In this paper,an approach is developed to optimize the quality of the training samples in the conventional Artificial Neural Network(ANN)by incorporating expert knowledge in the means of constructing expert-rule sampl...In this paper,an approach is developed to optimize the quality of the training samples in the conventional Artificial Neural Network(ANN)by incorporating expert knowledge in the means of constructing expert-rule samples from rules in an expert system,and through training by using these samples,an ANN based on expert-knowledge is further developed.The method is introduced into the field of quantitative identification of potential seismic sources on the basis of the rules in an expert system.Then it is applied to the quantitative identification of the potential seismic sources in Beijing and its adjacent area.The result indicates that the expert rule based on ANN method can well incorporate and represent the expert knowledge in the rules in an expert system,and the quality of the samples and the efficiency of training and the accuracy of the result are optimized.展开更多
We demonstrate theoretically and experimentally how changes of a terahertz （THz） beam induced by the sample affect the accuracy of the determination of THz dielectric properties in THz time-domain transmission spect...We demonstrate theoretically and experimentally how changes of a terahertz （THz） beam induced by the sample affect the accuracy of the determination of THz dielectric properties in THz time-domain transmission spectros- copy （TDTS）. We apply a Gaussian beam and the ABCD matrix formalism to describe the propagation of the THz beam in a focused beam setup. The insertion of the sample induces a focus displacement which is absent in the reference t without a sample. We show how the focus displacement can be corrected. The THz optical properties after focus displacement correction reported in this Letter are in quantitative agreement with those obtained using collimated beam THz-TDTSinpreviouswork.展开更多
文摘In this paper,an approach is developed to optimize the quality of the training samples in the conventional Artificial Neural Network(ANN)by incorporating expert knowledge in the means of constructing expert-rule samples from rules in an expert system,and through training by using these samples,an ANN based on expert-knowledge is further developed.The method is introduced into the field of quantitative identification of potential seismic sources on the basis of the rules in an expert system.Then it is applied to the quantitative identification of the potential seismic sources in Beijing and its adjacent area.The result indicates that the expert rule based on ANN method can well incorporate and represent the expert knowledge in the rules in an expert system,and the quality of the samples and the efficiency of training and the accuracy of the result are optimized.
基金the support from the Center for Applied Photonics (CAP) at the University of Konstanzthe DFG through the SFB 767 (Germany)the China Scholarship Council (CSC)
文摘We demonstrate theoretically and experimentally how changes of a terahertz （THz） beam induced by the sample affect the accuracy of the determination of THz dielectric properties in THz time-domain transmission spectros- copy （TDTS）. We apply a Gaussian beam and the ABCD matrix formalism to describe the propagation of the THz beam in a focused beam setup. The insertion of the sample induces a focus displacement which is absent in the reference t without a sample. We show how the focus displacement can be corrected. The THz optical properties after focus displacement correction reported in this Letter are in quantitative agreement with those obtained using collimated beam THz-TDTSinpreviouswork.
