• Communications for Statistical Applications and Methods
• /
• 제30권2호
• /
• pp.119-133
• /
• 2023

With the rapid growth of the economy and fossil fuel consumption, the concentration of air pollutants has increased significantly and the air pollution problem is no longer limited to small areas. We conduct statistical analysis with the actual data related to air quality that covers the entire of South Korea using R and Python. Some factors such as SO₂, CO, O₃, NO₂, PM₁₀, precipitation, wind speed, wind direction, vapor pressure, local pressure, sea level pressure, temperature, humidity, and others are used as covariates. The main goal of this paper is to predict air quality index (AQI) spatio-temporal data. The observations of spatio-temporal big datasets like AQI data are correlated both spatially and temporally, and computation of the prediction or forecasting with dependence structure is often infeasible. As such, the likelihood function based on the spatio-temporal model may be complicated and some special modelings are useful for statistically reliable predictions. In this paper, we propose several methods for this big spatio-temporal AQI data. First, random effects with spatio-temporal basis functions model, a classical statistical analysis, is proposed. Next, neural networks model, a deep learning method based on artificial neural networks, is applied. Finally, random forest model, a machine learning method that is closer to computational science, will be introduced. Then we compare the forecasting performance of each other in terms of predictive diagnostics. As a result of the analysis, all three methods predicted the normal level of PM_2.5 well, but the performance seems to be poor at the extreme value.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권7호
• /
• pp.2488-2511
• /
• 2015

Data redundancy has high impact on Wireless Sensor Network's (WSN) performance and reliability. Spatial and temporal similarity is an inherent property of sensory data. By reducing this spatio-temporal data redundancy, substantial amount of nodal energy and bandwidth can be conserved. Most of the data gathering approaches use either temporal correlation or spatial correlation to minimize data redundancy. In Collective Prediction exploiting Spatio Temporal correlation (CoPeST), we exploit both the spatial and temporal correlation between sensory data. In the proposed work, the spatial redundancy of sensor data is reduced by similarity based sub clustering, where closely correlated sensor nodes are represented by a single representative node. The temporal redundancy is reduced by model based prediction approach, where only a subset of sensor data is transmitted and the rest is predicted. The proposed work reduces substantial amount of energy expensive communication, while maintaining the data within user define error threshold. Being a distributed approach, the proposed work is highly scalable. The work achieves up to 65% data reduction in a periodical data gathering system with an error tolerance of 0.6℃ on collected data.

• Smart Structures and Systems
• /
• 제14권1호
• /
• pp.1-16
• /
• 2014

In this work, we address the so-called sensor reachback problem for Wireless Sensor Networks, which consists in collecting the measurements acquired by a large number of sensor nodes into a sink node which has major computational and power capabilities. Focused on applications such as Structural Health Monitoring, we propose a cooperative communication protocol that exploits the spatio-temporal correlations of the sensor measurements in order to save energy when transmitting the information to the sink node in a non-stationary environment. In addition to cooperative communications, the protocol is based on two well-studied adaptive filtering techniques, Least Mean Squares and Recursive Least Squares, which trade off computational complexity and reduction in the number of transmissions to the sink node. Finally, experiments with real acceleration measurements, obtained from the Canton Tower in China, are included to show the effectiveness of the proposed method.

• 한국환경과학회지
• /
• 제25권11호
• /
• pp.1563-1573
• /
• 2016

The aim of this study was to evaluate the occurrence of vanadium in Jeju Island groundwater, focusing on the spatio-temporal patterns and geochemical controlling factors of vanadium. For this, we collected two sets of groundwater data: 1) concentrations of major constituents of 2,595 groundwater samples between 2008 and 2014 and 2) 258 groundwater samples between December 2006 and June 2008. The concentrations of groundwater vanadium were in the range of $0.2{\sim}71.0{\mu}g/L$ (average, $12.0{\mu}g/L$) and showed local enrichments without temporal/seasonal variation. This indicated that vanadium distribution was controlled by 1) the geochemical/mineralogical composition and dissolution processes of original materials (i.e., volcanic rock) and 2) the flow and chemical properties of groundwater. Vanadium concentration was significantly positively correlated with that of major ions ($Cl^-$, $Na^+$, and $K^+$) and trace metals (As, Cr, and Al), and with pH, but was negatively correlated with $NO_3-N$ concentration. The high concentrations of vanadium (>$15{\mu}g/L$) occurred in typically alkaline groundwater with high pH (${\geq}8.0$), indicating that a higher degree of water-rock interaction resulted in vanadium enrichment. Thus, higher concentrations of vanadium occurred in groundwater of $Na-Ca-HCO_3$, $Na-Mg-HCO_3$ and $Na-HCO_3$ types and were remarkably lower in groundwater of $Na-Ca-NO_3$(Cl) type that represented the influences from anthropogenic pollution.

