• Korean Journal of Cognitive Science
• /
• v.28 no.4
• /
• pp.299-314
• /
• 2017

Predicting compound-protein interactions in-silico is significant for the drug discovery. In this paper, we propose an scalable machine learning model to predict compound-protein interaction. The key idea of this scalable machine learning model is the architecture of pairwise neural network model and feature embedding method from the raw data, especially for protein. This method automatically extracts the features without additional knowledge of compound and protein. Also, the pairwise architecture elevate the expressiveness and compact dimension of feature by preventing biased learning from occurring due to the dimension and type of features. Through the 5-fold cross validation results on large scale database show that pairwise neural network improves the performance of predicting compound-protein interaction compared to previous prediction models.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.7
• /
• pp.646-656
• /
• 2008

In this paper, the prototype of haptic-based NVE (networked virtual environment) QoE (quality of experience) is proposed. The proposed framework plays a role of providing users with realistic feeling by managing the system and network resources efficiently under time-varying networks and heterogeneous systems. Therefore, first IoPH (importance of presence for haptic interaction) is defined to quantitate the sense of real that the users feel now. Then we define the haptic-based NVE components required to satisfy the haptic interaction QoE requirements. Finally, QoE adaptation scheme is suggested, which adapts the haptic-based NVE components to current network and system constraints for better haptic interaction quality.

• Journal of Society of Preventive Korean Medicine
• /
• v.17 no.1
• /
• pp.1-11
• /
• 2013

Objectives : The aim of this research is to produce information by gathering up the data on the interaction between medicinal herbs which lie scattered in oriental medical books, and to provide people with easy access to the information by visualizing it. Methods : For this purpose, this study established the fundamental data by organizing the patterns of interaction into some kinds after selecting a part of Bonchogangmok(本草綱目) and extracting its text. In addition, in an effort to visualize the data, the study converted the data into 'net' file and visualized the interaction between medicinal herbs on Pajek. The visualization was done targeting a total of three patterns, such as 1 medicinal herb, 2 medicinal herbs, and 1 prescription. With the data on 'Chinese Lacquer(乾漆)' for 1 medicinal herb, data on 'Licorice(甘草)' and 'Chinese Lacquer(乾漆)' for 2 medicinal herbs, and data on 'Iijin-tang(二陳湯)' for prescription, the research conducted the analysis of the network using 'Kamada-Kawaii Algorithm' on Pajek. Results : As a result of the analysis, it was possible to see the meanings at a single glance as the scattered and fractional meanings were integrated with focus on medicinal herbs, but the increasing number of analyzed medicinal herbs tended to more and more complicate their relationships, thus, requiring additional work like filtering. Conclusions : Such results are fairly applicable in on-line database, and it is judged that if further research expands its scope to include systematic classification of medicinal herbs or cover other medical books than Bonchogangmok, it will create more objective, abundant information.

• The KIPS Transactions:PartB
• /
• v.15B no.3
• /
• pp.219-228
• /
• 2008

For real-time social networking service, mobile interaction in a usable-unified-ubiquitous (U3) web service was studied. Both as a convenient mobile HCI for real-time social networks and as indexing keys to metadata information in ubiquitous web service, the multi-lingual single-character domain names (e.g. 김.net, 이.net, 가.net, ㄱ.net, ㄴ.net, ㅎ.net, ㅏ.net, ㅔ.net, ㄱ.com, ㅎ.com) are convenient mobile interfaces when searching for social information and registering information. We introduce the sketched design goals and experience of mobile interaction in Korea, Japan and China, with the implementation of real-time social networking service as an example of U3 Web service. We also introduce the possibility of extending the application to the metadata directory service in IP-USN (IP-based Ubiquitous Sensor Network) for a unified information management in the service of social networking and sensor networking.

