TY - GEN
T1 - Species distribution modeling with scalability: The case study of P-GARP, a parallel genetic algorithm for rule-set production
AU - Santana, Fabiana
AU - Bravo Pariente, César Alberto
AU - Saraiva, Antonio Mauro
PY - 2017/8/4
Y1 - 2017/8/4
N2 - Species distribution modeling (SDM) calculates a species' probabilistic distribution by combining Environmental raster layers with species datasets. Such models can help to answer complex questions in Ecology/Biology/Health, e.g., by calculating impacts of climate changes in Biodiversity, or the potential for a disease spread (vectors' modeling). Machine learning is largely applied in SDM, being the Genetic Algorithm for Rule-set Production (GARP) one of the most reliable solutions. However, GARP's convergence needs to speedup under certain conditions (high resolution or number of layers), for which this paper proposes P-GARP, a parallel, scalable implementation of GARP. P-GARP was implemented onto a SGI Altix XE 1300 cluster with 2 quad-core processors/node. Preliminary results show an expressive 3.2/node speedup. Premature convergence is not observed in PGARP and its accuracy is very similar to GARP´s. Effective solutions to improve this speedup in even larger scale are proposed, along with a discussion about P-GARP correctness and efficiency.
AB - Species distribution modeling (SDM) calculates a species' probabilistic distribution by combining Environmental raster layers with species datasets. Such models can help to answer complex questions in Ecology/Biology/Health, e.g., by calculating impacts of climate changes in Biodiversity, or the potential for a disease spread (vectors' modeling). Machine learning is largely applied in SDM, being the Genetic Algorithm for Rule-set Production (GARP) one of the most reliable solutions. However, GARP's convergence needs to speedup under certain conditions (high resolution or number of layers), for which this paper proposes P-GARP, a parallel, scalable implementation of GARP. P-GARP was implemented onto a SGI Altix XE 1300 cluster with 2 quad-core processors/node. Preliminary results show an expressive 3.2/node speedup. Premature convergence is not observed in PGARP and its accuracy is very similar to GARP´s. Effective solutions to improve this speedup in even larger scale are proposed, along with a discussion about P-GARP correctness and efficiency.
KW - Ecological niche modeling
KW - GARP
KW - Genetic algorithms
KW - Parallel algorithms
KW - PGARP
KW - Species distribution modeling
UR - http://www.scopus.com/inward/record.url?scp=85044212324&partnerID=8YFLogxK
UR - http://www.mendeley.com/research/species-distribution-modeling-scalability-case-study-pgarp-parallel-genetic-algorithm-ruleset-produc
U2 - 10.1109/IRI.2017.93
DO - 10.1109/IRI.2017.93
M3 - Conference contribution
SN - 9781538615638
VL - 2017-January
T3 - Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017
SP - 162
EP - 170
BT - Proceedings - 2017 IEEE International Conference on Information Reuse and Integration, IRI 2017
PB - IEEE, Institute of Electrical and Electronics Engineers
CY - San Diego, US
ER -