• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.5 no.1
• /
• pp.59-65
• /
• 1987

This paper presented the procedures for detecting gross errors, and described the influence of the number and distribution of points on the internal reliability in photogrammetric adjustment, such as relative orientation and independent model method. The use of the standard six points for relative orientation and the regular four points for independent model method result in low internal reliability. With such a distribution, gross erors in measured points might not be detected But using cluster of double or triple points instead of individual point, internal reliability improves remarkably.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2002.06a
• /
• pp.139-152
• /
• 2002

Gene-shaving(Hastie et al, 2000) is a very useful method to identify a meaningful group of genes when the variation of expression is large. By shaving off the low-correlated genes with the leading principal component, the primary genes with the coherent expression pattern can be identified. Gene-shaving method works well If expression levels are varied enough, but it may not catch the meaningful cluster in low expression level or different expression time even with coherent patterns. The sliding window gene-shaving method which is to apply gene-shaving in each sliding window after hierarchical clustering is to compensate losing a meaningful set of genes whose variation is not large but distinct. The performance to identify expression patterns is compared for the simulated profile data by the different variance and expression level.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.4
• /
• pp.435-442
• /
• 2001

According to the current increase of the usefulness of information in Internet, several methods are proposed in which multimedia information may be efficiently managed and retrieved. The purpose of this paper is to propose the new grouping method by SRR(Semantic Reference Relation) and DRR(Direct Reference Relation). The important point of this method proposed in this paper is to group MDI(Multimedia Document Informations) as a cluster of this multimedia objects. According to the result of experimental simulation, which has been tested by by the 1,000 multimedia items in internet, this method has made more efficiently the service and grouping of MDI possible than any other methods do in internet.

• Journal of KIISE:Computing Practices and Letters
• /
• v.15 no.12
• /
• pp.993-997
• /
• 2009

Clustering is well known method and that can be used in statistical machine translation. In this paper we propose a corpus clustering method using syntactic structure and POS information of dependency grammar. And using this cluster language model as additional feature to phrased-based statistical machine translation system to improve translation Quality.

• Proceedings of the Korean Statistical Society Conference
• /
• 2003.10a
• /
• pp.9-14
• /
• 2003

A Bayesian model-based clustering method is proposed for clustering objects on the basis of dissimilarites. This combines two basic ideas. The first is that tile objects have latent positions in a Euclidean space, and that the observed dissimilarities are measurements of the Euclidean distances with error. The second idea is that the latent positions are generated from a mixture of multivariate normal distributions, each one corresponding to a cluster. We estimate the resulting model in a Bayesian way using Markov chain Monte Carlo. The method carries out multidimensional scaling and model-based clustering simultaneously, and yields good object configurations and good clustering results with reasonable measures of clustering uncertainties. In the examples we studied, the clustering results based on low-dimensional configurations were almost as good as those based on high-dimensional ones. Thus tile method can be used as a tool for dimension reduction when clustering high-dimensional objects, which may be useful especially for visual inspection of clusters. We also propose a Bayesian criterion for choosing the dimension of the object configuration and the number of clusters simultaneously. This is easy to compute and works reasonably well in simulations and real examples.

• Journal of the Korean Data and Information Science Society
• /
• v.18 no.2
• /
• pp.481-488
• /
• 2007

The k-means algorithm is well known for its efficiency in clustering large data sets. However, working only on numeric values prohibits it from being used to cluster real world data containing categorical values. The k-modes algorithm is to extend the k-means paradigm to categorical domains. The algorithm requires a pre-setting or random selection of initial points (modes) of the clusters. This paper improved the problem of k-modes algorithm, using the Max-Min method that is a kind of methods to decide initial values in k-means algorithm. we introduce new similarity measures to deal with using the categorical data for clustering. We show that the mushroom data sets and soybean data sets tested with the proposed algorithm has shown a good performance for the two aspects(accuracy, run time).

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.5 no.1
• /
• pp.88-93
• /
• 2005

Web usage mining is a research field for searching potentially useful and valuable information from web log file. Web log file is a simple list of pages that users refer. Therefore, it is not easy to analyze user's current interest field from web log file. This paper presents web usage mining method for finding users' current interest based on fuzzy categories. We consider not only how many times a user visits pages but also when he visits. We describe a user's current interest with a fuzzy interest degree to categories. Based on fuzzy categories and fuzzy interest degrees, we also propose a method to cluster users according to their interests for user modeling. For user clustering, we define a category vector space. Experiments show that our method properly reflects the time factor of users' web visiting as well as the users' visit number.

• The Transactions of the Korean Institute of Electrical Engineers C
• /
• v.51 no.11
• /
• pp.511-516
• /
• 2002

$TiO_2-SnO_2$ thin films are fabricated using sol-gel method. In case the amount of water required hydrolysis smaller than that for stoichiometry, Ti sol forms clear sol which has normal chain structure. On the contrary, in case the amount of water required hydrolysis larger than that for stoichiometry, Ti sol forms suspended sol which has cluster structure. The thickness of thin films increase about $0.03{\sim}0.04{\mu}m$ every a dipping. The permittivity and dissipation factor of $TiO_2-SnO_2$ thin films decrease with increasing frequency. Thin films show semiconductive characteristics above $400^{\cric}C$.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.64 no.1
• /
• pp.41-47
• /
• 2015

In recent years, energy supply cases to take advantage of EMS(Energy Management System) are increasing according to high interest of energy efficiency. The important factor for essential and economical EMS operation is the supply and demand plan the hourly power demand of building load using the hierarchical clustering method of variety statistical techniques, and use the real historical data of target load. Also the estimated results of study are obtained the reliability through separate tests of validity.

• 제어로봇시스템학회:학술대회논문집
• /
• 1988.10b
• /
• pp.777-781
• /
• 1988

A method for weighing fruits without separating them from stem is proposed. The base of stem is fixed and a fruit or a cluster of fruits is forced to vibrate. The approximated vibration model is constructed by the use of Transfer Matrix Method. The natural frequency (w) in this model can be represented as a function of weight elements, and the length and stiffness of branch elements of stem. With this function, only w is possible to measure. However, several small weights whose weights are known are attached to weight elements in various combinations. From these equations, unknown parameters are determined so that the weight of each fruit can be obtained by a non-destructive method.