Japan STA Program: Expert Systems and Neural Networks for Agriculture and Crop Management

该奖项将提供补充支持，使科罗拉多州立大学的Israel Broner博士能够与Hirafiji Masayuki先生在日本筑波国家农业研究中心进行为期12个月的合作研究。Broner和Masayuki将把Masayuki的递归神经网络（RNN）知识应用到Broner的基于知识的专家系统（KBES）原型中，以开发农业应用的决策支持系统（DSS）。​用于作物生产的决策支助系统将以一种非常便携和易于获取的媒介提供管理咨询意见。KBES是系统的记忆部分，它使用来自人类专家和研究结果的知识，但往往是不完整和不确定的。RNN组件使用其反馈连接，将被编程为调用历史信息来预测不同管理操作的时间表。在进行任何RNN预测之前，需要对归纳和模糊逻辑进行基础研究，以解决KBES与RNN之间的冲突。Masayuki博士开发的神经网络计算将补充Broner博士在KBES方面的专业知识，并将促进DSS的发展，为美国小规模农民提供长期帮助。

This award will provide supplementary support to enable Dr. Israel Broner of the Colorado State University to conduct collaborative research with Mr. Hirafiji Masayuki for 12 months at the National Agriculture Research Center in Tsukuba, Japan. Broner and Masayuki will apply Masayuki's knowledge of recurrent neural networks (RNN) to Broner's prototype of a knowledge based expert system (KBES) to develop a decision support system (DSS) for agricultural applications. A DSS for crop production will provide management advice in a very portable and easily accessible medium. The KBES is the memory component of the system which uses knowledge derived from human experts and research results, but is often incomplete and uncertain. The RNN component, using its feedback connections, will be programmed to call on the historical information to predict schedules for different management operations. Before any RNN predictions can be made, basic research on induction and fuzzy logic is needed to resolve the conflicts between the KBES and RNN. The neural network computations developed by Dr. Masayuki will complement Dr. Broner's expertise in KBES and will facilitate the development of DSS with a long-term benefit of aiding the small scale U.S. farmer.