{"corpus_id":11652370,"paper_sha":"1517e9695cb0bf98ad69ad7fcbff720dd4f79d7d","doi":"10.1111/J.0906-7590.2008.5203.X","arxiv_id":null,"pmid":null,"pmcid":null,"mag_id":2033686454,"dblp_id":null,"acl_id":null,"title":"Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation","year":2008,"publication_date":"2008-04-01","venue":"","journal":{"name":"Ecography","pages":"161-175","volume":"31"},"journal_issn":null,"journal_title":null,"publication_types":[],"pubmed_pub_types":null,"s2_fields_of_study":["Biology","Computer Science","Environmental Science"],"reference_count":40,"citation_count":6739,"influential_citation_count":987,"is_open_access":true,"arxiv_categories":null,"arxiv_license":null,"arxiv_journal_ref":null,"mesh_headings":null,"chemicals":null,"comments_corrections":null,"source_flags":1,"s2_open_access_pdf_url":"https://onlinelibrary.wiley.com/doi/pdfdirect/10.1111/j.0906-7590.2008.5203.x","s2_open_access_landing_url":"https://www.semanticscholar.org/paper/1517e9695cb0bf98ad69ad7fcbff720dd4f79d7d","s2_open_access_license":null,"s2_open_access_status":"BRONZE","pmc_open_access_pdf_url":null,"pmc_open_access_landing_url":null,"pmc_open_access_license":null,"pmc_open_access_status":null,"unpaywall_open_access_pdf_url":null,"unpaywall_open_access_landing_url":null,"unpaywall_open_access_license":null,"unpaywall_open_access_status":null,"abstract":"Accurate modeling of geographic distributions of species is crucial to various applications in ecology and conservation. The best performing techniques often require some parameter tuning, which may be prohibitively time‐consuming to do separately for each species, or unreliable for small or biased datasets. Additionally, even with the abundance of good quality data, users interested in the application of species models need not have the statistical knowledge required for detailed tuning. In such cases, it is desirable to use “default settings”, tuned and validated on diverse datasets. Maxent is a recently introduced modeling technique, achieving high predictive accuracy and enjoying several additional attractive properties. The performance of Maxent is influenced by a moderate number of parameters. The first contribution of this paper is the empirical tuning of these parameters. Since many datasets lack information about species absence, we present a tuning method that uses presence‐only data. We evaluate our method on independently collected high‐quality presence‐absence data. In addition to tuning, we introduce several concepts that improve the predictive accuracy and running time of Maxent. We introduce “hinge features” that model more complex relationships in the training data; we describe a new logistic output format that gives an estimate of probability of presence; finally we explore “background sampling” strategies that cope with sample selection bias and decrease model‐building time. Our evaluation, based on a diverse dataset of 226 species from 6 regions, shows: 1) default settings tuned on presence‐only data achieve performance which is almost as good as if they had been tuned on the evaluation data itself; 2) hinge features substantially improve model performance; 3) logistic output improves model calibration, so that large differences in output values correspond better to large differences in suitability; 4) “target‐group” background sampling can give much better predictive performance than random background sampling; 5) random background sampling results in a dramatic decrease in running time, with no decrease in model performance.","claims":[{"public_id":"cl_d3a913863a870fe5ed6ccbf44acea94a","status":"active","text":"Default settings tuned on presence-only data perform almost as well as settings tuned on the evaluation data itself.","confidence":0.98,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_d3a913863a870fe5ed6ccbf44acea94a"},{"public_id":"cl_00f2fdb9e0036d5cc0283681a4b3a066","status":"active","text":"Hinge features substantially improve Maxent model performance.","confidence":0.97,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_00f2fdb9e0036d5cc0283681a4b3a066"},{"public_id":"cl_bddd9e1706274dcb6cecfdfb49d8c024","status":"active","text":"Logistic output improves model calibration, making differences in output values correspond better to differences in suitability.","confidence":0.96,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_bddd9e1706274dcb6cecfdfb49d8c024"},{"public_id":"cl_f18a9e0a299b0b950f06a4706640932c","status":"active","text":"Random background sampling dramatically decreases running time without reducing model performance.","confidence":0.95,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_f18a9e0a299b0b950f06a4706640932c"},{"public_id":"cl_9005c83ac146f63ed0fe2e941ab828e9","status":"active","text":"Target-group background sampling can yield much better predictive performance than random background sampling.","confidence":0.96,"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/claims/cl_9005c83ac146f63ed0fe2e941ab828e9"}],"concepts":[{"public_id":"co_163273f7b3403fdfd4fd4cc7865be453","status":"active","name":"default settings","description":"Baseline parameter values used for Maxent without species-specific manual tuning.","types":["parameter setting"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_163273f7b3403fdfd4fd4cc7865be453"},{"public_id":"co_18daba5da7acb9274cc6013a7b4c869d","status":"active","name":"evaluation data","description":"Independently collected high-quality presence-absence data used to assess model