• Journal of Korea Multimedia Society
• /
• v.22 no.4
• /
• pp.432-442
• /
• 2019

Most current human action recognition methods based on deep learning methods. It is required, however, a very high computational cost. In this paper, we propose an action change detection method to reduce repetitive human action recognition tasks. In reality, simple actions are often repeated and it is time consuming process to apply high cost action recognition methods on repeated actions. The proposed method decides whether action has changed. The action recognition is executed only when it has detected action change. The action change detection process is as follows. First, extract the number of non-zero pixel from motion history image and generate one-dimensional time-series data. Second, detecting action change by comparison of difference between current time trend and local extremum of time-series data and threshold. Experiments on the proposed method achieved 89% balanced accuracy on action change data and 61% reduced action recognition repetition.

• The Journal of Korean Medicine
• /
• v.43 no.4
• /
• pp.145-158
• /
• 2022

Objectives: To identify changes in the subjects and methods of Herbal Medicine Consumption Survey, and analyze trend of the key results Methods: The population, methods, and items of the basic reports of all Surveys on Consumption of Herbal Medicine(HM) were organized in a time-series manner. The analysis items were trend in the purchase of standardized HM; consumption value share, and price of prepared HM; type of herbal dispensary; and awareness of HM policy in Koran Medicine(KM) institutions. Results: The price of HM preparations showed an upward trend in 2011, 2014, and 2017 surveys, and decreased in the 2020 survey. However, despite this recent decrease, the 2021 survey also saw the highest proportion of HM users reporting that price of herbal decoction is expensive. Furthermore, the demand for expanded coverage of herbal decoction was the greatest for the expansion of health insurance benefits. Efforts such as adjusting the number of covered diseases and the cost of health insurance coverage would be necessary. Regarding decoction dispensaries the proportion of HM hospitals using only extramural herbal dispensaries increased. Finally, the consumption of HM and the size of the HM industry has continued to expand due to the large-scale branding of KM institutions and the expansion of health insurance coverage. Conclusion: Future surveys must standardize and unify the items for the time-series continuity and compare the results with government statistics reports on HM to increase reliability. Moreover, specialized survey items may be developed for KM, to establish a better and efficient distribution system for domestic HMs.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.49 no.3
• /
• pp.274-284
• /
• 2022

The purpose of this study was to analyze the trend of the number of births in Gwangju and the number of outpatients in Pediatric Dentistry at Chonnam National University Dental Hospital over the past 10 years (2010 - 2019) and predict the next year using time series analysis. The number of births showed an unstable downward trend with monthly variations, with the highest in January and the lowest in December. The average number of births in 2020 was predicted to be 682 (595 to 782, 95% CI), and the actual number of births was an average of 610. The number of outpatients was relatively stable, showing a month-to-month variation, with highest in August and the lowest in June. The average number of patients in 2020 was predicted to be 603 (505 to 701, 95% CI), and the average number of actual visits was 587. Despite the decrease in the number of births, the number of outpatients was expected to increase somewhat. Due to the special situation of COVID-19, the actual number of births and patients was to be slightly lower than the predicted values, but it was that they were within the predicted confidence interval. Time series analysis can be used as a basic tool to prepare for the low fertility era in the field of pediatric dentistry.

• Journal of Korean Society of Transportation
• /
• v.19 no.6
• /
• pp.65-73
• /
• 2001

Until recently, we use only weekly and monthly adjustment factors in order to estimate the AADT. By the way. we can suppose that the traffic is time series data related to flow of time. So we tried to analyse traffic patterns using time series analysis and apply them to estimate the AADT. We could divide traffic patterns into trend, cyclic variation, seasonal variation and irregular variation like as time series data. Also, in order to reduce random error components, we have looked for the weather conditions as an influential factor. There are many weather conditions such as rainfalls, but, temperatures, and sunshine hours among others but we selected rainfalls and lowest temperatures. And then, we have estimated the AADT using time series factors. To compare the results of, we have applied both irregular variation joined to weather factors and that not joined to. RMSE and U-test were opted at methods to appreciate results of AADT estimation.

• Advanced Industrial SCIence
• /
• v.3 no.1
• /
• pp.17-22
• /
• 2024

History, which began with the emergence of mankind, has a means of recording. Today, we can check the past through data. Generated data may only be generated and stored at a certain moment, but it is not only continuously generated over a certain time interval from the past to the present, but also occurs in the future, so making predictions using it is an important task. In order to find out trends in the use of time series data among numerous data, this paper analyzes the concept of time series data, analyzes Recurrent Neural Network and Long-Short Term Memory, which are mainly used for time series data analysis in the machine learning field, and analyzes the use of these models. Through case studies, it was confirmed that it is being used in various fields such as medical diagnosis, stock price analysis, and climate prediction, and is showing high predictive results. Based on this, we will explore ways to utilize it in the future.

