• Journal of Information Processing Systems
• /
• v.17 no.4
• /
• pp.754-771
• /
• 2021

In the task of continuous dimension emotion recognition, the parts that highlight the emotional expression are not the same in each mode, and the influences of different modes on the emotional state is also different. Therefore, this paper studies the fusion of the two most important modes in emotional recognition (voice and visual expression), and proposes a two-mode dual-modal emotion recognition method combined with the attention mechanism of the improved AlexNet network. After a simple preprocessing of the audio signal and the video signal, respectively, the first step is to use the prior knowledge to realize the extraction of audio characteristics. Then, facial expression features are extracted by the improved AlexNet network. Finally, the multimodal attention mechanism is used to fuse facial expression features and audio features, and the improved loss function is used to optimize the modal missing problem, so as to improve the robustness of the model and the performance of emotion recognition. The experimental results show that the concordance coefficient of the proposed model in the two dimensions of arousal and valence (concordance correlation coefficient) were 0.729 and 0.718, respectively, which are superior to several comparative algorithms.

• The KIPS Transactions:PartC
• /
• v.12C no.7 s.103
• /
• pp.1047-1056
• /
• 2005

The existing data broadcasting system has the mechanism that the receiving device should download repetitively the entire broadcasting application data until it acquires the whole of the application data. This mechanism takes long time for the download. To solve this problem this paper proposes a scheme that the device effectively downloads the missing parts of broadcasting application data through the return channel and shows the performance evaluations of the proposed scheme with respect to the experiment.

• Journal of The Korean Astronomical Society
• /
• v.46 no.2
• /
• pp.93-96
• /
• 2013

Modified Newtonian Dynamics (MOND) is a possible solution for the missing mass problem in galactic dynamics; its predictions are in good agreement with observations in the limit of weak accelerations. However, MOND does not derive from a physical mechanism and does not make predictions on the transitional regime from Newtonian to modified dynamics; rather, empirical transition functions have to be constructed from the boundary conditions and comparisons with observations. I compare the formalism of classical MOND to the scaling law derived from a toy model of gravity based on virtual massive gravitons (the "graviton picture") which I proposed recently. I conclude that MOND naturally derives from the "graviton picture" at least for the case of non-relativistic, highly symmetric dynamical systems. This suggests that-to first order-the "graviton picture" indeed provides a valid candidate for the physical mechanism behind MOND and gravity on galactic scales in general.

• Korean Journal of School Psychology
• /
• v.17 no.3
• /
• pp.243-261
• /
• 2020

Fear of missing out (FoMO) refers to the desire to stay continually connected with what others are doing because they are afraid of missing the flow or not being included. Although FoMO has been recently proposed as a widespread phenomenon within social media, there has been few empirical research by which mechanism FoMO leads to social networking service (SNS) addiction. Therefore, this study examined whether FoMO mediated the relationship between adolescents' attachment and SNS addiction proneness. A total of 517 middle and high school students participated in the study and completed self-report questionnaires. Results showed that FoMO partially mediated the relationship between adolescents' parental attachment and SNS addiction proneness, whereas the relationship between adolescents' peer attachment and SNS addiction proneness was fully mediated by FoMO. These results suggest that unstable attachment might contribute to SNS addiction through FoMO, especially in case of peer attachment. Lastly, we discussed theoretical and practical implications of this study and suggestions for future research.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11a
• /
• pp.403-405
• /
• 2005

무선 센서 네트워크에서 이동하는 물체를 에너지 효율적으로 추적하기 위하여 많은 연구가 진행되고 있다. 그 중 대표적인 것은 물체의 이동에 따라 동적으로 클러스터링을 구성해 나가는 방법이다. 물체의 이동에 따라 클러스터를 구성한 후 클러스터 내부에서는 모든 센서 노드들이 연속적으로 물체를 모니터링하거나 혹은 일반적인 스케줄링 기법을 사용하여 에너지 소모를 분산시킨다. 이런 스케줄링 기법들은 환경 모니터링 등 일반적인 센서 네트워크를 대상으로 개발되고 있기 때문에 이동하는 물체를 추적하는 응용에서는 적합하지 않다. 본 논문에서는 물체의 이동경로를 따른 동적 클러스터링 환경에서 물체의 이동 정보를 고려한 클러스터 내부에서의 스케줄링 기법을 제안함으로써 이동하는 물체에 대한 missing-rate를 최소화하는 동시에 에너지 소모를 최대한 줄임으로써 전체 센서 네트워크의 생명주기를 연장시키고자 한다. 시뮬레이션 결과가 증명하는 바와 같이 제안한 방안은 보다 낮은 에너지 소모와 missing-rate를 달성하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.06a
• /
• pp.360-363
• /
• 2007

