• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1056-1059
• /
• 2003

Self-Oranizing Map(SOM) is an unsupervised neural network providing cluster analysis of high dimensional input data. The output from the SOM is represented in map that help us to explore data. The weak point of conventional SOM is when the map is large, it take a long time to train the data. The computing time is known to be O(MN) for trainning to find the winning node (M,N are the number of nodes in width and height of the map). This paper presents a new method to reduce the computing time by creating new map. Each node in a new map is the centroid of nodes' group that are in the original map. After create a new map, we find the winning node of this map, then find the winning node in original map only in nodes that are represented by the winning node from the new map. This new method is called "High Speed Self-Oranizing Map"(HS-SOM). Our experiment use HS-SOM to cluster documents and compare with SOM. The results from the experiment shows that HS-SOM can reduce computing time by 30%-50% over conventional SOM.

• Korean Journal of Environment and Ecology
• /
• v.32 no.3
• /
• pp.274-288
• /
• 2018

This study investigated the environmental factors and fish assemblage in 42 wetlands between spring and autumn of 2017 to evaluate the fish distribution and deduce the management strategy for improving biodiversity in created wetlands located at the Nakdong River basin. The investigation identified a total of 30 fish species and found that the most of wetlands were dominated by exotic fishes such as Micropterus salmoides and Lepomis macrochirus. Fish species such as Rhinogobius brunneus, Opsariichthys uncirostris amurensis, Zacco platypus were less abundant in the area with high density of Micropterus salmoides (static area) because they preferred the environment with active water flow. The pattern analysis of fish distribution in each wetland using the self-organizing map (SOM) showed a total of 24 variables (14 fish species and 10 environmental variables). The comparison of variables indicated that the distribution of fish species varied according to water depth and plant cover rate and was less affected by water temperature, pH, and dissolved oxygen. The plant cover rate was strongly associated with high fish density and species diversity. However, wetlands with low plant biomass had diversity and density of fish species. The results showed that the microhabitat structure, created by macrophytes, was an important factor in determining the diversity and abundance of fish communities because the different species compositions of macrophytes supported diverse fish species in these habitats. Based on the results of this study, we conclude that macrophytes are the key components of lentic freshwater ecosystem heterogeneity, and the inclusion of diverse plant species in wetland construction or restoration schemes will result in ecologically healthy food webs.