Development of a feedback neural network model of expert's decision-making process

专家决策过程反馈神经网络模型的开发

基本信息

批准号：
15300270
负责人：
NAKAGAWA Masanori
金额：
$ 3.01万
依托单位：
Tokyo Institute of Technology
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (B)
财政年份：
2003
资助国家：
日本
起止时间：
2003 至 2004
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-15300270/
关键词：
Decision-making Neural network Feed back Psychological experiment Computer Simulation Expert

项目摘要

Research results of 2003 year(1)Examined existing models, i.e.corporative model, parallel constraints satisfaction model, and risk taking model, sufficiently.(2)Based on the examination in (1), studied a foundational feed forward neural network model constructed in the computer through its simulation.(3)Studied a feed back neural network model based on the system constructed in (2).(4)Extended the model constructed in (3), based on the variation of feed back structures.(5)Executed psychological experiments of decision-making process including psychological feed back process supposed in the present neural network model.(6)Executed the computer simulation of the model constructed in (4), using the same materials and method as psychological experiments in (5). Then, compared the results of the simulations and the results of experiments.(7)Improved the model in order to simulate the more human-like decision-making process, based on the comparison in (6).Research results of 2004 year(1)Developed a new model based on the system constructed in 2003, in order to simulate the decision making process of experts.(2)Executed psychological experiments of experts' decision making process including psychological feed back process.(3)Executed the computer simulation of the model constructed in (1), using the same materials and method as psychological experiments in (2). Then, compared the results of the simulations and the results of experiments.Based on the comparison in (3), improved the model in order to simulate the more expert-like decision-making process. Then, examined the possibility of application of the model in the educational technology.

2003年（1）的研究结果检查了现有模型，即企业模型，平行约束满意度模型和冒险模型。（2）基于（1）的检查，研究了建立的基础饲料前馈神经网络模型，该模型构建了通过其仿真。（3）根据（2）中构建的系统研究了一个馈回神经网络模型。（4）扩展了基于（3）中构建的模型，基于Feed Back Back结构的变化。（5）。执行决策过程的心理实验，包括当前神经网络模型中假设的心理馈回过程。（6）使用（4）中构建的模型的计算机模拟，使用与（5）中的心理实验相同的材料和方法。然后，比较了模拟的结果和实验的结果。（7）改进了模型，以模拟更类似人类的决策过程，基于（6）中的比较。研究2004年的研究结果（1 ）开发了一种基于2003年构建的系统的新模型，以模拟专家的决策过程。（2）执行专家决策过程的心理实验，包括心理馈回过程。（3）执行计算机模拟的计算机模拟在（1）中构建的模型，使用（2）中的心理实验相同的材料和方法。然后，比较了模拟的结果和实验结果。基于（3）中的比较，改进了模型，以模拟更专家样的决策过程。然后，检查了该模型在教育技术中应用的可能性。