Jintao Kong , Hongru Yu , Junyi Sun , Huan Zhang , Miaomiao Zhang , Zhi Xia
{"title":"Classifying zircon: A machine-learning approach using zircon geochemistry","authors":"Jintao Kong , Hongru Yu , Junyi Sun , Huan Zhang , Miaomiao Zhang , Zhi Xia","doi":"10.1016/j.gr.2024.09.010","DOIUrl":null,"url":null,"abstract":"<div><div>This study presented a novel, rapid, and accurate method for determining zircon origin via a comprehensive analysis of a dataset containing 27,818 zircon trace element sets. This method integrated back propagation neural networks with the AdaBoost algorithm. The optimal classifier characterized as a linear combination of a two-layer neural network model, comprised 100 base classifiers and 400 hidden neurons. It was rigorously trained over 1000 iterations, which resulted in an unbiased error rate of 8.31%. To facilitate practical application, the classifier was integrated into a macro-enabled Excel spreadsheet.</div></div>","PeriodicalId":12761,"journal":{"name":"Gondwana Research","volume":"137 ","pages":"Pages 227-233"},"PeriodicalIF":7.2000,"publicationDate":"2024-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Gondwana Research","FirstCategoryId":"89","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S1342937X24002788","RegionNum":1,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GEOSCIENCES, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
Abstract
This study presented a novel, rapid, and accurate method for determining zircon origin via a comprehensive analysis of a dataset containing 27,818 zircon trace element sets. This method integrated back propagation neural networks with the AdaBoost algorithm. The optimal classifier characterized as a linear combination of a two-layer neural network model, comprised 100 base classifiers and 400 hidden neurons. It was rigorously trained over 1000 iterations, which resulted in an unbiased error rate of 8.31%. To facilitate practical application, the classifier was integrated into a macro-enabled Excel spreadsheet.
期刊介绍：
Gondwana Research (GR) is an International Journal aimed to promote high quality research publications on all topics related to solid Earth, particularly with reference to the origin and evolution of continents, continental assemblies and their resources. GR is an "all earth science" journal with no restrictions on geological time, terrane or theme and covers a wide spectrum of topics in geosciences such as geology, geomorphology, palaeontology, structure, petrology, geochemistry, stable isotopes, geochronology, economic geology, exploration geology, engineering geology, geophysics, and environmental geology among other themes, and provides an appropriate forum to integrate studies from different disciplines and different terrains. In addition to regular articles and thematic issues, the journal invites high profile state-of-the-art reviews on thrust area topics for its column, ''GR FOCUS''. Focus articles include short biographies and photographs of the authors. Short articles (within ten printed pages) for rapid publication reporting important discoveries or innovative models of global interest will be considered under the category ''GR LETTERS''.
京公网安备 11010802042870号
京ICP备2023020795号-1
分享
求助内容：
应助结果提醒方式：
扫码关注我们
