Changes in the water cycle on the Tibetan Plateau(TP)have a significant impact on local agricultural production and livelihoods and its downstream regions.Against the background of widely reported warming and wetting,...Changes in the water cycle on the Tibetan Plateau(TP)have a significant impact on local agricultural production and livelihoods and its downstream regions.Against the background of widely reported warming and wetting,the hydrological cycle has accelerated and the likelihood of extreme weather events and natural disasters occurring(i.e.,snowstorms,floods,landslides,mudslides,and ice avalanches)has also intensified,especially in the highelevation mountainous regions.Thus,an accurate estimation of the intensity and variation of each component of the water cycle is an urgent scientific question for the assessment of plateau environmental changes.Following the transformation and movement of water between the atmosphere,biosphere and hydrosphere,the authors highlight the urgent need to strengthen the three-dimensional comprehensive observation system(including the eddy covariance system;planetary boundary layer tower;profile measurements of temperature,humidity,and wind by microwave radiometers,wind profiler,and radiosonde system;and cloud and precipitation radars)in the TP region and propose a practical implementation plan.The construction of such a three-dimensional observation system is expected to promote the study of environmental changes and natural hazards prevention.展开更多
Laser-accelerated high-flux-intensity heavy-ion beams are important for new types of accelerators.A particle-in-cell program(Smilei) is employed to simulate the entire process of Station of Extreme Light(SEL) 100 PW l...Laser-accelerated high-flux-intensity heavy-ion beams are important for new types of accelerators.A particle-in-cell program(Smilei) is employed to simulate the entire process of Station of Extreme Light(SEL) 100 PW laser-accelerated heavy particles using different nanoscale short targets with a thickness of 100 nm Cr, Fe, Ag, Ta, Au, Pb, Th and U, as well as 200 nm thick Al and Ca. An obvious stratification is observed in the simulation. The layering phenomenon is a hybrid acceleration mechanism reflecting target normal sheath acceleration and radiation pressure acceleration, and this phenomenon is understood from the simulated energy spectrum,ionization and spatial electric field distribution. According to the stratification, it is suggested that high-quality heavy-ion beams could be expected for fusion reactions to synthesize superheavy nuclei. Two plasma clusters in the stratification are observed simultaneously, which suggest new techniques for plasma experiments as well as thinner metal targets in the precision machining process.展开更多
A rotor CFD solver is developed for simulating the aerodynamic interaction phenomenon among rotor, wing and fuselage of a tilt rotor aircraft in its helicopter mode. The unsteady Navier-Stokes equations are discretize...A rotor CFD solver is developed for simulating the aerodynamic interaction phenomenon among rotor, wing and fuselage of a tilt rotor aircraft in its helicopter mode. The unsteady Navier-Stokes equations are discretized in inertial frame and embedded grid system is adopted for describing the relative motion among blades and nacelle/wing/fuselage. A combination of multi-layer embedded grid and 'extended hole fringe' technique is complemented in original grid system to tackle grid assembly difficulties arising from the narrow space among different aerodynamic components, and to improve the interpolation precision by decreasing the cell volume discrepancy among different grid blocks. An overall donor cell searching and automatic hole cutting technique is used for grid assembly, and the solution processes are speeded up by introduction of OpenMP parallel method. Based on this solver, flow fields and aerodynamics of a tilt rotor aircraft in hover are simulated with several rotor collective angles, and the corresponding states of an isolated rotor and rotor/wing/fuselage model are also computed to obtain reference solution. Aerodynamic interference influences among the rotor and wing/fuselage/nacelle are analyzed, and some meaningful conclusions are drawn. (C) 2016 Chinese Society of Aeronautics and Astronautics. Production and hosting by Elsevier Ltd.展开更多
Technologically, multi-layer fluid models are important in understanding fluid-fluid or fluid-nanoparticle interactions and their effects on flow and heat transfer characteristics. However, to the best of the authors...Technologically, multi-layer fluid models are important in understanding fluid-fluid or fluid-nanoparticle interactions and their effects on flow and heat transfer characteristics. However, to the best of the authors＇ knowledge, little attention has been paid to the study of three-layer fluid models with nanofluids. Therefore, a three-layer fluid flow model with nanofluids is formulated in this paper. The governing coupled nonlinear differential equations of the problem are non-dimensionalized by using appropriate fundamental quantities. The resulting multi-point boundary value problem is solved numerically by quasi-linearization and Richardson＇s extrapolation with modified boundary conditions. The effects of the model parameters on the flow and heat transfer are obtained and analyzed. The results show that an increase in the nanoparticle concentration in the base fluid can modify the fluid-velocity at the interface of the two fluids and reduce the shear not only at the surface of the clear fluid but also at the interface between them. That is, nanofluids play a vital role in modifying the flow phenomena. Therefore, one can use nanofluids to obtain the desired qualities for the multi-fluid flow and heat transfer characteristics.展开更多
