{"title":"迈向所有现存爬行动物物种的数字化描述","authors":"PETER UETZ, YAA ADARKWA DARKO, DUSTIN ZELIFF","doi":"10.11646/megataxa.10.1.6","DOIUrl":null,"url":null,"abstract":"Vertebrate databases have been slow to digitize species descriptions. One of them, the Reptile Database (http://www.reptile-database.org), has accumulated ~8,000 species descriptions for ~3,000 species of snakes, ~5,000 species of lizards, and ~150 species of turtles and crocodiles. Here we discuss how this data contributes to character analysis, species identification, but also to integration with other data sources such as citizen science observations (which depend on correct identifications). Importantly, the data described here may serve as training data for machine learning projects and we present examples of species comparisons using ChatGPT. While these AI-driven comparisons are still erroneous, we expect substantial improvements in the near future. We request the herpetological community to help complete our public collection of species descriptions and suggest that other species databases follow suit and provide similar data for their taxa.","PeriodicalId":52569,"journal":{"name":"Megataxa","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2023-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Towards digital descriptions of all extant reptile species\",\"authors\":\"PETER UETZ, YAA ADARKWA DARKO, DUSTIN ZELIFF\",\"doi\":\"10.11646/megataxa.10.1.6\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Vertebrate databases have been slow to digitize species descriptions. One of them, the Reptile Database (http://www.reptile-database.org), has accumulated ~8,000 species descriptions for ~3,000 species of snakes, ~5,000 species of lizards, and ~150 species of turtles and crocodiles. Here we discuss how this data contributes to character analysis, species identification, but also to integration with other data sources such as citizen science observations (which depend on correct identifications). Importantly, the data described here may serve as training data for machine learning projects and we present examples of species comparisons using ChatGPT. While these AI-driven comparisons are still erroneous, we expect substantial improvements in the near future. We request the herpetological community to help complete our public collection of species descriptions and suggest that other species databases follow suit and provide similar data for their taxa.\",\"PeriodicalId\":52569,\"journal\":{\"name\":\"Megataxa\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-10-31\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Megataxa\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.11646/megataxa.10.1.6\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Megataxa","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.11646/megataxa.10.1.6","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Towards digital descriptions of all extant reptile species
Vertebrate databases have been slow to digitize species descriptions. One of them, the Reptile Database (http://www.reptile-database.org), has accumulated ~8,000 species descriptions for ~3,000 species of snakes, ~5,000 species of lizards, and ~150 species of turtles and crocodiles. Here we discuss how this data contributes to character analysis, species identification, but also to integration with other data sources such as citizen science observations (which depend on correct identifications). Importantly, the data described here may serve as training data for machine learning projects and we present examples of species comparisons using ChatGPT. While these AI-driven comparisons are still erroneous, we expect substantial improvements in the near future. We request the herpetological community to help complete our public collection of species descriptions and suggest that other species databases follow suit and provide similar data for their taxa.