• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.7
• /
• pp.3112-3127
• /
• 2018

Since time series clustering is performed without prior information, it is used for exploratory data analysis. In particular, clusters of time series with similar shapes can be used in various fields, such as business, medicine, finance, and communications. However, existing time series clustering algorithms have a problem in that time series with different shapes are included in the clusters. The reason for such a problem is that the existing algorithms do not consider the limitations on the size of the generated clusters, and use a dimension reduction method in which the information loss is large. In this paper, we propose a method to alleviate the disadvantages of existing methods and to find a better quality of cluster containing similarly shaped time series. In the data preprocessing step, we normalize the time series using z-transformation. Then, we use piecewise aggregate approximation (PAA) to reduce the dimension of the time series. In the clustering step, we use density-based spatial clustering of applications with noise (DBSCAN) to create a precluster. We then use a modified K-means algorithm to refine the preclusters containing differently shaped time series into subclusters containing only similarly shaped time series. In our experiments, our method showed better results than the existing method.

• Journal of applied mathematics & informatics
• /
• v.25 no.1_2
• /
• pp.525-531
• /
• 2007

We newly introduce the concept of summable in measure and investigate on some its properties. In addition to this, we consider a size of given series by means of we are giving Lebesgue measure to an associated series.

• 제어로봇시스템학회:학술대회논문집
• /
• 1986.10a
• /
• pp.108-112
• /
• 1986

Several algorithms for bivariate time series modeling are reviewed : linear least square, nonlinear least squares, generalized least square, and multi-stage least square methods. Estimation results of simulated data by the above methods are discussed.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.27 no.11
• /
• pp.37-46
• /
• 2013

This paper proposes the boost type bidirectional zero current switching(ZCS) DC/DC converter of transformer series construction for electric vehicle operation using low voltage battery. This converter can high boost through the double voltage circuit and series construction of output part using two converters. This converter system has the advantages that bidirectional power transfer is excellent, size and making of transformer because of this converter keeps the transformation ratio to 1:1. Proposed DC/DC converter uses the ZCS method to decrease the switching loss. By replacing reactance ingredients of L-C resonance circuit for ZCS with leakage inductance ingredients of high frequency transformer and half-bridge capacitor it reduces system size and expense because of not add special reactor. It can confirm to output of high voltage to operate the electric vehicle with low voltage of input and operation of ZCS in all load region through the result of PSIM simulation and experiment.

• Proceedings of the Korea Electromagnetic Engineering Society Conference
• /
• 2005.11a
• /
• pp.245-248
• /
• 2005

In this paper, small sized half-wavelength circular loop antenna which attached beneath a visor of helmet is designed and fabricated at the resonant frequency of 449MHz. To reduce the size of the antenna, double series-shorted stubs are inserted inside the loop for practical use of a vacant space of loop inside. Also, It is designed for the printed type to install the helmet easily. The size of antenna on helmet is reduced to 87.75%(diameter : 70mm, height :.36mm) compare with general type antenna(diameter : 200mm height 101mm). The return loss, -10dB bandwidth and gain are -13.2dB, 17.6MHz(3.9%), and -1.78dBd. And, radiation pattern is omni-directional pattern at H-plane. Therefore, it can be seen that the half-wavelength circular loop antenna using double series-shorted stubs is proper structure for the miniaturization and the installed antenna of the helmet.

• Journal of Welding and Joining
• /
• v.12 no.4
• /
• pp.127-140
• /
• 1994

Resistance spot welding has been widely used in the sheet metal joining processes because of its high productivity and convenience. In the resistance spot welding processes the size of molten nugget is a criterion to assess weld quality. Many research have founded on measuring weld nugget size at the same time monitoring welding process parameters such as dynamic resistance and electrode movement. With increasing demand of energy saving, many efforts were made to employ aluminum alloys that are lighter than steel and have relatively equivalent strength to steel in the automobile industry. In this paper, spot weldability of aluminum alloys for various welding conditions were examined by series of experiments. One of the 6000 series (Mg-Si) aluminum alloy, 6383-T4 was chosen, which is currently considered as a substitute for the galvanized steel. Dynamic resistance, electrode movement and corresponding nugget size were observed and compared to the case of steel. Finally, resistance spot welding of dissimilar material (galvanized steel-aluminum alloy) was attempted.