Seismic response of ground supported baseisolated liquid storage tanks are evaluated under bi-directional earthquakes. The base-isolated liquid storage tanks are modeled using mechanical analogs with two and three lum...Seismic response of ground supported baseisolated liquid storage tanks are evaluated under bi-directional earthquakes. The base-isolated liquid storage tanks are modeled using mechanical analogs with two and three lumped masses (Model 1 and Model 2). Two types of isolation systems, such as sliding system and elastomeric system, are considered for the present study. The isolation systems are modeled using Wen’s equation for hysteretic isolation systems. Response of base-isolated liquid storage tanks, evaluated through two different modeling approaches, is compared. Both the models predict similar sloshing displacement. The effect of interaction between the mutually perpendicular seismic responses of the isolator is investigated for both the models. It is observed that interaction affects the peak seismic response of the base-isolated liquid storage tanks significantly, under the bi-directional earthquake components.展开更多
Discovering new drugs is a complicated,timeconsuming,costly,risky and failure-prone process.However,about 80%of the drugs that have been approved so far are targeted at protein targets,and 99%of them only target speci...Discovering new drugs is a complicated,timeconsuming,costly,risky and failure-prone process.However,about 80%of the drugs that have been approved so far are targeted at protein targets,and 99%of them only target specific proteins.This means that there are still a large number of protein targets that are considered“useless”.By exploring miRNA as a potential therapeutic target,we can expand the range of target selection and improve the efficiency of drug development.Therefore,it is of great significance to search for potential miRNA-drug interactions(MDIs)through reasonable computational methods.In this paper,a dual-channel network model,MDIDCN,based on Temporal Convolutional Network(TCN)and Bi-directional Long Short-Term Memory(BiLSTM),was proposed to predict MDIs.Specifically,we first used a known bipartite network to represent the interaction between miRNAs and drugs,and the graph embedding technique of BiNE was applied to learn the topological features of both.Secondly,we used TCN to learn the MACCS fingerprints of drugs,BiLSTM to learn the k-mer of miRNA,and concatenated the topological and structural features of the two together as their fusion features.Finally,the fusion features of miRNA and drug underwent max-pooling,and they were input into the Softmax layer to obtain the predicted scores of both,so as to obtain the potential miRNA-drug interaction pairs.In this paper,the prediction performance of the model was evaluated on three different datasets by using 5-fold crossvalidation,and the average AUC were 0.9567,0.9365,and 0.8975,respectively.In addition,case studies on the drugs Gemcitabine and hsa-miR-155-5p were also conducted in this paper,and the results showed that the model had high accuracy and reliability.In conclusion,the MDIDCN model can accurately and efficiently predict MDIs,which has important implications for drug development.展开更多
Earthquake performance of a flexible one-story building isolated with a variable frequency pendulum isolator （VFPI） under near-fault and far-field ground motions is investigated. The frictional forces mobilized at t...Earthquake performance of a flexible one-story building isolated with a variable frequency pendulum isolator （VFPI） under near-fault and far-field ground motions is investigated. The frictional forces mobilized at the interface of the VFPI are assumed to be velocity dependent. The interaction between frictional forces of the VFPI in two horizontal directions is duly considered and coupled differential equations of motion of the isolated system in the incremental form are solved iteratively. The response of the system with bi-directional interaction is compared with those without interaction. In addition, the effects of velocity dependence on the response of the isolated system are also investigated. Moreover, a parametric study is carried out to critically examine the influence of important parameters on bi-directional interaction effects of the frictional forces of the VFPI. These parameters are： the superstructure time period, frequency variation factor （FVF） and friction coefficient of the VFPI. From the above investigations, it is observed that the dependence of the friction coefficient on relative velocity of the system does not have a noticeable effect on the peak response of the system isolated with VFPI, and that the bi-directional interaction of frictional forces of the VFPI is important and if neglected, isolator displacements will be underestimated and the superstructure acceleration and base shear will be overestimated.展开更多