• The Journal of Economics, Marketing and Management
• /
• v.9 no.2
• /
• pp.1-10
• /
• 2021

Purpose: In this work, we examine several time series models to predict internet and mobile transaction amount in South Korea, whereas Jeong (2020) has obtained the optimal forecasts for online shopping transaction amount by using time series models. Additionally, optimal forecasts based on the model considered can be calculated and applied to the Corona 19 situation. Research design, data, and methodology: The data are extracted from the online shopping trend survey of the National Statistical Office, and homogeneous and comparable in size based on 46 realizations sampled from January 2007 to October 2020. To achieve the goal of this work, both multiplicative ARIMA model and Holt-Winters Multiplicative seasonality method are taken into account. In addition, goodness-of-fit measures are used as crucial tools of the appropriate construction of forecasting model. Results: All of the optimal forecasts for the next 12 months for two online shopping transactions maintain a pattern in which the slope increases linearly and steadily with a fixed seasonal change that has been subjected to seasonal fluctuations. Conclusions: It can be confirmed that the mobile shopping transactions is much larger than the internet shopping transactions for the increase in trend and seasonality in the future.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.8
• /
• pp.3757-3763
• /
• 2012

This study ultimately aims at proposing an IT-based efficient method to solve one of the biggest problems currently faced by human beings which is lack of water. As a trial, targeting residential water, a chain of efforts was added such as choosing an appropriate field area and a censor, installing a sensor and the communication systems and servers, and monitoring the real time residential water consumption data. Then, a series of residential water consumption models was developed through the analyses of data gathered. And using the developed models, a series of trend analyses was performed for the residential water consumption. The research results shows that the developed models can be generalized and utilized for the water supply management purpose individually or along with the ones from the other water categories.

• Journal of Digital Contents Society
• /
• v.15 no.5
• /
• pp.643-650
• /
• 2014

This study deals with the effects of the extension of terrestrial TV VOD holdback on the viewing behavior focusing on IPTV & digital cable TV. And this study presents the implication by analyzing a time series trend of the digital media service. Thus researcher has analyzed the VOD performance trend of IPTV & digital cable TV, based on the real data of media player's internal fact sheet. First, researcher confirmed that the repetition of the peak value goes to January and August seasonally. Also, it was confirmed that the viewing rate of terrestrial broadcasting companies is affected by the hit drama program. And researcher confirmed that the terrestrial VOD, movie and kids VOD, and terrestrial VOD purchases is falling down by analyzing a time series relationship. Terrestrial broadcasting companies did an extension of the holdback for the purpose of trying to increase the viewership of the broadcast, but it was confirmed that it was reduced both purchasing desire VOD viewing and viewing of the broadcast audience rather. Thus the researcher expect the customer familiar policy of media players in the future.

• Journal of Information Processing Systems
• /
• v.16 no.2
• /
• pp.508-521
• /
• 2020

We compared a novel encoding Lempel-Ziv complexity (ELZC) with three common complexity algorithms i.e., approximate entropy (ApEn), sample entropy (SampEn), and classic Lempel-Ziv complexity (CLZC) so as to determine a satisfied complexity and its corresponding quality indices for assessing quality of multi-lead electrocardiogram (ECG). First, we calculated the aforementioned algorithms on six artificial time series in order to compare their performance in terms of discerning randomness and the inherent irregularity within time series. Then, for analyzing sensitivity of the algorithms to content level of different noises within the ECG, we investigated their change trend in five artificial synthetic noisy ECGs containing different noises at several signal noise ratios. Finally, three quality indices based on the ELZC of the multi-lead ECG were proposed to assess the quality of 862 real 12-lead ECGs from the MIT databases. The results showed the ELZC could discern randomness and the inherent irregularity within six artificial time series, and also reflect content level of different noises within five artificial synthetic ECGs. The results indicated the AUCs of three quality indices of the ELZC had statistical significance (>0.500). The ELZC and its corresponding three indices were more suitable for multi-lead ECG quality assessment than the other three algorithms.

• Journal of the Korean Statistical Society
• /
• v.27 no.1
• /
• pp.67-90
• /
• 1998

In this paper we analytically study model misspecification that arises in regression analysis of nonstationary seasonal time series. We assume the underlying data generating process is a seasonally or a regularly and seasonally integrated process. We first study consequences of totally misspecified cases where seasonal indicator variables, a linear time trend, or another statistically independent seasonally integrated process are used as predictor variables in order to model the nonstationary seasonal behavior of the dependent variable. Then we study consequences of partially misspecified cases where the dependent variable and a predictor variable are cointegrated at some, but not all of the frequencies corresponding to the nonstationary roots.