• 한국수산과학회지
• /
• 제47권6호
• /
• pp.1055-1062
• /
• 2014

Hydroacoustic surveys were conducted for spatio-temporal distribution and size estimation of fish in the Yongdam reservoir, Korea, from April to July 2014. Acoustic data were collected along acoustic transects using a 120 kHz scientific echosounder. The received acoustic data were the in situ acoustic target strength (dB) and nautical area scattering coefficient ($m^2/mile^2$). Data on fish behavioral patterns and size were collected using a DIDSON acoustic camera at stationary stations. Fish were mainly distributed in the center channel and close to the outer Yongdam reservoir. Acoustic density of fish in the summer season were higher than in the spring season. The seasonal vertical distribution pattern of fish aggregations may be strongly related to the vertical temperature structure. The size distribution of fish obtained from an acoustic camera correlated well with the acoustic size of fish from an echosounder.

• 응용통계연구
• /
• 제36권5호
• /
• pp.361-380
• /
• 2023

Magnetoencephalography (MEG)는 뉴론 활동에 신경 세포들간 전류 흐름에 의해 유도된 자기장을 측정하는 비침습 뇌영상 기술이다. 기능적 뇌활동은 뇌영역간 또는 뉴런들의 연결로 기능적 연결로 수행된다. MEG 데이터는 상관성, 시공간성을 가지며 다중 다층적 동적 네트워크인 특징을 갖는다. 이러한 복잡성 때문에 MEG 네트워크에 대한 연구는 아직 많지 않은 편이다. 본 연구에서는 MEG 네트워크 모형과 분석법을 소개하고 실제 MEG 데이터 분석에 활용되어 해석된 경우를 요약하고 앞으로 MEG 네트워크 모형 개발 연구의 필요성을 설명하고자 한다. 그러므로 통계적 네트워크 분석이 뇌과학에서 신경학적 질병을 포함하여 뇌기능에 대한 이해에 중요한 역할을 할 수 있음을 알리고자 한다.

• 지적과 국토정보
• /
• 제47권2호
• /
• pp.19-37
• /
• 2017

VIIRS-DNB 데이터는 기존의 DMSP-OLS 데이터에 비해 야간에 발생하는 빛의 밝기를 측정하는데 더 우수한 성능을 보여준다. 하지만 지금까지 우리나라에서 VIIRS-DNB 데이터를 이용해 야간 빛의 분포 변화를 분석한 연구는 상당히 드물다. 이 연구에서는 우리나라의 수도권을 대상으로 2013~2016년간의 야간 빛의 분포 및 변화 패턴을 파악하고, 공간회귀모델을 통해 그 요인을 분석하였다. 이를 위해 두 시점 간의 변화를 살펴봄은 물론 계절간 변화 양상 또한 함께 분석하였다. 주요한 결과는 다음과 같다. 첫째, 2013년과 2016년 두 시점 모두 야간 빛은 서울과, 인천, 그리고 서울과 인접한 경기도의 도시에 집중되어 인구밀도 및 주거지관련요인, 경제토지이용관련요인 등과의 연관성을 나타내었다. 둘째, 2013년과 2016년을 비교해보면 야간 빛의 강도는 특히 서울에서 약화되는 경향을 보이고 있는데, 이는 인구밀도의 변화 및 산업용 건물의 비중과 관련된 것으로 나타났다. 셋째, 서울, 그리고 인천과 경기도의 주변 지역들은 야간 빛의 계절 변동성이 높게 나타났는데, 겨울(12월, 1월, 2월) 및 가을(10월, 11월)에 빛의 강도가 가장 강하게 나타났다. 넷째, 야간 빛의 계절간 변동은 적설면적 변화와 유의미하게 양적 상관관계 나타났고, 알베도의 변화와 유의미하게 음적 상관관계 나타났다.