Consider the inverse problem of recovering a multi-layered fluid-solid medium from many acoustic measurements corresponding to time-harmonic acoustic plane waves.We prove that both the supports of the embedded solid o...Consider the inverse problem of recovering a multi-layered fluid-solid medium from many acoustic measurements corresponding to time-harmonic acoustic plane waves.We prove that both the supports of the embedded solid obstacle and the surrounding layered fluid medium can be uniquely identified by means of acoustic far-field pattern for all incident wave fields at a fixed frequency.Our proof is based on the constructions of some well-posed partial differential equation systems in sufficiently small domains combined with the a priori estimates for the solutions of the forward scattering problem.展开更多
Bi-directional pedestrian flows are common at crosswalks, footpaths, and shopping areas. However, the properties of pedestrian movement may vary in urban areas according to the type of walking facility. In recent year...Bi-directional pedestrian flows are common at crosswalks, footpaths, and shopping areas. However, the properties of pedestrian movement may vary in urban areas according to the type of walking facility. In recent years,crowd movements at carnival events have attracted the attention of researchers. In contrast to pedestrian behavior in other walking facilities, pedestrians whose attention is attracted by carnival displays or activities may slow down and even stop walking. The Lunar New Year Market is a traditional carnival event in Hong Kong held annually one week before the Lunar New Year. During the said event,crowd movements can be easily identified, particularly in Victoria Park, where the largest Lunar New Year Market in Hong Kong is hosted. In this study, we conducted a videobased observational survey to collect pedestrian flow and speed data at the Victoria Park Lunar New Year Market on the eve of the Lunar New Year. Using the collected data, an extant mathematical model was calibrated to capture the relationships between the relevant macroscopic quantities,thereby providing insight into pedestrian behavior at the carnival event. Bayesian inference was employed to calibrate the model by using prior data obtained from a previous controlled experiment. Results obtained enhance our understanding of crowd behavior under different conditions at carnival events, thus facilitating the improvement of the safety and efficiency of similar events in the future.展开更多
he transition from traditional learning to practice-oriented programming learning will bring learners discomfort.The discomfort quickly breeds negative emotions when encountering programming difficulties,which leads t...he transition from traditional learning to practice-oriented programming learning will bring learners discomfort.The discomfort quickly breeds negative emotions when encountering programming difficulties,which leads the learner to lose interest in programming or even give up.Emotion plays a crucial role in learning.Educational psychology research shows that positive emotion can promote learning performance,increase learning interest and cultivate creative thinking.Accurate recognition and interpretation of programming learners’emotions can give them feedback in time,and adjust teaching strategies accurately and individually,which is of considerable significance to improve effects of programming learning and education.The existing methods of sensor-free emotion prediction include emotion prediction based on keyboard dynamic,mouse interaction data and interaction logs,respectively.However,none of the three studies considered the temporal characteristics of emotion,resulting in low recognition accuracy.For the first time,this paper proposes an emotion prediction model based on time series and context information.Then,we establish a Bi-recurrent neural network,obtain the time sequence characteristics of data automatically,and explore the application of deep learning in the field of Academic Emotion prediction.The results show that the classification ability of this model is much better than that of the original LSTM(Long-Short Term Memory),GRU(Gate Recurrent Unit)and RNN(Re-current Neural Network),and this model has better generalization ability.展开更多
We propose a learning architecture for integrating multi-modal information e.g., vision, audio information. In recent years, artificial intelligence (AI) is making major progress in key tasks like a language, vision, ...We propose a learning architecture for integrating multi-modal information e.g., vision, audio information. In recent years, artificial intelligence (AI) is making major progress in key tasks like a language, vision, voice recognition tasks. Most studies focus on how AI could achieve human-like abilities. Especially, in human-robot interaction research field, some researchers attempt to make robots talk with human in daily life. The key challenges for making robots talk naturally in conversation are to need to consider multi-modal non-verbal information same as human, and to learn with small amount of labeled multi-modal data. Previous multi-modal learning needs a large amount of labeled data while labeled multi-modal data are shortage and difficult to be collected. In this research, we address these challenges by integrating single-modal classifiers which trained each modal information respectively. Our architecture utilized knowledge by using bi-directional associative memory. Furthermore, we conducted the conversation experiment for collecting multi-modal non-verbal information. We verify our approach by comparing accuracies between our system and conventional system which trained multi-modal information.展开更多
基金This research was jointly funded by the Second Tibetan Plateau Scientific Expedition and Research Program(Grant Nos.2019QZKK0103 and 2019QZKK0105)the National Natural Science Foundation of China(Grant Nos.91837208 and 42075085).
