publisher = {Nature Publishing Group},
+  issn = {2041-1723},
+  doi = {10.1038/s41467-024-48434-7},
+  urldate = {2024-06-04},
+  abstract = {Understanding how biodiversity has changed through time is a central goal of evolutionary biology. However, estimates of past biodiversity are challenged by the inherent incompleteness of the fossil record, even when state-of-the-art statistical methods are applied to adjust estimates while correcting for sampling biases. Here we develop an approach based on stochastic simulations of biodiversity and a deep learning model to infer richness at global or regional scales through time while incorporating spatial, temporal and taxonomic sampling variation. Our method outperforms alternative approaches across simulated datasets, especially at large spatial scales, providing robust palaeodiversity estimates under a wide range of preservation scenarios. We apply our method on two empirical datasets of different taxonomic and temporal scope: the Permian-Triassic record of marine animals and the Cenozoic evolution of proboscideans. Our estimates provide a revised quantitative assessment of two mass extinctions in the marine record and reveal rapid diversification of proboscideans following their expansion out of Africa and a\,{$>$}70\% diversity drop in the Pleistocene.},
+  copyright = {2024 The Author(s)},
+  langid = {english},
+  keywords = {Biodiversity,Machine learning,Palaeontology,Taxonomy},
+  file = {/Users/tanyastrydom/Zotero/storage/TPYM2GVT/Cooper et al. - 2024 - DeepDive estimating global biodiversity patterns .pdf}
+}
+
 @article{daleGraphsSpatialGraphs2010,
   title = {From {{Graphs}} to {{Spatial Graphs}}},
   author = {Dale, M.R.T. and Fortin, M.-J.},