• 한국수산과학회지
• /
• 제21권6호
• /
• pp.351-360
• /
• 1988

Based on historic data of monthly means of sea surface temperatures (SST) for 24 years $(1921\~1944) $ at 23 Korean and Japanese coastal stations in the East Sea (the Japan Sea), we analyzed spatio-temporal characteristics of coastal SST and SST anomalies. The means of SST at Korean coast are higher than those at Japanese coast of the same latitudes, and the annual range of SST at Korean coast are larger than those at Japanese coast. Empirical orthogonal function analysis shows that almost all $(96\%)$ of the SST fluctuations are described by simultaneous seasonal variations. The flurtuations of SST anomalies are small in the Korea Strait and large at the boundaries between the warm and told currents in the basin. The fluctuations of SST anomalies along Korean coast are correlated each other The same is true for SST anomalies along Japanese coast. However, there is only weak correlation between the SST anomalies at Korean coast and those at Japanese coast. Empirical orthogonal function analysis shows that $27\%$ of the coastal SST anomalies in the East Sea are described by simultaneous fluctuations, and $12\%$ of them are described by alternating fluctuations between Korean and Japanese coasts.

• 인터넷정보학회논문지
• /
• 제10권4호
• /
• pp.35-44
• /
• 2009

데이터 스트림이란 새로운 개념과 기존의 단순 데이터 사이에 존재하는 개념적 차이를 극복하기 위해서는 많은 연구가 필요하다. 대표적인 예로써 센서 네크워크에서의 데이터 스트림 처리를 들 수 있는 데, 이를 위해서는 대역폭이나 에너지, 메모리와 같은 자원적 한계에서 부터 연속 질의를 포함하는 질의처리의 특수성까지 고려해야 할 대상이 광범위하다. 본 논문에서는 데이터 스트림 처리에서의 물리적 제약사항에 해당하는 한정된 메모리 문제를 해결하기 위해 PCA 기법을 기반으로 하는 데이터 스트림 축소 방안을 제안하다. PCA는 상호 관련된 다수의 변수들을 관련이 없는 적은 수의 변수로 변환해준다. 본 논문에서는 질의 처리 엔진의 협력을 가정하고서 센서 네크워크의 스트림 데이터 처리를 위해 PCA 기법을 적용하며, 다른 센서로부터 얻어진 많은 측정값 사이에 시공간적 관련성을 이용한다. 최종적으로 그러한 데이터 처리를 위한 프레임워크를 제시하고 다양한 실험을 통하여 기법의 성능을 분석한다.

• 말소리와 음성과학
• /
• 제11권3호
• /
• pp.31-38
• /
• 2019

The current electromagnetic articulography study analyzes several articulatory measures and examines whether, and if so, how they are interconnected, with a focus on cluster types and an additional consideration of speech rates and morphosyntactic contexts. Using articulatory data on non-assimilating contexts from three Seoul-Korean speakers, we examine how speaker-dependent gestural overlap between C1 and C2 in a low vowel context (/a/-to-/a/) and their resulting intergestural coordination are realized. Examining three C1C2 sequences (/k(#)t/, /k(#)p/, and /p(#)t/), we found that three normalized gestural overlap measures (movement onset lag, constriction onset lag, and constriction plateau lag) were correlated with one another for all speakers. Limiting the scope of analysis to C1 velar stop (/k(#)t/ and /k(#)p/), the results are recapitulated as follows. First, for two speakers (K1 and K3), i) longer normalized constriction plateau lags (i.e., less gestural overlap) were observed in the pre-/t/ context, compared to the pre-/p/ (/k(#)t/>/k(#)p/), ii) the tongue dorsum at the constriction offset of C1 in the pre-/t/ contexts was more anterior, and iii) these two variables are correlated. Second, the three speakers consistently showed greater horizontal distance between the vertical tongue dorsum and the vertical tongue tip position in /k(#)t/ sequences when it was measured at the time of constriction onset of C2 (/k(#)t/>/k(#)p/): the tongue tip completed its constriction onset by extending further forward in the pre-/t/ contexts than the uncontrolled tongue tip articulator in the pre-/p/ contexts (/k(#)t/>/k(#)p/). Finally, most speakers demonstrated less variability in the horizontal distance of the lingual-lingual sequences, which were taken as the active articulators (/k(#)t/=/k(#)p/ for K1; /k(#)t/