[Objective] The aim was to investigate the status of wild Platycodon grandiflourus resources in Changbai Mountain area. [ Method] The habitats and growth environment of the wild Platycodon grandiflourus were investiga...[Objective] The aim was to investigate the status of wild Platycodon grandiflourus resources in Changbai Mountain area. [ Method] The habitats and growth environment of the wild Platycodon grandiflourus were investigated, and made a collection of the germplasm resources at 24 sites in Changbai Mountain area. [ Result ] Most of the exiting wild Platycodon grandiflourus survived because of its small roots that means no value and grew in poor or remote conditions that led to much difficulty to dig. The character of wild Platycodon grandiflourus was obviously inferior to cultivated Platycodon grandiflourus. [ Condusion] This study lays a foundation for the genetic diversity analysis and the establishment of germplasm resource pool of wild Platycodon grandflourus in Changbai Mountain Area.展开更多
In order to recognize people＇s annoyance emotions in the working environment and evaluate emotional well- being, emotional speech in a work environment is induced to obtain adequate samples of emotional speech, and a...In order to recognize people＇s annoyance emotions in the working environment and evaluate emotional well- being, emotional speech in a work environment is induced to obtain adequate samples of emotional speech, and a Mandarin database with two thousands samples is built. In searching for annoyance-type emotion features, the prosodic feature and the voice quality feature parameters of the emotional statements are extracted first. Then an improved back propagation （BP） neural network based on the shuffled frog leaping algorithm （SFLA） is proposed to recognize the emotion. The recognition capability of the BP, radical basis function （RBF） and the SFLA neural networks are compared experimentally. The results show that the recognition ratio of the SFLA neural network is 4. 7% better than that of the BP neural network and 4. 3% better than that of the RBF neural network. The experimental results demonstrate that the random initial data trained by the SFLA can optimize the connection weights and thresholds of the neural network, speed up the convergence and improve the recognition rate.展开更多
Through exploring the limitation of the neoclassical theory of economic growth,which classifies growth as a homogenous process,this paper reconciles various theories of economic development and explains the rises and ...Through exploring the limitation of the neoclassical theory of economic growth,which classifies growth as a homogenous process,this paper reconciles various theories of economic development and explains the rises and falls of economic growth under a unified framework,focusing on incentives of the accumulation of physical and human capital.This paper classifies instances of economic growth into four categories—the Malthusian poverty trap,the Lewis dual model of economic development,the Lewis turning point,and Solow neoclassical growth model.This paper conducts empirical analysis of these categories of economic development as they are relevant to Chinese economic growth and discusses policy implications therein.展开更多
To fully make use of information from different representation subspaces,a multi-head attention-based long short-term memory(LSTM)model is proposed in this study for speech emotion recognition(SER).The proposed model ...To fully make use of information from different representation subspaces,a multi-head attention-based long short-term memory(LSTM)model is proposed in this study for speech emotion recognition(SER).The proposed model uses frame-level features and takes the temporal information of emotion speech as the input of the LSTM layer.Here,a multi-head time-dimension attention(MHTA)layer was employed to linearly project the output of the LSTM layer into different subspaces for the reduced-dimension context vectors.To provide relative vital information from other dimensions,the output of MHTA,the output of feature-dimension attention,and the last time-step output of LSTM were utilized to form multiple context vectors as the input of the fully connected layer.To improve the performance of multiple vectors,feature-dimension attention was employed for the all-time output of the first LSTM layer.The proposed model was evaluated on the eNTERFACE and GEMEP corpora,respectively.The results indicate that the proposed model outperforms LSTM by 14.6%and 10.5%for eNTERFACE and GEMEP,respectively,proving the effectiveness of the proposed model in SER tasks.展开更多
Background： Lower body positive pressure （LBPP） treadmills can be used in rehabilitation programs and/or to supplement tun mileage in healthy runners by reducing the effective body weight and impact associated with...Background： Lower body positive pressure （LBPP） treadmills can be used in rehabilitation programs and/or to supplement tun mileage in healthy runners by reducing the effective body weight and impact associated with running. The purpose of this study is to determine if body weight support influences the stride length （SL）-velocity as well as leg impact acceleration relationship during running. Methods： Subjects （n = 10, 21.4 ± 2.0 years, 72.4 ± 10.3 kg, 1.76 ± 0.09 m） completed 16 run conditions consisting of specific body weight support and velocity combinations. Velocities tested were 100%, 110%, 120%, and 130% of the preferred velocity （2.75± 0.36 m/s）. Body weight support conditions consisted of 0, 60%,5, 70%, and 80% body weight support. SL and leg impact accelerations were determined using a light-weight accelerometer mounted on the surface of the anterior-distal aspect of the tibia. A 4 × 4 （velocity x body weight support） repeated measures ANOVA was used for each dependent variable （a = 0.05）. Results： Neither SL nor leg impact acceleration were influenced by the interaction of body weight support and velocity （p 〉 0.05）. SL was least during no body weight support （p 〈 0.05） but not different between 60%, 70%, and 80% support （p 〉 0.05）. Leg impact acceleration was greatest during no body weight support （p 〈 0.05） but not different between 60%, 70%, and 80% support （p 〉 0.05）. SL and leg impact accelerations increased with velocity regardless of support （p 〈 0.05）. Conclusion： The relationships between SL and leg impact accelerations with velocity were not influenced by body weight support.展开更多
基金Supported by National Natural Science Foundation (30660016)~~
文摘[Objective] The aim was to investigate the status of wild Platycodon grandiflourus resources in Changbai Mountain area. [ Method] The habitats and growth environment of the wild Platycodon grandiflourus were investigated, and made a collection of the germplasm resources at 24 sites in Changbai Mountain area. [ Result ] Most of the exiting wild Platycodon grandiflourus survived because of its small roots that means no value and grew in poor or remote conditions that led to much difficulty to dig. The character of wild Platycodon grandiflourus was obviously inferior to cultivated Platycodon grandiflourus. [ Condusion] This study lays a foundation for the genetic diversity analysis and the establishment of germplasm resource pool of wild Platycodon grandflourus in Changbai Mountain Area.