• Fisheries and Aquatic Sciences
• /
• v.2 no.2
• /
• pp.210-217
• /
• 1999

Chemical oxygen demand (COD), phytoplankton cell number and chlorophyll-a concentration (Chl-a), sediment mean grain size and ignition loss were studied to determine their temporal trends in the study area. Historical data of COD, cell number and Chl-a were gathered from the late 1960s or early 1980s to 1997, and trends in temporal domain were obtained from a simple regression. Sediments for grain size and ignition loss (as organic contents in sediments) were sampled from the Chokchon macrotidal flat bimonthly from September 1990 to November 1996, and were analyzed using the decomposition method of time series analysis. In general, the first three data showed increasing trends based on regression analysis. The trends of sediment grain size fluctuated in a neutral pathway while those of ignition loss yielded no increasing pattern. In contrast with the suggestions from Ahn and Choi (1998) who reported a coarsening variation in sediment grain size to be a cause of the directional and remarkable changes of macrofaunal communities in this area, we could not find such a corresponding variation pattern from our samples. In diagnosing eutrophication, a paradoxical phenomenon was encountered between the trends in water column (COD, cell number and Chl-a) and sediment (ignition loss) data. In this paper, we inferred the possible processes that produce the discrepancy. Some explanations and biological responses to eutrophication were predicted and discussed.

• Korean Journal of Microbiology
• /
• v.17 no.3
• /
• pp.97-130
• /
• 1979

Many industrial processes those employ bacteria are subjected to phage infestations. In L-glutamic acid fermentions using acetic acid, the phage infestations of the organisms have been recently recognized. In efforts to elucidate the sources of phage contamination involved in the abnormal fermentation, a series of study was conducted to isolate the phages both from the contents of abnormally fermented tanks and the soil or sewage samples from the surroundings of a fermentation factory, to define major charateristics of the phage isolates, and finally to determine the correlation between the phage isolates and temperate phages originating from the miscellaneous bacterial species isolated from the soil or sewage samples. The results are summarized as follows; 1) All phages were isolated from the irregular fermentation tanks and soil or sewage samples, and they were designated as phage PR-1, PR-2, PR-3, PR-4, PR-5, PR-6, and PR-7, in the order of isolation. These PR-series phages were proved to be highly specific for the variant strains of Br. lactofermentum only, namely, phage PR-1 and PR-2 for Br. lactofermentum No. 468-5 and phage PR-3~PR-7 for Br. lactofemrentum No. 2256. By cross-neutralization test, the 7 phagescould be subdivided into 3 groups, i. e., phage PR-I and PR-2 the first, phage PR-3, PR-4, PR-5, PR-6 the second, and the phage PR-7 the third. 2) The 7 phages were virulent under the experimental conditions. They produced plaques with clear and relatively sharp margins without distinct halo. The mean sizes of plaques were 1.5mm in diameter for phage PR-1 and PR-2, and 1. Omm for phages PR-3~PR-7. Double layer technique modified by Hongo and described by Adams, was applied to assay of the PR-series phages. The factors influencing the plaques were as follows;young age cells of host bacteria cultured for 3-6 hours represented the largest number and size, optimum was pH 7.0, incubation temperature was $30^{\circ}C$, and agar concentration and amount of overlayer medium were 0.6% and 0.2ml, respectively. 3) PR-series phages were stable in 0.05M tris buffer and 0.1M ammonium acetate buffer solution. The addition of $5{\times}10^{-3}M$ magnesium ion effectively increased the stability. Thermostability experiments indicated that PR-series phages were stable at the teinperture between $50^{\circ}{\sim}55^{\circ}C$ in nutrient medium, $45^{\circ}{\sim}50^{\circ}C$ in buffer solution. However, the phages mere completely inactivated at 603C and 65$^{\circ}$C within 10 minutes. The phages were stable at the range of pH6~9 in nutrient medium and of pH 8-9 in buffer solution, respectively. Exposure of the phages to UV for 25, 60 and 100 seconds resulted in the complete loss of infectivily, respectively. 4) Electron microscopy showed that PR-series phage particles exhibited rather similar morphology, differing in the size All of PR-series phages had a multilateral head and had a simple long tiil about three to five times long as compared with head. By the size, phage PR-1 and PR-2, PR-3, PR-4, PR-5, and PR-6 and PR-7 were classified into same groups, respectively. The head and tail size of phage PR-1, PR-5, PR-5(T) and PR-7 were 85nm, 74nm and 235nm and 350mm, and 72nm and 210nm, respectively. 5) Nucleic acids of PR-series phages were double stranded DNA. The G+C contents of phage PR-1, PR-5 and PR-7 were 56.1, 52.9 and 53.7, respectively. The values of G+C contents derived from the $T_m$ were in agreement with the chemically determined values. 6) PR-series phages effectively adsorbed on their host bacteria at the rate of more than 90% during 5 min. K value for phage PR-1, PR-5 and PR-7 were calculated to be $6{\times}10^9 ml$ per minute, respectiveky. The pH of the medium did effect adsorption rate, but both temperature and age of host cells did not. Generally, optimum adsorption condition of phages seemed to be almost same as optimum growth conditions of host bacteria. 7) In one-step growth experiments, the latent periods at $30^{\circ}C$ for PR-1, and PR-7 were about 70, 50 and 55 min, respectively. The corresponding average burst size was 200, 70 and 90, respectively. Lpsis period according to the multiplicity of infection and a phage series. In case of m. o. i. 100, strain No. 2256 (PR-5) and No. 468-5(PR-1) failed to grow and turbidity decreased after 50 and 70min, respectively. 8) In the lysate of a plaque purified phage PR-5 infected bacteria, there observed 2 types ofphage particles, i. e., phage PR-5 and PR-5 (T) of similar morphology but differing at the length of phage tail, and phage tail like particles. The phage taillike particles could be divided into 4 types by the length. Induction experiments of Br. lactofermentum with UV irradiation, mitomycin C or bacitracin treatment produced neither phage PR-5 (T) or phage tail-like particles. 9) No lysis occured when the growth of 7 strains of miscellaneous bacteria, isolated from soil and sewage samples, were inoculated with either phage PR-5 (T) or phage tail-like particles the inoculation of phage PR-5 pellet resulted in the growth inhibition of the orgainsms in the spot test. The lysates obtained from 3 miscellaneous soil derived bacteria following mitomycin C treatment the growth of Br. lactofermentum, but did not lyze the bacterium.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.11 no.3 s.25
• /
• pp.89-99
• /
• 1987

