mirror of
https://github.com/zoo-apps/papers.git
synced 2026-11-02 02:29:15 +00:00
56 lines
2.7 KiB
TeX
56 lines
2.7 KiB
TeX
\section{Related Work}
|
|
|
|
\label{sec:related}
|
|
% ══════════════════════════════════════════════════════════════════════════════
|
|
|
|
\subsection{Citizen Science Platforms}
|
|
|
|
Major biodiversity citizen science platforms include:
|
|
\begin{itemize}
|
|
\item \textbf{eBird} \citep{sullivan2014ebird}: Structured bird
|
|
checklists with 800M+ observations. Strong protocol design but
|
|
limited to birds.
|
|
\item \textbf{iNaturalist} \citep{inaturalist2024}: Multi-taxon
|
|
platform with AI identification. Broad taxonomic coverage but
|
|
unstructured opportunistic data.
|
|
\item \textbf{Zooniverse} \citep{simpson2014zooniverse}: Platform
|
|
for classification tasks (e.g., galaxy classification, camera
|
|
trap identification).
|
|
\item \textbf{GBIF} \citep{gbif2024}: Global aggregator of
|
|
biodiversity data from multiple sources including citizen science.
|
|
\end{itemize}
|
|
|
|
\subsection{Gamification in Citizen Science}
|
|
|
|
\citet{bowser2013gamifying} identify four gamification strategies:
|
|
competition (leaderboards), cooperation (team goals), exploration
|
|
(discovery quests), and collection (digital specimen albums).
|
|
\citet{eveleigh2014designing} demonstrate that intrinsic motivation
|
|
(learning, contribution to science) outperforms extrinsic rewards for
|
|
sustained engagement. \citet{preece2019citizen} survey
|
|
gamification across 70 citizen science projects, finding that narrative
|
|
elements and progressive skill development are most effective for
|
|
long-term retention.
|
|
|
|
\subsection{Data Quality in Citizen Science}
|
|
|
|
\citet{kelling2015can} demonstrate that structured protocols (complete
|
|
checklists, standardized effort) substantially improve data quality.
|
|
\citet{johnston2020analytical} develop analytical methods for
|
|
accounting for observer skill in eBird data. \citet{isaac2014statistics}
|
|
formalize occupancy models that jointly estimate species presence
|
|
and detection probability from citizen science data.
|
|
|
|
\subsection{AI-Assisted Identification}
|
|
|
|
AI identification assistants in citizen science include iNaturalist's
|
|
computer vision \citep{van2021benchmarking}, Merlin for bird audio
|
|
\citep{kahl2021birdnet}, and PlantNet for plant identification
|
|
\citep{goeau2022overview}. These systems improve identification accuracy
|
|
but introduce new biases: AI suggestions may anchor observer decisions,
|
|
and AI errors propagate through community validation
|
|
\citep{terry2020think}.
|
|
|
|
|
|
% ══════════════════════════════════════════════════════════════════════════════
|