• Communications for Statistical Applications and Methods
• /
• v.17 no.3
• /
• pp.377-390
• /
• 2010

In the case of time-series analysis, it is often more convenient to rely on the frequency domain than the time domain. Spectral density is the core of the frequency-domain analysis that describes autocorrelation structures in a time-series process. Possible ways to estimate spectral density are to compute a periodogram or to average the periodogram over some frequencies with (un)equal weights. This can be an attractive tool to measure the similarity between time-series processes. We employ the metrics based on a smoothed periodogram proposed by Park and Kim (2008) for the classification of different classes of time-series processes. We consider several lag windows with unequal weights instead of a modified Daniel's window used in Park and Kim (2008). We evaluate the performance under various simulation scenarios. Simulation results reveal that the metrics used in this study split the time series into the preassigned clusters better than do the raw-periodogram based ones proposed by Caiado et al. 2006. Our metrics are applied to an economic time-series dataset.

• Journal of the Korean Mathematical Society
• /
• v.34 no.4
• /
• pp.809-824
• /
• 1997

With the primitive orders in quaternion algebra, theta series associated with these orders are constructed. Here, we studied the space of modular forms generated by these theta series.

• Korean Journal of Mathematics
• /
• v.30 no.1
• /
• pp.25-42
• /
• 2022

In this paper, we derive analogues of a couple of classes of infinite series identities with the confluent hypergeometric functions involving Dirichlet characters.

• Korean Journal of Clinical Pharmacy
• /
• v.30 no.4
• /
• pp.226-233
• /
• 2020

Background: Currently, the over-the-counter (OTC) drug market is flooded with series OTC products. The pharmacist must follow the OTC product's indication, given that the most critical role of a pharmacist is the right selection and recommendation of an OTC drug for a patient's symptoms in a dynamic pharmacy environment. Therefore, pharmacists must know each OTC product information precisely to avoid any ambiguity due to several OTC series brand names. Objective: We evaluated the risk and effectiveness of OTC series medicines. Methods: From December 5 to December 18, 2019, an online survey was conducted among 145 community pharmacists. Results: A total of 51.0% of pharmacists knew the difference between products named in a series and could explain it spontaneously. Only 0.7% of the pharmacists admitted to not knowing the difference between products named in a series. While 42.9% of pharmacists who owned a pharmacy opined that the OTC medicines named in a series have health benefits for patients, 50.0% of employee pharmacists admitted that they were rather confused because there are several OTC series medicines. In contrast, 69.2% of pharmacists who owned pharmacies and 72.2% of employee pharmacists admitted that OTC series drugs with names similar to popular OTC drugs sell better. Conclusion: While pharmacists had different opinions regarding OTC series drugs per employment status, they opined that OTC series are more helpful in pharmacy management than completely new brand names. Further studies in this regard are needed.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.6
• /
• pp.1530-1544
• /
• 2023

Data monitoring is an important foundation of modern science. In most cases, the monitoring data is time-series data, which has high application value. The deep learning algorithm has a strong nonlinear fitting capability, which enables the recognition of time series by capturing anomalous information in time series. At present, the research of time series recognition based on deep learning is especially important for data monitoring. Deep learning algorithms require a large amount of data for training. However, abnormal sample is a small sample in time series, which means the number of abnormal time series can seriously affect the accuracy of recognition algorithm because of class imbalance. In order to increase the number of abnormal sample, a data augmentation method called GANBATS (GAN-based Bi-LSTM and Attention for Time Series) is proposed. In GANBATS, Bi-LSTM is introduced to extract the timing features and then transfer features to the generator network of GANBATS.GANBATS also modifies the discriminator network by adding an attention mechanism to achieve global attention for time series. At the end of discriminator, GANBATS is adding averagepooling layer, which merges temporal features to boost the operational efficiency. In this paper, four time series datasets and five data augmentation algorithms are used for comparison experiments. The generated data are measured by PRD(Percent Root Mean Square Difference) and DTW(Dynamic Time Warping). The experimental results show that GANBATS reduces up to 26.22 in PRD metric and 9.45 in DTW metric. In addition, this paper uses different algorithms to reconstruct the datasets and compare them by classification accuracy. The classification accuracy is improved by 6.44%-12.96% on four time series datasets.

• Korean Journal of Soil Science and Fertilizer
• /
• v.9 no.4
• /
• pp.235-244
• /
• 1976