基金The National Natural Science Foundation of China(No.61375028,61301219)China Postdoctoral Science Foundation(No.2012M520973)the Scientific Research Funds of Nanjing Institute of Technology(No.ZKJ201202)
文摘In order to recognize people＇s annoyance emotions in the working environment and evaluate emotional well- being, emotional speech in a work environment is induced to obtain adequate samples of emotional speech, and a Mandarin database with two thousands samples is built. In searching for annoyance-type emotion features, the prosodic feature and the voice quality feature parameters of the emotional statements are extracted first. Then an improved back propagation （BP） neural network based on the shuffled frog leaping algorithm （SFLA） is proposed to recognize the emotion. The recognition capability of the BP, radical basis function （RBF） and the SFLA neural networks are compared experimentally. The results show that the recognition ratio of the SFLA neural network is 4. 7% better than that of the BP neural network and 4. 3% better than that of the RBF neural network. The experimental results demonstrate that the random initial data trained by the SFLA can optimize the connection weights and thresholds of the neural network, speed up the convergence and improve the recognition rate.
文摘Through exploring the limitation of the neoclassical theory of economic growth,which classifies growth as a homogenous process,this paper reconciles various theories of economic development and explains the rises and falls of economic growth under a unified framework,focusing on incentives of the accumulation of physical and human capital.This paper classifies instances of economic growth into four categories—the Malthusian poverty trap,the Lewis dual model of economic development,the Lewis turning point,and Solow neoclassical growth model.This paper conducts empirical analysis of these categories of economic development as they are relevant to Chinese economic growth and discusses policy implications therein.
基金The National Natural Science Foundation of China(No.61571106,61633013,61673108,81871444).
文摘To fully make use of information from different representation subspaces,a multi-head attention-based long short-term memory(LSTM)model is proposed in this study for speech emotion recognition(SER).The proposed model uses frame-level features and takes the temporal information of emotion speech as the input of the LSTM layer.Here,a multi-head time-dimension attention(MHTA)layer was employed to linearly project the output of the LSTM layer into different subspaces for the reduced-dimension context vectors.To provide relative vital information from other dimensions,the output of MHTA,the output of feature-dimension attention,and the last time-step output of LSTM were utilized to form multiple context vectors as the input of the fully connected layer.To improve the performance of multiple vectors,feature-dimension attention was employed for the all-time output of the first LSTM layer.The proposed model was evaluated on the eNTERFACE and GEMEP corpora,respectively.The results indicate that the proposed model outperforms LSTM by 14.6%and 10.5%for eNTERFACE and GEMEP,respectively,proving the effectiveness of the proposed model in SER tasks.
文摘Background： Lower body positive pressure （LBPP） treadmills can be used in rehabilitation programs and/or to supplement tun mileage in healthy runners by reducing the effective body weight and impact associated with running. The purpose of this study is to determine if body weight support influences the stride length （SL）-velocity as well as leg impact acceleration relationship during running. Methods： Subjects （n = 10, 21.4 ± 2.0 years, 72.4 ± 10.3 kg, 1.76 ± 0.09 m） completed 16 run conditions consisting of specific body weight support and velocity combinations. Velocities tested were 100%, 110%, 120%, and 130% of the preferred velocity （2.75± 0.36 m/s）. Body weight support conditions consisted of 0, 60%,5, 70%, and 80% body weight support. SL and leg impact accelerations were determined using a light-weight accelerometer mounted on the surface of the anterior-distal aspect of the tibia. A 4 × 4 （velocity x body weight support） repeated measures ANOVA was used for each dependent variable （a = 0.05）. Results： Neither SL nor leg impact acceleration were influenced by the interaction of body weight support and velocity （p 〉 0.05）. SL was least during no body weight support （p 〈 0.05） but not different between 60%, 70%, and 80% support （p 〉 0.05）. Leg impact acceleration was greatest during no body weight support （p 〈 0.05） but not different between 60%, 70%, and 80% support （p 〉 0.05）. SL and leg impact accelerations increased with velocity regardless of support （p 〈 0.05）. Conclusion： The relationships between SL and leg impact accelerations with velocity were not influenced by body weight support.
