• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.185-199
• /
• 2024

In this paper, an improved automated spectral clustering (IASC) algorithm is proposed to address the limitations of the traditional spectral clustering (TSC) algorithm, particularly its inability to automatically determine the number of clusters. Firstly, a cluster number evaluation factor based on the optimal clustering principle is proposed. By iterating through different k values, the value corresponding to the largest evaluation factor was selected as the first-rank number of clusters. Secondly, the IASC algorithm adopts a density-sensitive distance to measure the similarity between the sample points. This rendered a high similarity to the data distributed in the same high-density area. Thirdly, to improve clustering accuracy, the IASC algorithm uses the cosine angle classification method instead of K-means to classify the eigenvectors. Six algorithms-K-means, fuzzy C-means, TSC, EIGENGAP, DBSCAN, and density peak-were compared with the proposed algorithm on six datasets. The results show that the IASC algorithm not only automatically determines the number of clusters but also obtains better clustering accuracy on both synthetic and UCI datasets.

• The Journal of Engineering Geology
• /
• v.29 no.4
• /
• pp.451-460
• /
• 2019

In this study, a complex behavior connector is proposed to overcome the problems that may occur when small pile pipe and micro pile is used as a friction pile concept in the lower foundation of an oil sand plant where a piloti foundation is used. The individual settlement and heaving of piles were connected in one group to allow the composite behavior. This study performed to analyze the load carrying capacity to identify a complex behavior. In addition, the shape of the composite behavior connector was examined to apply the advantages of pile-group and piled raft foundations to oil sand plants. A scale model was constructed to measure the behavior of the load. The stability and weakness of the device were selected to determine the shape of the connector using the scale model testing.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.1
• /
• pp.148-158
• /
• 1999

ABR traffic form ATM LAN is controlled through RM cell, and the interface function to public ATM network is necessary to provide ABR service efficiently. This paper presnets new interface architecture, which is based on "Projected Node" [6]. using AIPU(ABR Interface Proxy Unit) to support ABR traffic streams incoming from ATM LAN in the public ATM network. For the efficient interworking, the AIPU has designed for interworking functions with ERICA switch mechanism. Conventional ERICA switch mechanism specified in TM 4.0 is basically used for short distance comparative to public network, however AIPU adopts the novel control mechanism to cover logng roud trip time (RTT). To improve the problems and to provide a dynamic range of UCI(Update Count Interval), this paper proposes, a novel control scheme, DUCI ( Dynamic Update Count Interval. And the paper shows inefficiencies of ERICA mechanism with fixed UCI through the simulation results, and represents the performance enhancement of DUCI mechanism developed to adjust the update count interval dynamically.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.11
• /
• pp.101-107
• /
• 2010

As digital environment rapidly changes, various identification infrastructures have been introduced. It inevitably caused the Interoperability issue between the identifiers. Especially, Interoperability issue between the same kinds of identifiers raised the problems such as decreasing usefulness and cost overhead for making bridge system. In this paper, we resolve this issue by suggesting ID mashup service based on XRI. Although both UCI and ICN are the dominant identification infrastructures in the soundsource domain, the modification of identifiers or the requirement of complex system are essential for Interoperability. The ID mashup service suggested in the paper is able to provide interoperable functions to outer world without modifying the structures and resolution service of both identification infrastructures.