In sensor network, the importance of transporting data with reliability is growing gradually to support communications. Data flow from sink to nodes needs reliability for the control or management, that is very sensitive and intolerable, however relatively, data flow from nodes to sink is tolerable. In this paper, with emphasis of the data flow from sink to nodes, we proposed the mechanism that establishes confidence interval for transport. Establishing confidence interval hop-by-hop, not end to end, if errors happen or there's missing data, this mechanism recovers them with selective acknowledgement using fixed window. In addition, this mechanism supports traffic congestion control depending on the buffer condition. Through the simulation, we showed that this mechanism has an excellent performance for error recovery in sensor network.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.6
• /
• pp.1190-1194
• /
• 2007

In sensor network, the importance of transporting data with reliability is growing gradually to support communications. Data flow from sink to nodes needs reliability for the control or management, that is very sensitive and intolerable, however relatively, data flow from nodes to sink is tolerable. In this paper, with emphasis of the data flow from sink to nodes, we proposed the mechanism that establishes confidence interval for transport. Establishing confidence interval herby-hop, not end to end, if errors happen or there's missing data, this mechanism recovers them with selective acknowledgement using fixed window. In addition, this mechanism supports franc congestion control depending on the buffer condition. Through the simulation, we showed that this mechanism has an excellent performance for error recovery in sensor network.

• The Korean Journal of Applied Statistics
• /
• v.32 no.4
• /
• pp.543-559
• /
• 2019

Data often include missing values due to various reasons. If the missing data mechanism is not MCAR, analysis based on fully observed cases may an estimation cause bias and decrease the precision of the estimate since partially observed cases are excluded. Especially when data include many variables, missing values cause more serious problems. Many imputation techniques are suggested to overcome this difficulty. However, imputation methods using parametric models may not fit well with real data which do not satisfy model assumptions. In this study, we review imputation methods using nonlinear models such as kernel, resampling, and spline methods which are robust on model assumptions. In addition, we suggest utilizing imputation classes to improve imputation accuracy or adding random errors to correctly estimate the variance of the estimates in nonlinear imputation models. Performances of imputation methods using nonlinear models are compared under various simulated data settings. Simulation results indicate that the performances of imputation methods are different as data settings change. However, imputation based on the kernel regression or the penalized spline performs better in most situations. Utilizing imputation classes or adding random errors improves the performance of imputation methods using nonlinear models.

• Journal of agriculture & life science
• /
• v.51 no.1
• /
• pp.187-192
• /
• 2017

In various crops, seedlings are preferred to seeds for faster and more effective growth, so transplanters are used for them. This 2-row transplanter was developed to promote the mechanization of vegetable transplanting work which depends on human power. and it can automatically supply seedling tray and transport picked up seedling to the planting hopper. Also, we judged performance of transplanter with comparing seedling missing plant ration according to two types of pick-up method. Result of experiment, in finger-type picking up of 265 seedlings, missing plant ratio was 13.7% with 17 failures of pick-up and 15 collapse of bed soil. and In fork-type picking up of 200 seedlings, failure of pick-up was not appeared and missing plant ratio was low as 4% with 6 dropped during transfer. Therefore for 2-row automatic transplanter, fork-type pick-up device was found to be compatible.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.9A
• /
• pp.838-849
• /
• 2005

Fast handover protocol provides seamless handover in mobile Ipv6 networks by reducing handover latency. This paper proposes FMIP-M mechanism to apply this advantage of fast handover protocol to multicast service. The FMIP-M is a fast handover mechanism that supports a reliable multicast service in mobile Ipv6 networks. When hosts move other networks, they may have seine data missing and out-of-synch problems of multicast data. The proposed mechanism provides reliable multicast transmission by compensating data losses from the previous AR. Also it provides an additional function that is able to change multicasting service types dynamically in accordance with network status related with multicasting. So that it is able to make multicast paths very close to the optimum and more efficient multicast service is possible. The performance of the proposed mechanism is evaluated by a time analysis and simulations in various conditions.