• Healthcare Informatics Research
• /
• v.24 no.4
• /
• pp.309-316
• /
• 2018

Objectives: Both the valence and arousal components of affect are important considerations when managing mental healthcare because they are associated with affective and physiological responses. Research on arousal and valence analysis, which uses images, texts, and physiological signals that employ deep learning, is actively underway; research investigating how to improve the recognition rate is needed. The goal of this research was to design a deep learning framework and model to classify arousal and valence, indicating positive and negative degrees of emotion as high or low. Methods: The proposed arousal and valence classification model to analyze the affective state was tested using data from 40 channels provided by a dataset for emotion analysis using electrocardiography (EEG), physiological, and video signals (the DEAP dataset). Experiments were based on 10 selected featured central and peripheral nervous system data points, using long short-term memory (LSTM) as a deep learning method. Results: The arousal and valence were classified and visualized on a two-dimensional coordinate plane. Profiles were designed depending on the number of hidden layers, nodes, and hyperparameters according to the error rate. The experimental results show an arousal and valence classification model accuracy of 74.65 and 78%, respectively. The proposed model performed better than previous other models. Conclusions: The proposed model appears to be effective in analyzing arousal and valence; specifically, it is expected that affective analysis using physiological signals based on LSTM will be possible without manual feature extraction. In a future study, the classification model will be adopted in mental healthcare management systems.

• The Korean Journal of Applied Statistics
• /
• v.32 no.6
• /
• pp.899-908
• /
• 2019

Massive in-depth neural networks with numerous parameters are powerful machine learning methods, but they have overfitting problems due to the excessive flexibility of the models. Dropout is one methods to overcome the problem of oversized neural networks. It is also an effective method that randomly drops input and hidden nodes from the neural network during training. Every sample is fed to a thinned network from an exponential number of different networks. In this study, instead of feeding one sample for each thinned network, two or more samples are used in fitting for one thinned network known as a Hybrid Dropout. Simulation results using real data show that the new method improves the stability of estimates and reduces the minimum error for the verification data.

• The KIPS Transactions:PartD
• /
• v.14D no.2
• /
• pp.145-156
• /
• 2007

Recently, the needs for effective real time change detection algorithms for XML/HTML documents and increased in such fields as the detection of defacement attacks to web documents, the version management, and so on. Especially, those applications of real time change detection for large number of XML/HTML documents require fast heuristic algorithms to be used in real time environment, instead of algorithms which compute minimal cost-edit scripts. Existing heuristic algorithms are fast in execution time, but do not provide satisfactory edit script. In this paper, we present existing algorithms XyDiff and X-tree Diff, analyze their problems and propose algorithm X-tree Diff which improve problems in existing ones. X-tree Diff+ has similar performance in execution time with existing algorithms, but it improves matching ratio between nodes from two documents by refining matching process based on the notion of efficiency of matching.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.6B
• /
• pp.338-348
• /
• 2005

Currently, IEEE 802.11 Wireless LAN (WLAN) is rising as the most popular means for the broadband wireless access network. In this thesis, we propose a QoS(Quality of Service)-Oriented mechanism using handshaking method of scanning phase in IEEE 802.11 handoff. In conventional process for handoff, the major criterion to select the best AP(Access Point) among candidates is normally based on the RSS(Received Signal Strength), which does not always make the selected network guarantee the maximum achievable performance. Even though the link quality with a neighboring AP is excellent, the AP may not be a good candidate to handoff to simply because MAC(Medium Access Control) protocol of IEEE 802.11 standard is contention-based such as CSMA/CA. Therefore, if we apply network level information to AP selection criteria, we can achieve better handoff efficiency rather than before. The analysis and simulation results applied to our new mechanism show clearly better performance than AP selection based on traditional handoff method.

