外文摘要：A method for predicting field specific crop yield recommendations. A server computer system receives over a network, digital agricultural data records, including remotely sensed spectral properties of plants and soil moisture records. The computer system aggregates the digital records to create and store geo-specific time series over a specified time. The computer system selects representative features from the geo-specific time series and creates, for each specific geographic area, a covariate matrix comprising the representative features. The computer system assigns a probability value to a component group in a set of parameter component groups, where each component group includes regression coefficients calculated from a probability distribution. The computer system generates the probability distributions used to determine the regression coefficients, the probability distribution used to generate the error term is defined with a mean parameter set at zero and a variance parameter set to a field specific bias coefficient.
申请号： CA3007202
专利号：CA3007202C
公开/公告号：CA3007202C
申请日：2016-12-01
公开/公告日：2023-01-17
申请/专利权人：The Climate Corporation;
发明/设计人：Xu, Lijuan; Xu, Ying
主分类号： A01B79/00;
法律状态：授权
原文传递申请：江苏省科技资源（工程技术文献）统筹服务平台