• Korean Journal of Social Welfare
• /
• v.62 no.2
• /
• pp.279-306
• /
• 2010

This study is to compare and analyze the policy decision process between Korea Long-term Care Insurance and Japan's using policy network theory as an analysis tool, and to lead political and theoretical implications. The result of the study is summarized as follows. First, a policy agenda about Long-term Care problem for the Elderly set up by the government-leading both Korea and Japan. and the number of policy participation(actors) increased to characteristics of policy process stage. but there is a difference between Korea and Japan in a background of setting up policy agenda about Long-term Care problem. Second, interaction among policy actors is corporative from early policy agenda establishment stage in Korea and Japan. but it changes to dissenting or critical rapidly as being announced the frame of system to the people. Also, it shows that main policy actors play a role in policy making decision and the connection structures of network is similar to both countries although it has a difference in the interaction frequency. Fourth, although the number of policy actors and the extent of their opinion reflection to the policy effect are different, it shows that policy network pattern is very similar to from policy agenda setting stage to parliament(The National Assemble) policy decision stage as a result of government-leading policy process. The theoretical and political implications of this study are as follows. The number and the variety of policy participation, the importance of establishing opened-interaction system, lots of limitation of policy making-decision process of Japan Long-term Care system, and exclusively the government-leading network has demerits about the reflection of too much government's opinion to the policy outcomes.

• Asia pacific journal of information systems
• /
• v.6 no.1
• /
• pp.39-74
• /
• 1996

Most of the research on a group decision support system [GDSS] has focused on directly examining its effect on the decision outcomes. Under this research framework, however, the role of group interaction process is largely ignored. This study focuses on the effect of the group interaction process on decision-making performance when a GDSS is used as the only medium for group interaction. Specifically, this study sought to determine whether significant relationships exist between the quality of the decision and the decision functions, contingent phases, and different decision paths. Natural interaction processes of decision -making groups was simulated in an experimental setting in which volunteer subjects from several business classes were assigned to dispersed three-person groups undertook the experimental task via a decision network. A baseline GDSS was developed for this setting. The results of this study confirmed earlier studies in a non - GDSS setting to suggest significant effects of decision functions and contingent phases on the quality of decision but no significant relationship between decision path and the quality of group decision.

• Proceedings of the Korea Contents Association Conference
• /
• 2014.06a
• /
• pp.233-234
• /
• 2014

• Proceedings of the Korea Water Resources Association Conference
• /
• 2007.05a
• /
• pp.253-257
• /
• 2007

Channel-aquifer interaction is one of the key hydrological processes that determine water flows in the stream/river channel. Field measurements of channel-aquifer interaction, however, is very difficult and costly, particularly when one intends to understand its variations across a catchment for a long period. Hydrological simulations using a catchment model are a relatively easier and cheaper alternative provided the model structure is appropriate for describing channel-aquifer interaction. In this study, a catchment model called CAMEL (Chemicals from Agricultural Management and Erosion Losses) is used for estimating channel-aquifer interaction over time and space. CAMEL is a distributed catchment model to simulate transformation and transport processes of sediment and pollutants as well as water flows at the catchment scale. In the model, a catchment is represented using a network of square columns each of which is comprised of various storages of water. CAMEL explicitly simulates both surface and subsurface processes including channel-aquifer interaction. This paper presents an application study results of CAMEL for the Tarland Burn Catchment, a small (catchment area $52\;km^2$) rural catchment in Scotland, UK, demonstrating some of the channel-aquifer interaction dynamics across the catchment during a 2-year period.

• Journal of the Korea Computer Graphics Society
• /
• v.17 no.4
• /
• pp.23-30
• /
• 2011

We propose an intelligent ribbon road network for automatic vehicle simulation, and a real-time algorithm for large-scale, realistic traffic simulation based on artificial energy functions. Our method reconstructs a road network automatically from both GIS (Geographic Information System) real-world data and synthetic models. Such automatic road network helps us to easily simulate almost every possible scenario such as intersections, ramps, etc. In order to simulate agents' movement, we design car-environment interaction energy and car-car interaction energy functions. Car agents move along the road network according to the proposed energy functions while avoiding collisions with other car agents.

• Journal of the Korea Computer Graphics Society
• /
• v.4 no.1
• /
• pp.47-54
• /
• 1998

The multi-user VR system can more enhance the sense of reality than existing single user VR system, but it also cause a significant problem about network structure and protocol, since the current Internet using TCP/IP can not afford to deliver such a massive information required to the real-time constraint of multi-user VR system. In this paper, we introduce new network structure and protocol for multiple user's virtual world navigation and interaction through their avatars. The virtual worlds are based on virtual building and managed by distributed servers. The avatars have 9 behaviors and recognize the state of each other by pre-defined network protocol.