• Journal of KIISE:Information Networking
• /
• v.35 no.4
• /
• pp.345-354
• /
• 2008

Recently, small-scale mobile network which is composed of many mobile devices in a man becomes popular. Also, Examples of large-scale mobile network can be thought access networks deployed on public transportation such as ships, trains and buses. To provide seamless mobility for mobile nodes in this mobile network, binding update messages must be exchanged frequently. However, it incurs network overhead increasingly and decreases energy efficiency of mobile router. If we try to reduce the number of the messages to cope with the problem, it may happen the security -related problems conversely Thus, mobile router needs a effective algorithm to update location information with low cost and to cover security problems. In this paper, mobility management scheme based on mobile router's mobility pattern is proposed. Whenever each mobile router leaves a visiting network, it records related information as moving log. And then it periodically computes mean resident time for all visited network, and saves them in the profile. If each mobile router moves into the visited network hereafter, the number of binding update messages can be reduced since current resident time may be expected based on the profile. At this time, of course, security problems can happen. The problems, however, are solved using key credit, which just sends some keys once. Through extensive experiments, bandwidth usages are measured to compare binding update messages in proposed scheme with that in existing scheme. From the results, we can reduce about 65% of mobility-management-related messages especially when mobile router stays more than 50 minutes in a network. Namely, the proposed scheme improves network usage and energy usage of mobile router by decreasing the number of messages and authorization procedure.

• Journal of Internet Computing and Services
• /
• v.9 no.5
• /
• pp.153-163
• /
• 2008

Almost literature has so far proposed the modifications to IEEE 802.11 DCF MAC protocol to improve system throughput and average delay. Jitter(variance in packet delays) is, however, considered as an important QoS parameter for real-time streaming service which is popularized increasingly. Unfortunately, legacy access protocol for WLAN can't support the jitter. This paper proposes two schemes modifying the DCF MAC protocol to reduce the average value of jitter for multimedia frames arrived at received nodes. The one scheme, referred to as DCW-DCF(Decreasing Contention Window-DCF), reduces CW(Contention Window) size by half whenever a transmitted frame collided against other frames and back-off happens. Also, it increase initial CW size according as the link utilization and the collision probability increase. The other scheme, referred to as D-DIFS(Differential-DIFS), assigns different values to DIFS. That is, the DIFS interval is deceased as the number of back-off for a frame increases. In this paper, the two schemes are evaluated through simulation using ns2 and simulation results show that the proposed schemes are effective for improving the jitter property of standard WLAN.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.6
• /
• pp.1433-1442
• /
• 1996

Data distribution of SPMD(Single Program Multiple Data) pattern is one of main features of HPF (High Performance Fortran). This paper describes design is sues for such data distribution and its efficient execution model on TICOM IV computer, named SPAX(Scalable Parallel Architecture computer based on X-bar network). SPAX has a hierarchical clustering structure that uses distributed shared memory(DSM). In such memory structure, it cannot make a full system utilization to apply unanimously either SMDD(shared Memory Data Distribution) or DMDD(Distributed Memory Data Distribution). Here we propose another data distribution model, called DSMDD(Distributed Shared Memory Data Distribution), a data distribution model based on hierarchical masters-slaves scheme. In this model, a remote master and slaves are designated in each node, shared address scheme is used within a node and message passing scheme between nodes. In our simulation, assuming a node size in which system performance degradation is minimized,DSMDD is more effective than SMDD and DMDD. Especially,the larger number of logical processors and the less data dependency between distributed data,the better performace is obtained.