文摘Changes in the water cycle on the Tibetan Plateau(TP)have a significant impact on local agricultural production and livelihoods and its downstream regions.Against the background of widely reported warming and wetting,the hydrological cycle has accelerated and the likelihood of extreme weather events and natural disasters occurring(i.e.,snowstorms,floods,landslides,mudslides,and ice avalanches)has also intensified,especially in the highelevation mountainous regions.Thus,an accurate estimation of the intensity and variation of each component of the water cycle is an urgent scientific question for the assessment of plateau environmental changes.Following the transformation and movement of water between the atmosphere,biosphere and hydrosphere,the authors highlight the urgent need to strengthen the three-dimensional comprehensive observation system(including the eddy covariance system;planetary boundary layer tower;profile measurements of temperature,humidity,and wind by microwave radiometers,wind profiler,and radiosonde system;and cloud and precipitation radars)in the TP region and propose a practical implementation plan.The construction of such a three-dimensional observation system is expected to promote the study of environmental changes and natural hazards prevention.
基金support from the Strategic Priority Research Program of the Chinese Academy of Sciences (No.XDB34030000)the National Key R & D Program of China (No.2022YFA1602404)+2 种基金National Natural Science Foundation of China (No. U1832129)the Youth Innovation Promotion Association of the Chinese Academy of Sciences (No.2017309)the Program for Innovative Research Team (in Science and Technology) in University of Henan Province of China (No.21IRTSTHN011)。
文摘Laser-accelerated high-flux-intensity heavy-ion beams are important for new types of accelerators.A particle-in-cell program(Smilei) is employed to simulate the entire process of Station of Extreme Light(SEL) 100 PW laser-accelerated heavy particles using different nanoscale short targets with a thickness of 100 nm Cr, Fe, Ag, Ta, Au, Pb, Th and U, as well as 200 nm thick Al and Ca. An obvious stratification is observed in the simulation. The layering phenomenon is a hybrid acceleration mechanism reflecting target normal sheath acceleration and radiation pressure acceleration, and this phenomenon is understood from the simulated energy spectrum,ionization and spatial electric field distribution. According to the stratification, it is suggested that high-quality heavy-ion beams could be expected for fusion reactions to synthesize superheavy nuclei. Two plasma clusters in the stratification are observed simultaneously, which suggest new techniques for plasma experiments as well as thinner metal targets in the precision machining process.
文摘A rotor CFD solver is developed for simulating the aerodynamic interaction phenomenon among rotor, wing and fuselage of a tilt rotor aircraft in its helicopter mode. The unsteady Navier-Stokes equations are discretized in inertial frame and embedded grid system is adopted for describing the relative motion among blades and nacelle/wing/fuselage. A combination of multi-layer embedded grid and 'extended hole fringe' technique is complemented in original grid system to tackle grid assembly difficulties arising from the narrow space among different aerodynamic components, and to improve the interpolation precision by decreasing the cell volume discrepancy among different grid blocks. An overall donor cell searching and automatic hole cutting technique is used for grid assembly, and the solution processes are speeded up by introduction of OpenMP parallel method. Based on this solver, flow fields and aerodynamics of a tilt rotor aircraft in hover are simulated with several rotor collective angles, and the corresponding states of an isolated rotor and rotor/wing/fuselage model are also computed to obtain reference solution. Aerodynamic interference influences among the rotor and wing/fuselage/nacelle are analyzed, and some meaningful conclusions are drawn. (C) 2016 Chinese Society of Aeronautics and Astronautics. Production and hosting by Elsevier Ltd.