Following results are obtained by re-evaluating N. P. K. and soil improvement trials conducted from 1964 to 1969 in valley paddy soils in rolling lands (Jisan-series, imperfectly drained and Yongji-series, moderately well drained). 1. Average grain yield of rice in no fertilizer plots and the highest yield plots for Yongji-series (31 experiments) were 319 and 507kg/10a respectively, and that of Jisan-series (15 experiments) were 396 and 567kg/10a respectively. The fertility difference between two series may have been a result of the cultivation history. Jisan-series is a mature soil which has a long cultivation history and Yongji-series is sub-mature soil 2. Soil chemical characteristics for Jisan-series are charaterized by 12.8meq/100g in CEC, 6.5meq/100g in exchangeable Ca, 3.9% in OM, and 64 ppm in available $P_2O_5$ For Yongji-series they were 10.4meq/100g in CEC, 4.7meq/100g in exchangeable Ca, 3.2% in OM and 103ppm in available $P_2O_5$. 3. Deep plowing and application of organic matter and lime are expected to be effective in increasing fertility level of soils of Yongji-series. The same will be effective in some soils of Jisan series where the fertility level is low. 4. Jisan-series shows high response to nitrogen, while Yongji series shows sharp decrease in rice yield at the high levels of nitrogen. Both series, however, showed high response to nitrogen only when the OM level was higher than 3%. 5. The optimum level of nitrogen was 8~9kg for Jisan-series, and 10~11kg/10a for Yongji-series. The yield increase per kg of applied nitrogen was 12kg for Jisan-series and 13kg for Yongji series. 6. The optimum level of phosphorus at the optimum level of nitrogen was 6kg/10a for Yongji-series and 3kg/10a for Jisan-series. The optimum level of phosphorus, however, was different depending upon the nitrogen level. It was assumed that Yongji-series required more fertilizer (available $P_2O_5$ was 110ppm) than Jisan-series (available $P_2O_5$ was 64ppm) because the availability of P was higher in Jisan-series than Yongji-series due to the severe reduction of Jisan-series. 7. The response of potassium was also depending upon the nitrogen level. In Yongji-series the potassium response at 8kg/10a nitrogen level decreased with increasing levels of potassium, but the higher level of introgen, potassium response was also higher. In Jisan-series potassium response was recognized at all nitrogen levels. The optimum level of potassium at the optimum level of nitrogen was 8kg/10a in both serieses. 8. The reasonable ratio of NPK fertilizer seems to be 1:0.6:0.6:for Yongji-series and 1:0.4:1 for Jisan-series as N:$P_2O_5$:K.

• Journal for History of Mathematics
• /
• v.30 no.4
• /
• pp.233-246
• /
• 2017

$Ces{\grave{a}}ro$ summability is a generalized convergence criterion for infinite series. We have investigated the classical results of summability for Fourier series from 1897 to 1957. In this paper, we are concerned with the summability and summation methods for Fourier Series from 1960 to 2010. Many authors have studied the subject during this period. Especially, G.M. Petersen,$K{\hat{o}}si$ Kanno, S.R. Sinha, Fu Cheng Hsiang, Prem Chandra, G. D. Dikshit, B. E. Rhoades and others had studied neoclassical results on the summability of Fourier series from 1960 to 1989. We investigate the results on the summability for Fourier series from 1990 to 2010 in section 3. In conclusion, we present the research minor lineage on summability for Fourier series from 1960 to 2010. $H{\ddot{u}}seyin$ Bor is the earliest researcher on ${\mid}{\bar{N}},p_n{\mid}_k$-summability. Thus we consider his research results and achievements on ${\mid}{\bar{N}},p_n{\mid}_k$-summability and ${\mid}{\bar{N}},p_n,{\gamma}{\mid}_k$-summability.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.10 no.4
• /
• pp.39-47
• /
• 2015

"Series A Crunch"problem has become worse off as funding gap between angel seed investment and traditional VC Series A investment is getting the larger due to big boom over early startup investment both in Korea and US. The strong needs for new concept of fund alternative such as bridge fund is outstanding to fill up the funding gap in the early stage of venture investment. This research is brought to define the concept of 'Series A Crunch' problem and to diagnose its causes, eventually, popping up 'the concept of Micro VC Fund' to come up with this problem in Korea. Also, this paper suggests the policy alternatives to introduce Mivro VC Fund and accommodate its successful performance.

• Journal of Korea Water Resources Association
• /
• v.35 no.5
• /
• pp.631-638
• /
• 2002

Stochastic model has been widely used for the forecasting of time series. However, this study tries to perform the precipitation forecasting by fuzzy time series model using fuzzy concept. The published fuzzy based models are used for the forecasting of time series and also we suggest that the combination of fuzzy time series models and neuro-fuzzy system can increase the forecastibility of the models. The precipitation time series in illinois, USA is analyzed for the forecasting by the known fuzzy time series models and the suggested methodology in this study. As a result, we know that the suggested methodology shows more exact results than the known models.

• Journal for History of Mathematics
• /
• v.28 no.6
• /
• pp.321-332
• /
• 2015

This paper is concerned with the convergence of double trigonometric series and Fourier series. Since the beginning of the 20th century, many authors have studied on those series. Also, Ferenc $M{\acute{o}}ricz$ has studied the convergence of double trigonometric series and double Fourier series so far. We consider $L^p(T^2)$-convergence results focused on the Ferenc $M{\acute{o}}ricz^{\prime}s$ studies from the second half of the 20th century up to now. In section 2, we reintroduce some of Ferenc $M{\acute{o}}ricz^{\prime}s$ remarkable theorems. Also we investigate his several important results. In conclusion, we investigate his research trends and the simple minor genealogy from J. B. Joseph Fourier to Ferenc $M{\acute{o}}ricz$. In addition, we present the research minor lineage of his study on $L^p(T^2)$-convergence.