tuning and performance.","types":["evaluation dataset"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_18daba5da7acb9274cc6013a7b4c869d"},{"public_id":"co_1a2137590a430f8b6fb2052a1d51371e","status":"active","name":"target-group background sampling","description":"A background sampling strategy that uses records from a group of related species or surveys to reduce sampling bias.","types":["sampling strategy"],"aliases":["target group background sampling"],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_1a2137590a430f8b6fb2052a1d51371e"},{"public_id":"co_3f658d1e1d2476b3557fe26991fba884","status":"active","name":"suitability","description":"The degree to which a location is suitable for the species being modeled.","types":["ecological property"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_3f658d1e1d2476b3557fe26991fba884"},{"public_id":"co_456f71b4482c7cccc76de03b348e4714","status":"active","name":"model calibration","description":"The correspondence between predicted output values and the magnitude of differences in habitat suitability.","types":["evaluation property"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_456f71b4482c7cccc76de03b348e4714"},{"public_id":"co_60f4848bdfdba766f482fd297b2a06ae","status":"active","name":"model performance","description":"The effectiveness of a model in predicting species distributions or occurrence.","types":["evaluation metric"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_60f4848bdfdba766f482fd297b2a06ae"},{"public_id":"co_721e941bd7b094da35aac5fed0383861","status":"active","name":"random background sampling","description":"A background sampling strategy that selects background points at random from the study region.","types":["sampling strategy"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_721e941bd7b094da35aac5fed0383861"},{"public_id":"co_967801d6b21dc3f5cbfc6924056d9cd2","status":"active","name":"presence-only data","description":"Occurrence data that record where a species is observed but do not explicitly record absences.","types":["data type"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_967801d6b21dc3f5cbfc6924056d9cd2"},{"public_id":"co_a2d212a18523a57aadbbca4691107767","status":"active","name":"predictive performance","description":"The ability of a model to correctly predict species distribution or occurrence.","types":["evaluation metric"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_a2d212a18523a57aadbbca4691107767"},{"public_id":"co_affcac10964618c62b4485227bc65a54","status":"active","name":"hinge features","description":"Feature functions in Maxent that allow more flexible, piecewise relationships between predictors and species occurrence.","types":["feature type"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_affcac10964618c62b4485227bc65a54"},{"public_id":"co_bc8ce8e055e31ce1bb00cdc8f6874f67","status":"active","name":"logistic output format","description":"A Maxent output transformation intended to estimate probability of presence.","types":["output format"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_bc8ce8e055e31ce1bb00cdc8f6874f67"},{"public_id":"co_bdee32a97ca8a22a65cb26e3e8861dc2","status":"active","name":"Maxent","description":"A species distribution modeling technique used to estimate geographic suitability from occurrence data.","types":["modeling technique"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_bdee32a97ca8a22a65cb26e3e8861dc2"},{"public_id":"co_eba9312b1fcdc46470cde58bcf55aea7","status":"active","name":"running time","description":"The computational time required to build or fit the model.","types":["computational measure"],"aliases":[],"contributors":[{"id":1,"public_id":"12632b8b5f","public_label":"Anonymous (12632b8b5f)","roles":["extraction"],"url":"https://sah.borca.ai/u/12632b8b5f"}],"url":"https://sah.borca.ai/concepts/co_eba9312b1fcdc46470cde58bcf55aea7"}],"external_ids":{"DOI":"10.1111/J.0906-7590.2008.5203.X","ArXiv":null,"PubMed":null,"PubMedCentral":null,"MAG":2033686454,"DBLP":null,"ACL":null},"open_access":{"is_open_access":true,"pdf_url":"https://onlinelibrary.wiley.com/doi/pdfdirect/10.1111/j.0906-7590.2008.5203.x","landing_url":"https://www.semanticscholar.org/paper/1517e9695cb0bf98ad69ad7fcbff720dd4f79d7d","source":"semantic_scholar","pdf_url_source":"semantic_scholar_open_access_pdf","license":null,"status":"BRONZE","reason":null},"reference_availability":{"status":"available","references_indexed":true,"full_text_available":false,"full_text_source":null,"count_basis":"semantic_scholar_metadata","extraction_status":"not_applicable","reason":null},"source":{"provider":"episteme2","base_corpus":"semantic_scholar_dump","freshness_mode":"unknown","basis":["semantic_scholar_metadata","postgres_metadata"],"limits":["paper metadata is based on indexed upstream scholarly datasets","claims and concepts are available only for extracted papers","absence of claims or concepts means no extracted graph data is available in this response"],"status":"available","degraded":false,"degraded_reasons":[],"diagnostics":{"status":"available","degraded":false,"degraded_reasons":[],"metadata_status":"available","graph_status":"available","abstract_status":"available"},"source_flags":1},"paper_id":635806,"paper_uid":"8d5df3d6-84df-4722-9180-b0de38fd8866","canonical_identity":{"paper_id":635806,"paper_uid":"8d5df3d6-84df-4722-9180-b0de38fd8866","identity_status":"available","lookup_basis":"semantic_scholar_external_id","compatibility_path":"corpus_id"},"url":"https://sah.borca.ai/papers/11652370"}