• 유통과학연구
• /
• 제21권7호
• /
• pp.41-50
• /
• 2023

Purpose: The purpose of the research is identified factors influencing the competitiveness of the copper industry in Kazakhstan. Research design, data and methodology: A few studies are dedicated to the analysis in developing countries, particularly Kazakhstan. The algorithm was chosen for research provision: statistical and comparative analysis, correlation, and regression analysis. The data of 1999-2021 obtained from the World Bank, Bureau of National Statistics, National Bank of Kazakhstan. Results: The obtained results demonstrate the trends in the development of the industry since 2000. The development of the copper industry is strongly influenced by the distribution and state of the business environment, economic situation, and trends in the global commodity markets. Conclusions: According to econometric modeling, there is a correlation between the profitability of the copper industry, GDP, copper prices, liquidity, and energy resource prices. Trends in global commodity and energy markets have a significant impact on the state of the industry. Further research should be conducted to include an analysis and forecast of internal factors that may affect the development of the industry, such as copper reserves, condition of fixed assets, government programs, etc. It is also important to examine the correlation with the trends in the development of the global green economy and the revival of the Chinese market.

• 한국환경과학회지
• /
• 제26권6호
• /
• pp.719-728
• /
• 2017

In order to investigate the adsorption characteristics for Sr ion using the Na-X zeolite synthesized from coal fly ash, batch tests and response surface analyses were carried out. The adsorption kinetic data for Sr ions, using Na-X zeolite, fitted well with the pseudo-second-order model. The uptake of Sr ions followed the Langmuir isotherm model, with a maximum adsorption capacity of 196.46 mg/g. Thermodynamic studies were conducted at different reaction temperatures, with the results indicating that Sr ion adsorption by Na-X zeolite was an endothermic (${\Delta}H^o$>0) and spontaneous (${\Delta}G^o$<0) process. Using the response surface methodology of the Box-Behnken method, initial Sr ion concentration ($X_1$), initial temperature ($X_2$), and initial pH ($X_3$) were selected as the independent variables, while the adsorption of Sr ions by Na-X zeolite was selected as the dependent variable. The experimental data fitted well with a second-order polynomial equation by multiple regression analysis. The value of the determination coefficient ($R^2=0.9937$) and the adjusted determination coefficient (adjusted $R^2=0.9823$) was close to 1, indicating high significance of the model. Statistical results showed the order of Sr removal based on experimental factors to be initial pH > initial concentration > temperature.

• Journal of Forest and Environmental Science
• /
• 제30권4호
• /
• pp.351-361
• /
• 2014

Leaf area ($A_0$) and leaf biomass ($M_0$) estimation are significant prerequisites to studying tree physiological processes and modeling in the forest ecosystem. The objective of this study was to develop allometric models for estimating $A_0$ and $M_0$ of Swietenia mahagoni L. from different tree parameters such as DBH and tree height of mahogany plantations in the northeastern region of Bangladesh. A total of 850 healthy and well formed trees were selected randomly for sampling in the five study sites. Then, twenty two models were developed based on different statistical criteria that propose reliable and accurate models for estimating the $A_0$ and $M_0$ using non-destructive measurements. The results exposed that model iv and xv were selected on a single predictor of DBH and showed more statistically accuracy than other models. The selected models were also validated with an additional test data set on the basis of linear regression and t-test for mean difference between observed and predicted values. After that, a comparison between the best logarithmic and non-linear allometric model shows that the non-linear model produces systematic biases and underestimates $A_0$ and $M_0$ for larger trees. As a result, it showed that the bias-corrected logarithmic model iv and xv can be used to help quantify forest structure and functions, particularly valuable in future research for estimating $A_0$ and $M_0$ of S. mahagoni in this region.

• 산업융합연구
• /
• 제21권1호
• /
• pp.167-173
• /
• 2023

경제지표를 분석하는 방법으로 회귀 분석이나, 인공지능을 활용하여 미래의 데이터를 예측하는 연구가 발표되었다. 본 연구에서는 토픽모델링을 사용하여 과거 뉴스 기사로부터 얻은 주제 확률 데이터를 이용한 인공지능으로 미래 선물 가격을 예측하는 시스템을 구상하였다. 과거 뉴스 기사로부터 비지도학습을 통한 문서의 주제를 추출할 수 있는 LDA 방법으로 각 뉴스 기사 주제 확률 분포 데이터를 얻을 수 있고, 해당 데이터를 인공지능의 RNN의 파생 구조인 LSTM의 입력 데이터로 활용함으로써 미래 선물 가격을 예측하였다. 본 연구에서 제안한 방법에서는 선물 가격의 추세를 예측할 수 있었고, 이를 활용하여 추후 옵션 상품 등의 파생 상품에 대한 가격 추세도 예측할 수 있을 것으로 보인다. 다만, 일부 데이터에 대해 오차가 발생하는 것이 확인되어 정확도 향상을 위한 추가적인 연구가 필요하다.

