{"title":"CSIML: a cost-sensitive and iterative machine-learning method for small and imbalanced materials data sets","authors":"Shengzhou Li, Ayako Nakata","doi":"10.1093/chemle/upae090","DOIUrl":null,"url":null,"abstract":"Materials science research benefits from the powerful machine-learning (ML) surrogate models, but it is also limited by the implicit requirement for sufficiently big and balanced data distribution for ML. In this paper, we propose a model to obtain more credible results for small and imbalanced materials data sets as well as chemical knowledge. Taking 2 bandgaps imbalanced data sets as instances, we demonstrate the usability and performance of our model compared with common ML models with normal sampling and resampling methods.","PeriodicalId":9862,"journal":{"name":"Chemistry Letters","volume":"67 1","pages":""},"PeriodicalIF":1.4000,"publicationDate":"2024-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Chemistry Letters","FirstCategoryId":"92","ListUrlMain":"https://doi.org/10.1093/chemle/upae090","RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"CHEMISTRY, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
Abstract
Materials science research benefits from the powerful machine-learning (ML) surrogate models, but it is also limited by the implicit requirement for sufficiently big and balanced data distribution for ML. In this paper, we propose a model to obtain more credible results for small and imbalanced materials data sets as well as chemical knowledge. Taking 2 bandgaps imbalanced data sets as instances, we demonstrate the usability and performance of our model compared with common ML models with normal sampling and resampling methods.
材料科学研究得益于强大的机器学习(ML)代用模型,但也受限于 ML 对足够大且均衡的数据分布的隐性要求。在本文中,我们提出了一种模型,以获得更可信的结果,适用于小而不平衡的材料数据集以及化学知识。以 2 个带隙不平衡数据集为例,我们展示了我们的模型与采用正常采样和重采样方法的普通 ML 模型相比的可用性和性能。