基于模型的潜在因素分析

IF 2.3 3区 环境科学与生态学 Q2 ECOLOGY
Web Ecology Pub Date : 2018-11-14 DOI:10.5194/WE-18-153-2018
H. Gregorius
{"title":"基于模型的潜在因素分析","authors":"H. Gregorius","doi":"10.5194/WE-18-153-2018","DOIUrl":null,"url":null,"abstract":"Abstract. The detection of community or population structure\nthrough analysis of explicit cause–effect modeling of given observations has\nreceived considerable attention. The complexity of the task is mirrored by the\nlarge number of existing approaches and methods, the applicability of which\nheavily depends on the design of efficient algorithms of data analysis. It is\noccasionally even difficult to disentangle concepts and algorithms. To add\nmore clarity to this situation, the present paper focuses on elaborating the\nsystem analytic framework that probably encompasses most of the common\nconcepts and approaches by classifying them as model-based analyses of latent\nfactors. Problems concerning the efficiency of algorithms are not of primary\nconcern here. In essence, the framework suggests an input–output model system\nin which the inputs are provided as latent model parameters and the output is\nspecified by the observations. There are two types of model involved, one of\nwhich organizes the inputs by assigning combinations of potentially\ninteracting factor levels to each observed object, while the other specifies\nthe mechanisms by which these combinations are processed to yield the\nobservations. It is demonstrated briefly how some of the most popular methods\n(Structure, BAPS, Geneland) fit into the framework and how they differ\nconceptually from each other. Attention is drawn to the need to formulate and\nassess qualification criteria by which the validity of the model can be\njudged. One probably indispensable criterion concerns the cause–effect\ncharacter of the model-based approach and suggests that measures of\nassociation between assignments of factor levels and observations be\nconsidered together with maximization of their likelihoods (or posterior\nprobabilities). In particular the likelihood criterion is difficult to realize\nwith commonly used estimates based on Markov chain Monte Carlo (MCMC)\nalgorithms. Generally applicable MCMC-based alternatives that\nallow for approximate employment of the primary qualification criterion and\nthe implied model validation including further descriptors of model\ncharacteristics are suggested.\n","PeriodicalId":54320,"journal":{"name":"Web Ecology","volume":"14 1","pages":""},"PeriodicalIF":2.3000,"publicationDate":"2018-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"Model-based analysis of latent factors\",\"authors\":\"H. Gregorius\",\"doi\":\"10.5194/WE-18-153-2018\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract. The detection of community or population structure\\nthrough analysis of explicit cause–effect modeling of given observations has\\nreceived considerable attention. The complexity of the task is mirrored by the\\nlarge number of existing approaches and methods, the applicability of which\\nheavily depends on the design of efficient algorithms of data analysis. It is\\noccasionally even difficult to disentangle concepts and algorithms. To add\\nmore clarity to this situation, the present paper focuses on elaborating the\\nsystem analytic framework that probably encompasses most of the common\\nconcepts and approaches by classifying them as model-based analyses of latent\\nfactors. Problems concerning the efficiency of algorithms are not of primary\\nconcern here. In essence, the framework suggests an input–output model system\\nin which the inputs are provided as latent model parameters and the output is\\nspecified by the observations. There are two types of model involved, one of\\nwhich organizes the inputs by assigning combinations of potentially\\ninteracting factor levels to each observed object, while the other specifies\\nthe mechanisms by which these combinations are processed to yield the\\nobservations. It is demonstrated briefly how some of the most popular methods\\n(Structure, BAPS, Geneland) fit into the framework and how they differ\\nconceptually from each other. Attention is drawn to the need to formulate and\\nassess qualification criteria by which the validity of the model can be\\njudged. One probably indispensable criterion concerns the cause–effect\\ncharacter of the model-based approach and suggests that measures of\\nassociation between assignments of factor levels and observations be\\nconsidered together with maximization of their likelihoods (or posterior\\nprobabilities). In particular the likelihood criterion is difficult to realize\\nwith commonly used estimates based on Markov chain Monte Carlo (MCMC)\\nalgorithms. Generally applicable MCMC-based alternatives that\\nallow for approximate employment of the primary qualification criterion and\\nthe implied model validation including further descriptors of model\\ncharacteristics are suggested.\\n\",\"PeriodicalId\":54320,\"journal\":{\"name\":\"Web Ecology\",\"volume\":\"14 1\",\"pages\":\"\"},\"PeriodicalIF\":2.3000,\"publicationDate\":\"2018-11-14\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Web Ecology\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://doi.org/10.5194/WE-18-153-2018\",\"RegionNum\":3,\"RegionCategory\":\"环境科学与生态学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"ECOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Web Ecology","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.5194/WE-18-153-2018","RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ECOLOGY","Score":null,"Total":0}
引用次数: 1