The purpose of this present study is not only to investigate mutal relationship between the preference of clothing design and the nature of Reflectiveness-Impulsiveness, but to offer clothing design for desirably developing of child, especially of preschool children. And the practical research was performed for 166 preschool children who are in kindergarten located in Seoul and are selected according to their sex and social class. This study used Kagan's MFFT and Personality Inventory of Hwang Eung-Yeon for the test of Reflectiveness-Impulsiveness, while for the preference of clothing design used the photo deck which is showed to children. The data analysis was based on frequency, percentage, mean, standard deviation, chi-square ($x^2$) test, t-test, F-test, and Duncan's Multiple Range Test. It can be concluded that girls rather than toys, high class children rather than middle or low class children were more reflective as a general trend. The relationship between the preference of clothing design and the nature of Reflectiveness and Impulsiveness is as follows : 1. In color, reflective boy liked blue series and reflective girl liked pastel coloring series, while they disgusted red series and non-coloring series. The group of impulsiveness liked red series, while they disgusted non-coloring series. 2. In color combination, group of reflectiveness liked similar color, while group of impulsiveness liked contrasting color. 3. In texture, group of reflectiveness liked soft and shiny texture, while they disgusted thick and rough texture. Also Group of impulsiveness liked soft texture. 4. In pattern, reflective boy liked lettering pattern and reflective girl liked flower pattern, while she disgusted lettering pattern. Impulsive boy liked stripe pattern and limpulsive girl liked flower pattern. 5. In pattern size, group of reflectivenss liked small size, while group of impulsiveness liked large one. 6. In style, group of reflectiveness liked formal style.

• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.103-123
• /
• 1999

Input filtering as a preprocessing method is so much crucial to get good performance in time series forecasting. There are a few preprocessing methods (i.e. ARMA outputs as time domain filters, and Fourier transform or wavelet transform as time-frequency domain filters) for handling time series. Specially, the time-frequency domain filters describe the fractal structure of financial markets better than the time domain filters due to theoretically additional frequency information. Therefore, we, first of all, try to describe and analyze specially some issues on the effectiveness of different filtering methods from viewpoint of the performance of a neural network based forecasting. And then we discuss about neural network model architecture issues, for example, what type of neural network learning architecture is selected for our time series forecasting, and what input size should be applied to a model. In this study an input selection problem is limited to a size selection of the lagged input variables. To solve this problem, we simulate on analyzing and comparing a few neural networks having different model architecture and also use an embedding dimension measure as chaotic time series analysis or nonlinear dynamic analysis to reduce the dimensionality (i.e. the size of time delayed input variables) of the models. Throughout our study, experiments for integration methods of joint time-frequency analysis and neural network techniques are applied to a case study of daily Korean won / U. S dollar exchange returns and finally we suggest an integration framework for future research from our experimental results.