• Korean Journal of Plant Resources
• /
• v.26 no.2
• /
• pp.303-309
• /
• 2013

This study was performed to clarify the effect of medium compositions and plant growth regulators on the shoot, root formation and growth of 'Jarang' grape for mass propagation of virus-free plant. The formation and growth of shoot were considerably favorable in half-concentration of MS medium. However, the formation of adventitious root per explants (avg. 2.1) was effective in higher concentration (two times) of MS medium. For sucrose concentration, 1% for the shoot formation, 3% for the adventitious root formation and 1% for the growth were observed as yield significant results. With the addition of 0.05% of activated carbon, the shoot growth was improved, and it was effective for the adventitious root formation and growth as well. A pH of 6.8 in the medium was the most suitable for mass propagation; the results showed significant enhancement in the number of nodes and the length of the shoot, 3.9 and 1.3 cm, respectively. The shoot growth was the most vigorous in BA 1.0 mg/L due to the impact of the growth regulator on the mass propagation in it. Consequently, 16.9 shoots per explant were formed in NAA 1.0 mg/L so good results were obtained.

• The KIPS Transactions:PartC
• /
• v.16C no.6
• /
• pp.783-790
• /
• 2009

The necessity of wireless sensor networks is increasing in the recent years. So many researches are studied in wireless sensor networks. The clustering algorithm provides an effective way to prolong the lifetime of the wireless sensor networks. The one-hop routing of LEACH algorithm is an inefficient way in the energy consumption of cluster-head, because it transmits a data to the BS(Base Station) with one-hop. On the other hand, other clustering algorithms transmit data to the BS with multi-hop, because the multi-hop transmission is an effective way. But the multi-hop routing of other clustering algorithms which transmits data to BS with multi-hop have a data bottleneck state problem. The unequal clustering algorithm solved a data bottleneck state problem by increasing the routing path. Most of the unequal clustering algorithms partition the nodes into clusters of unequal size, and clusters closer to the BS have small-size the those farther away from the BS. However, the energy consumption of cluster-head in unequal clustering algorithm is more increased than other clustering algorithms. In the thesis, I propose an energy efficient unequal clustering algorithm which decreases the energy consumption of cluster-head and solves the data bottleneck state problem. The basic idea is divided a three part. First of all I provide that the election of appropriate cluster-head. Next, I offer that the decision of cluster-size which consider the distance from the BS, the energy state of node and the number of neighborhood node. Finally, I provide that the election of assistant node which the transmit function substituted for cluster-head. As a result, the energy consumption of cluster-head is minimized, and the energy consumption of total network is minimized.

• Korean Journal of Veterinary Research
• /
• v.38 no.4
• /
• pp.793-802
• /
• 1998

Salmonellosis caused by a number of serotypes of Salmonella is an infectious, acute or chronic, zoonotic disease and characterized by enteritis and diarrhea, septicemia in animal. In these studies we investigated the prevalent serotypes of Salmonella causing animal salmonellosis in Korea and the 71 strains of Salmonella spp. were isolated from materials such as mesenteric lymph nodes, fecal samples from slaughtered animal. With the identification test results, the most prevalent serotypes were, in order, S stanley 31 strains (43.7%), S typhimurium 19 strains (26.8%) and S montevideo 11 strains (15.5%), respectively. And we could establish the method for detection of antibodies to broad variety of Salmonella serotypes. Lipopolysaccharide(LPS) antigen extracted from Salmonella was more sensitive and specific than outer membrane protein antigen from that for detection of Salmonella antibody by using an indirect ELISA. The optimal concentration of antigen was 100ng/ml of LPS, the dilutions of conjugate and serum were 1 : 1,000~2,000 and 1 : 200~400, respectively. The mix LPS-ELISA which was used by mixing LPS from S typhimurium (group B), S choleraesuis (group C) and S enteritidis (group D) were more rapid and effective than that used LPS from individual strain for detection of Salmonella serogroup O4, O7 and O9 antibody at the same time. We could obtain the high values of optical density ($0.73{\pm}0.32$) by mix LPS-ELISA on the farm which had occurred salmonellosis, but very low values of $0.17{\pm}0.06$ on the negative farm of salmonellosis. So, the mix LPS-ELISA may be used to monitor the serological surveillance for the presence of infection with a number of serotypes of Salmonella and would be useful for prevention and control of salmonellosis in animal.