基金supported by the Imam Khomeini International University of Iran(No.751166-91)
文摘Technologically, multi-layer fluid models are important in understanding fluid-fluid or fluid-nanoparticle interactions and their effects on flow and heat transfer characteristics. However, to the best of the authors＇ knowledge, little attention has been paid to the study of three-layer fluid models with nanofluids. Therefore, a three-layer fluid flow model with nanofluids is formulated in this paper. The governing coupled nonlinear differential equations of the problem are non-dimensionalized by using appropriate fundamental quantities. The resulting multi-point boundary value problem is solved numerically by quasi-linearization and Richardson＇s extrapolation with modified boundary conditions. The effects of the model parameters on the flow and heat transfer are obtained and analyzed. The results show that an increase in the nanoparticle concentration in the base fluid can modify the fluid-velocity at the interface of the two fluids and reduce the shear not only at the surface of the clear fluid but also at the interface between them. That is, nanofluids play a vital role in modifying the flow phenomena. Therefore, one can use nanofluids to obtain the desired qualities for the multi-fluid flow and heat transfer characteristics.
文摘Seismic response of ground supported baseisolated liquid storage tanks are evaluated under bi-directional earthquakes. The base-isolated liquid storage tanks are modeled using mechanical analogs with two and three lumped masses (Model 1 and Model 2). Two types of isolation systems, such as sliding system and elastomeric system, are considered for the present study. The isolation systems are modeled using Wen’s equation for hysteretic isolation systems. Response of base-isolated liquid storage tanks, evaluated through two different modeling approaches, is compared. Both the models predict similar sloshing displacement. The effect of interaction between the mutually perpendicular seismic responses of the isolator is investigated for both the models. It is observed that interaction affects the peak seismic response of the base-isolated liquid storage tanks significantly, under the bi-directional earthquake components.
基金supported by the National Natural Science Foundation of China(Grant No.62272288)the Fundamental Research Funds for the Central Universities,and Shaanxi Normal University(GK202302006).
文摘Discovering new drugs is a complicated,timeconsuming,costly,risky and failure-prone process.However,about 80%of the drugs that have been approved so far are targeted at protein targets,and 99%of them only target specific proteins.This means that there are still a large number of protein targets that are considered“useless”.By exploring miRNA as a potential therapeutic target,we can expand the range of target selection and improve the efficiency of drug development.Therefore,it is of great significance to search for potential miRNA-drug interactions(MDIs)through reasonable computational methods.In this paper,a dual-channel network model,MDIDCN,based on Temporal Convolutional Network(TCN)and Bi-directional Long Short-Term Memory(BiLSTM),was proposed to predict MDIs.Specifically,we first used a known bipartite network to represent the interaction between miRNAs and drugs,and the graph embedding technique of BiNE was applied to learn the topological features of both.Secondly,we used TCN to learn the MACCS fingerprints of drugs,BiLSTM to learn the k-mer of miRNA,and concatenated the topological and structural features of the two together as their fusion features.Finally,the fusion features of miRNA and drug underwent max-pooling,and they were input into the Softmax layer to obtain the predicted scores of both,so as to obtain the potential miRNA-drug interaction pairs.In this paper,the prediction performance of the model was evaluated on three different datasets by using 5-fold crossvalidation,and the average AUC were 0.9567,0.9365,and 0.8975,respectively.In addition,case studies on the drugs Gemcitabine and hsa-miR-155-5p were also conducted in this paper,and the results showed that the model had high accuracy and reliability.In conclusion,the MDIDCN model can accurately and efficiently predict MDIs,which has important implications for drug development.
文摘Earthquake performance of a flexible one-story building isolated with a variable frequency pendulum isolator （VFPI） under near-fault and far-field ground motions is investigated. The frictional forces mobilized at the interface of the VFPI are assumed to be velocity dependent. The interaction between frictional forces of the VFPI in two horizontal directions is duly considered and coupled differential equations of motion of the isolated system in the incremental form are solved iteratively. The response of the system with bi-directional interaction is compared with those without interaction. In addition, the effects of velocity dependence on the response of the isolated system are also investigated. Moreover, a parametric study is carried out to critically examine the influence of important parameters on bi-directional interaction effects of the frictional forces of the VFPI. These parameters are： the superstructure time period, frequency variation factor （FVF） and friction coefficient of the VFPI. From the above investigations, it is observed that the dependence of the friction coefficient on relative velocity of the system does not have a noticeable effect on the peak response of the system isolated with VFPI, and that the bi-directional interaction of frictional forces of the VFPI is important and if neglected, isolator displacements will be underestimated and the superstructure acceleration and base shear will be overestimated.