摘要

摘要通过对给定观测结果的显式因果模型的分析来检测群落或种群结构已经受到了相当大的关注。任务的复杂性反映在大量现有的方法和方法上,其适用性在很大程度上取决于有效的数据分析算法的设计。有时甚至很难将概念和算法区分开来。为了使这种情况更加清晰,本文将重点放在阐述系统分析框架上,该框架可能包含大多数常见的概念和方法,并将它们分类为基于模型的潜在因素分析。关于算法效率的问题在这里不是主要关注的。从本质上讲,该框架提出了一个输入-输出模型系统,其中输入作为潜在模型参数提供,输出由观测指定。有两种类型的模型,其中一种通过分配潜在的相互作用因素水平的组合来组织输入到每个观察对象,而另一种指定了处理这些组合以产生观察结果的机制。它简要地展示了一些最流行的方法(Structure, BAPS, Geneland)如何适应框架,以及它们在概念上如何彼此不同。提请注意需要制定和评估资格标准,通过这些标准可以判断模型的有效性。一个可能不可或缺的标准涉及基于模型的方法的因果特征,并建议将因素水平分配与观测值之间的关联度量与它们的可能性(或后验概率)的最大化一起考虑。特别是常用的基于马尔可夫链蒙特卡罗(MCMC)算法的估计难以实现似然准则。一般适用的基于mcmc的替代方案,允许近似使用主要资格标准和隐含的模型验证,包括模型特征的进一步描述符。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Model-based analysis of latent factors
Abstract. The detection of community or population structure through analysis of explicit cause–effect modeling of given observations has received considerable attention. The complexity of the task is mirrored by the large number of existing approaches and methods, the applicability of which heavily depends on the design of efficient algorithms of data analysis. It is occasionally even difficult to disentangle concepts and algorithms. To add more clarity to this situation, the present paper focuses on elaborating the system analytic framework that probably encompasses most of the common concepts and approaches by classifying them as model-based analyses of latent factors. Problems concerning the efficiency of algorithms are not of primary concern here. In essence, the framework suggests an input–output model system in which the inputs are provided as latent model parameters and the output is specified by the observations. There are two types of model involved, one of which organizes the inputs by assigning combinations of potentially interacting factor levels to each observed object, while the other specifies the mechanisms by which these combinations are processed to yield the observations. It is demonstrated briefly how some of the most popular methods (Structure, BAPS, Geneland) fit into the framework and how they differ conceptually from each other. Attention is drawn to the need to formulate and assess qualification criteria by which the validity of the model can be judged. One probably indispensable criterion concerns the cause–effect character of the model-based approach and suggests that measures of association between assignments of factor levels and observations be considered together with maximization of their likelihoods (or posterior probabilities). In particular the likelihood criterion is difficult to realize with commonly used estimates based on Markov chain Monte Carlo (MCMC) algorithms. Generally applicable MCMC-based alternatives that allow for approximate employment of the primary qualification criterion and the implied model validation including further descriptors of model characteristics are suggested.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Web Ecology
Web Ecology Agricultural and Biological Sciences-Ecology, Evolution, Behavior and Systematics
CiteScore
4.60
自引率
0.00%
发文量
6
审稿时长
17 weeks
期刊介绍: Web Ecology (WE) is an open-access journal issued by the European Ecological Federation (EEF) representing the ecological societies within Europe and associated members. Its special value is to serve as a publication forum for national ecological societies that do not maintain their own society journal. Web Ecology publishes papers from all fields of ecology without any geographic restriction. It is a forum to communicate results of experimental, theoretical, and descriptive studies of general interest to an international audience. Original contributions, short communications, and reviews on ecological research on all kinds of organisms and ecosystems are welcome as well as papers that express emerging ideas and concepts with a sound scientific background.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信