• Journal of Korean Dental Science
• /
• v.8 no.2
• /
• pp.82-88
• /
• 2015

Purpose: To evaluate the effect of light-curing on the immediate and delayed micro-shear bond strength (${\mu}SBS$) between yttria-tetragonal zirconia polycrystal (Y-TZP) ceramics and RelyX Ultimate when using Single Bond Universal (SBU). Materials and Methods: Y-TZP ceramic specimens were ground with #600-grit SiC paper. SBU was applied and RelyX Ultimate was mixed and placed on the Y-TZP surface. The specimens were divided into three groups depending on whether light curing was done after adhesive (SBU) and resin cement application: uncured after adhesive and uncured after resin cement application (UU); uncured after adhesive, but light cured after resin cement (UC); and light cured after adhesive and light cured resin cement (CC). The three groups were further divided depending on the timing of ${\mu}SBS$ testing: immediate at 24 hours (UUI, UCI, CCI) and delayed at 4 weeks (UUD, UCD, CCD). ${\mu}SBS$ was statistically analyzed using one-way ANOVA and Student-Newman-Keuls multiple comparison test (P<0.05). The surface of the fractured Y-TZP specimens was analyzed under a scanning electron microscope (SEM). Result: At 24 hours, ${\mu}SBS$ of UUI group ($8.60{\pm}2.06MPa$) was significantly lower than UCI group ($25.71{\pm}4.48MPa$) and CCI group ($29.54{\pm}3.62MPa$) (P<0.05). There was not any significant difference between UCI and CCI group (P>0.05). At 4 weeks, ${\mu}SBS$ of UUD group ($24.43{\pm}2.88MPa$) had significantly increased over time compared to UUI group (P<0.05). The SEM results showed mixed failure in UCI and CCI group, while UUI group showed adhesive failure. Conclusion: Light-curing of universal adhesive before or after application of RelyX Ultimate resin cement significantly improved the immediate ${\mu}SBS$ of resin cement to air-abrasion treated Y-TZP surface. After 4 weeks, the delayed ${\mu}SBS$ of the non-light curing group significantly improved to the level of light-cured groups.

• Smart Media Journal
• /
• v.1 no.1
• /
• pp.36-41
• /
• 2012

Feature selection is the important technique of selecting a subset of relevant features for building robust pattern recognition systems. Various methods have been studied for feature selection from sequential search algorithms to stochastic algorithms. In this work, we adopted a Quantum-inspired Genetic Algorithm (QGA) which is based on the concept and principles of quantum computing such as Q-bits and superposition of state for feature selection. The performance of QGA is compared to that of the Conventional Genetic Algorithm (CGA) with respect to the classification rates and the number of selected features. The experimental result using UCI data sets shows that QGA is superior to CGA.

• Journal of the Korean Society for Library and Information Science
• /
• v.44 no.2
• /
• pp.263-283
• /
• 2010

Free Use License is an effective means of allowing many people to freely use works in the public domain. It can also prevent copyright infringement problems. This article analyzes the current service and domestic and foreign technology as it applies to free use license. It then proposes a plan which incorporates diverse requirements including technical aspects. Strategies are formulated from the technical, service and the resources connection perspectives for applying license in addition to searching and using licensed works. This article suggests a three-stage utilization plan to encourage the use of free use license. First, a connection plan between digital archiving and free use license, is developed. Then, a connection plan between UCI identifier and free use license is developed, and finally, the plan for the use of public resources is presented.

• The KIPS Transactions:PartB
• /
• v.17B no.2
• /
• pp.157-162
• /
• 2010

We present a new distributed genetic algorithm that can be used to extract useful information from distributed, large data over the network. The main idea of the proposed algorithms is to determine how many and which individuals move between subpopulations at each site adaptively. In addition, we present a method to help individuals from other subpopulations not be weeded out but adapt to the new subpopulation. We used six data sets from UCI Machine Learning Repository to compare the performance of our approach with that of the single, centralized genetic algorithm. As a result, the proposed algorithm produced better performance than the single genetic algorithm in terms of the classification accuracy with the feature subsets.

• Journal of Digital Convergence
• /
• v.12 no.6
• /
• pp.223-229
• /
• 2014

In an attempt to retrieve useful information for an efficient decision in the large knowledge system, it is generally necessary and important for a refined feature selection. Rough set has difficulty in generating optimal reducts and classifying boundary objects. In this paper, we propose quick reduction algorithm generating optimal features by rough entropy analysis for condition and decision attributes to improve these restrictions. We define a new conditional information entropy for efficient feature extraction and describe procedure of feature selection to classify the significance of features. Through the simulation of 5 datasets from UCI storage, we compare our feature selection approach based on rough set theory with the other selection theories. As the result, our modeling method is more efficient than the previous theories in classification accuracy for feature selection.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.816-820
• /
• 2004

Clustering is a process of dividing similar data objects in data set into clusters and acquiring meaningful information in the data. The main issues related to clustering are the effective clustering of high dimensional data and optimization. This study proposed a method of measuring similarity based on SVM and a new method of calculating the number of clusters in an efficient way. The high dimensional data are mapped to Feature Space ones using kernel functions and then similarity between neighboring clusters is measured. As for created clusters, the desired number of clusters can be got using the value of similarity measured and the value of Δd. In order to verify the proposed methods, the author used data of six UCI Machine Learning Repositories and obtained the presented number of clusters as well as improved cohesiveness compared to the results of previous researches.