基金supported by the National Natural Science Foundation of China(Nos.11871416,11771349,91730306)Shandong Provincial Natural Science Foundation(Nos.ZR2019MA027,ZR2017MA044)。
文摘Consider the inverse problem of recovering a multi-layered fluid-solid medium from many acoustic measurements corresponding to time-harmonic acoustic plane waves.We prove that both the supports of the embedded solid obstacle and the surrounding layered fluid medium can be uniquely identified by means of acoustic far-field pattern for all incident wave fields at a fixed frequency.Our proof is based on the constructions of some well-posed partial differential equation systems in sufficiently small domains combined with the a priori estimates for the solutions of the forward scattering problem.
基金supported by a grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (Project No. Poly U 5243/13E)respectively supported by the Postdoctoral Fellow Scheme and Francis S. Y. Bong Professorship in Engineering of The University of Hong Kong
文摘Bi-directional pedestrian flows are common at crosswalks, footpaths, and shopping areas. However, the properties of pedestrian movement may vary in urban areas according to the type of walking facility. In recent years,crowd movements at carnival events have attracted the attention of researchers. In contrast to pedestrian behavior in other walking facilities, pedestrians whose attention is attracted by carnival displays or activities may slow down and even stop walking. The Lunar New Year Market is a traditional carnival event in Hong Kong held annually one week before the Lunar New Year. During the said event,crowd movements can be easily identified, particularly in Victoria Park, where the largest Lunar New Year Market in Hong Kong is hosted. In this study, we conducted a videobased observational survey to collect pedestrian flow and speed data at the Victoria Park Lunar New Year Market on the eve of the Lunar New Year. Using the collected data, an extant mathematical model was calibrated to capture the relationships between the relevant macroscopic quantities,thereby providing insight into pedestrian behavior at the carnival event. Bayesian inference was employed to calibrate the model by using prior data obtained from a previous controlled experiment. Results obtained enhance our understanding of crowd behavior under different conditions at carnival events, thus facilitating the improvement of the safety and efficiency of similar events in the future.
基金supported by the 2018-2020 Higher Education Talent Training Quality and Teaching Reform Project of Sichuan Province(Grant No.JG2018-46)the Science and Technology Planning Program of Sichuan University and Luzhou(Grant No.2017CDLZG30)the Postdoctoral Science fund of Sichuan University(Grant No.2019SCU12058).
文摘he transition from traditional learning to practice-oriented programming learning will bring learners discomfort.The discomfort quickly breeds negative emotions when encountering programming difficulties,which leads the learner to lose interest in programming or even give up.Emotion plays a crucial role in learning.Educational psychology research shows that positive emotion can promote learning performance,increase learning interest and cultivate creative thinking.Accurate recognition and interpretation of programming learners’emotions can give them feedback in time,and adjust teaching strategies accurately and individually,which is of considerable significance to improve effects of programming learning and education.The existing methods of sensor-free emotion prediction include emotion prediction based on keyboard dynamic,mouse interaction data and interaction logs,respectively.However,none of the three studies considered the temporal characteristics of emotion,resulting in low recognition accuracy.For the first time,this paper proposes an emotion prediction model based on time series and context information.Then,we establish a Bi-recurrent neural network,obtain the time sequence characteristics of data automatically,and explore the application of deep learning in the field of Academic Emotion prediction.The results show that the classification ability of this model is much better than that of the original LSTM(Long-Short Term Memory),GRU(Gate Recurrent Unit)and RNN(Re-current Neural Network),and this model has better generalization ability.
文摘We propose a learning architecture for integrating multi-modal information e.g., vision, audio information. In recent years, artificial intelligence (AI) is making major progress in key tasks like a language, vision, voice recognition tasks. Most studies focus on how AI could achieve human-like abilities. Especially, in human-robot interaction research field, some researchers attempt to make robots talk with human in daily life. The key challenges for making robots talk naturally in conversation are to need to consider multi-modal non-verbal information same as human, and to learn with small amount of labeled multi-modal data. Previous multi-modal learning needs a large amount of labeled data while labeled multi-modal data are shortage and difficult to be collected. In this research, we address these challenges by integrating single-modal classifiers which trained each modal information respectively. Our architecture utilized knowledge by using bi-directional associative memory. Furthermore, we conducted the conversation experiment for collecting multi-modal non-verbal information. We verify our approach by comparing accuracies between our system and conventional system which trained multi-modal information.
