排序方式: 共有3条查询结果,搜索用时 0 毫秒
1
1.
机器学习是一种面向机器的数据分析方法,自动化机器学习的研究促进了人工智能的发展。大数据的快速积累,促进了机器学习算法的井喷式发展。如何选择合适的机器学习算法解决行业问题,成为了当前应用的难点。笔者整理了机器学习新材料,对各种机器算法的特点和算法之间的差异,进行了仔细的梳理,总结了各种算法的需求背景和优缺点,以及主要的应用场合。在此基础上,分析了机器学习在农业应用的案例,综述了机器学习在农业应用,指出了目前存在的发展瓶颈,并提出了进一步研究应用的建议。 相似文献
2.
Alexandre M. J.-C. Wadoux 《European Journal of Soil Science》2023,74(3):e13370
Spectroscopic modelling of soil has advanced greatly with the development of large spectral libraries, computational resources and statistical modelling. The use of complex statistical and algorithmic tools from the field of machine learning has become popular for predicting properties from their visible, near- and mid-infrared spectra. Many users, however, find it difficult to trust the predictions made with machine learning. We lack interpretation and understanding of how the predictions were made, so that these models are often referred to as black boxes. In this study, I report on the development and application of a model-independent method for interpreting complex machine learning spectroscopic models. The method relies on Shapley values, a statistical approach originally developed in coalitional game theory. In a case study for predicting the total organic carbon from a large European mid-infrared spectroscopic database, I fitted a random forest machine learning model and showed how Shapley values can help us understand (i) the average contribution of individual wavenumbers, (ii) the contribution of the spectrum-specific wavenumbers, and (iii) the average contribution of groups of spectra taken together with similar characteristics. The results show that Shapley values revealed more insights than commonly used interpretation methods based on the variable importance. The most striking spectral regions identified as important contributors to the prediction corresponded to the molecular vibration of organic and inorganic compounds that are known to relate to organic carbon. Shapley values are a useful methodological development that will yield a better understanding and trust of complex machine learning and algorithmic tool in soil spectroscopy research. 相似文献
3.
Nathan J. Robinson Peter G. Dahlhaus Megan Wong Andrew MacLeod Dean Jones Cam Nicholson 《Soil Use and Management》2019,35(1):94-104
Soil data form the basis of soil information systems across the globe. Soil information needs, and the questions posed by users, are likely to evolve in response to advances in technology in this era of Big Data. This poses a challenge to the pedological community which is already experiencing a decline in soil knowledge and expertise. With a decrease in soil data collection by governments, it is timely to reconsider how and what soil information should be provided to future users. A public–private partnership is advocated to deliver timely and accessible soil information to users. Two public–private provisioning programs are presented, and advantages and considerations for sharing soil data and information amongst industry, government, research organizations, service providers and land managers for these are discussed. Interoperable, open‐source and agreed soil community standards are used to present soil data and information through spatial web portals with tools for interpretation of soil data for public and private beneficiaries. 相似文献
1