To address the problems that input variables should be reduced as much as possible and explain output variables fully in building neural network model of complicated system, a variable selection method based on cluste...To address the problems that input variables should be reduced as much as possible and explain output variables fully in building neural network model of complicated system, a variable selection method based on cluster (analysis) was investigated. Similarity coefficient which describes the mutual relation of variables was defined. The methods of the highest contribution rate, part replacing whole and variable replacement are put forwarded and deduced by information theory. The software of the neural network based on cluster analysis, which can provide many kinds of methods for defining variable similarity coefficient, clustering system variable and evaluating variable cluster, was developed and applied to build neural network forecast model of cement clinker quality. The results show that all the network scale, training time and prediction accuracy are perfect. The practical application demonstrates that the method of selecting variables for neural network is feasible and effective.展开更多
As the Internet continues to expand, the influence of network information on real life has gradually deepened. Research on propagation and evolution of Internet public opinion has become a hot topic. The force of inte...As the Internet continues to expand, the influence of network information on real life has gradually deepened. Research on propagation and evolution of Internet public opinion has become a hot topic. The force of intemet public opinion penetrates and influences on every aspect of the society. Compared with traditional public opinion, the network public opinion has features of immediate, multivariate and interactive, the propagation behavior has the new change compared with the traditional public opinion. In the propagation behavior of network public opinion, agenda setting is no longer an arbitrary, the influence of opinion leaders in the agenda setting becomes more and more complex and diversified. The formation time of network public opinion is short, and social influence becomes strong. Guide the public opinion to build a harmonious environment of network public opinion. Overall, our country＇ s network public opinion environment is a benign situation and steadily promoting the reforms of public policies. Although there is still a few not rational voices full of them, the network public opinion shows a general trend of positive thinking. Based on this philosophy, through the research of network public opinion dissemination and evolution mechanism, can be all-round good guidance and supervision of public opinion, building a harmonious network environment of public opinion.展开更多
Face-to-face communication is very important skill to share intentions. However, many people in the modem world feel that they are deficient in face-to-face communication. So, we feel that it is necessary to support t...Face-to-face communication is very important skill to share intentions. However, many people in the modem world feel that they are deficient in face-to-face communication. So, we feel that it is necessary to support their face-to-face communication using information technologies. We have developed a topic-providing system that can infer behaviors from daily life and provides users with information about their conversation partner, including that on his hometown, hobbies and life logs when face-to-face communication is initiated. The life logs are details about a user＇s life, and are generated using a Bayesian network on the basis of sensor data provided by our system. This system enables users to access other users＇ information of behaviors from the accumulated life logs and it utilizes this infbrmation to generate topics for conversation. We evaluated the accuracy with which proposal system inferred behaviors to confirm whether exact life log generation is possible. And we also evaluated the proposed system by administering a questionnaire to confirm whether the proposed system can support face-to-face communication.展开更多
文摘To address the problems that input variables should be reduced as much as possible and explain output variables fully in building neural network model of complicated system, a variable selection method based on cluster (analysis) was investigated. Similarity coefficient which describes the mutual relation of variables was defined. The methods of the highest contribution rate, part replacing whole and variable replacement are put forwarded and deduced by information theory. The software of the neural network based on cluster analysis, which can provide many kinds of methods for defining variable similarity coefficient, clustering system variable and evaluating variable cluster, was developed and applied to build neural network forecast model of cement clinker quality. The results show that all the network scale, training time and prediction accuracy are perfect. The practical application demonstrates that the method of selecting variables for neural network is feasible and effective.
文摘As the Internet continues to expand, the influence of network information on real life has gradually deepened. Research on propagation and evolution of Internet public opinion has become a hot topic. The force of intemet public opinion penetrates and influences on every aspect of the society. Compared with traditional public opinion, the network public opinion has features of immediate, multivariate and interactive, the propagation behavior has the new change compared with the traditional public opinion. In the propagation behavior of network public opinion, agenda setting is no longer an arbitrary, the influence of opinion leaders in the agenda setting becomes more and more complex and diversified. The formation time of network public opinion is short, and social influence becomes strong. Guide the public opinion to build a harmonious environment of network public opinion. Overall, our country＇ s network public opinion environment is a benign situation and steadily promoting the reforms of public policies. Although there is still a few not rational voices full of them, the network public opinion shows a general trend of positive thinking. Based on this philosophy, through the research of network public opinion dissemination and evolution mechanism, can be all-round good guidance and supervision of public opinion, building a harmonious network environment of public opinion.
文摘Face-to-face communication is very important skill to share intentions. However, many people in the modem world feel that they are deficient in face-to-face communication. So, we feel that it is necessary to support their face-to-face communication using information technologies. We have developed a topic-providing system that can infer behaviors from daily life and provides users with information about their conversation partner, including that on his hometown, hobbies and life logs when face-to-face communication is initiated. The life logs are details about a user＇s life, and are generated using a Bayesian network on the basis of sensor data provided by our system. This system enables users to access other users＇ information of behaviors from the accumulated life logs and it utilizes this infbrmation to generate topics for conversation. We evaluated the accuracy with which proposal system inferred behaviors to confirm whether exact life log generation is possible. And we also evaluated the proposed system by administering a questionnaire to confirm whether the proposed system can support face-to-face communication.