• 대한교통학회지
• /
• 제26권4호
• /
• pp.29-40
• /
• 2008

우리나라는 건설교통부에서 규정하는 지침에 의거하여 도로를 설계하는데 안전한 도로의 건설을 위해서는 사고와 연관된 기하구조요인의 정확한 이해와 분석을 통해 도로의 설계자가 도로 설계과정에 반영하도록 하고 있다. 따라서 본 연구의 목적은 고속도로 인터체인지에서 발생하는 교통사고를 실제로 수집이 가능한 도로기하구조 및 교통여건에 관한 자료만을 이용하여 신뢰성있게 분석할 수 있는 모형으로 정립하는데 있다. 본 연구에서는 고속도로상의 총 129개 트럼펫인터체인지와 35개 클로버인터체인지 상에서 연결로 형식별로 발생한 교통사고를 이용하여 다양한 교통사고요인들의 관계를 분석하기 위해 교통사고발생분포를 통계학적 방법을 통하여 분석한 바, 음이항분포가 가장 적합한 것으로 확인되었다. 따라서 트럼펫인터체인지의 연결로 형식별, 그리고 클로버인터체인지의 연결로 형식별로 교통사고를 분석할 수 있는 음이항회귀모형을 개발하였다. 본 모형은 고속도로를 계획 및 설계하는 초기단계에서 인터체인지의 유형과 인터체인지 연결로의 유형을 결정하는데 활용할 목적으로 개발되었으며, 본 모형의 적합성을 판단하는 여러 가지 통계학적 값들과 모형을 통해 예측한 값들, 그리고 실제로 현장에서 관측한 값들의 차이를 분석한 바 본 논문에서 구축한 모형이 이론적 및 실용적 측면에서 적합하게 구축되었음을 확인하였다.

• 한국산업정보학회논문지
• /
• 제14권5호
• /
• pp.281-288
• /
• 2009

본 논문에서는 수량화 방법과 AHP(Analytic Hierarchy Process) 기법을 사용하여 산사태 발생에 대한 통계적 예측모형을 구축하는데 목적이 있다. 수량화(Quantification) 방법은 질적변수에 수량을 부여하는 통계적 방법으로, 기 조사된 자료에 기반하여 분석을 수행하는 방법이다. 본 논문에서는 서구의 다변량분석 기법인 정준상관분석의 결과를 토대로 수량화 과정을 구체적으로 제안한다. 데이터에 기반한 수량화 방법과는 팔리 AHP 기법은 일종의 다기준 의사결정을 위해 사용되는 기법으로, 설문자료에 기반한 분석법이다. 실제자료에 대한 분석으로 산사태 발생여부를 측정한 자료(한국지질자원연구원 제공)와 전문가 설문을 통해 수집된 자료를 이용하였다. 이들 자료에 대해 수량화 분석과 AHP분석을 통해 산사태 발생여부를 예측할 수 있는 두 종류의 평가표와 함께 로지스틱 회귀를 통한 통계적 예측모형을 개발하였으며, 두 모형간의 성능비교와 안정성 평가를 수행하였다.

• Computers and Concrete
• /
• 제8권3호
• /
• pp.293-309
• /
• 2011

A long-term structural health monitoring (SHM) system comprising over 700 sensors of sixteen types has been implemented on the Guangzhou Television and Sightseeing Tower (GTST) of 610 m high for real-time monitoring of the structure at both construction and service stages. As part of this sophisticated SHM system, 48 temperature sensors have been deployed at 12 cross-sections of the reinforced concrete inner structure of the GTST to provide on-line monitoring via a wireless data transmission system. In this paper, the differential temperature profiles in the reinforced concrete inner structure of the GTST, which are mainly caused by solar radiation, are recognized from the monitoring data with the purpose of understanding the temperature-induced structural internal forces and deformations. After a careful examination of the pre-classified temperature measurement data obtained under sunny days and non-sunny days, common characteristic of the daily temperature variation is observed from the data acquired in sunny days. Making use of 60-day temperature measurement data obtained in sunny days, statistical patterns of the daily rising temperature and daily descending temperature are synthesized, and temperature distribution models of the reinforced concrete inner structure of the GTST are formulated using linear regression analysis. The developed monitoring-based temperature distribution models will serve as a reliable input for numerical prediction of the temperature-induced deformations and provide a robust basis to facilitate the design and construction of similar structures in consideration of thermal effects.

• Genomics & Informatics
• /
• 제19권1호
• /
• pp.11.1-11.8
• /
• 2021

For the novel coronavirus disease 2019 (COVID-19), predictive modeling, in the literature, uses broadly susceptible exposed infected recoverd (SEIR)/SIR, agent-based, curve-fitting models. Governments and legislative bodies rely on insights from prediction models to suggest new policies and to assess the effectiveness of enforced policies. Therefore, access to accurate outbreak prediction models is essential to obtain insights into the likely spread and consequences of infectious diseases. The objective of this study is to predict the future COVID-19 situation of Korea. Here, we employed 5 models for this analysis; SEIR, local linear regression (LLR), negative binomial (NB) regression, segment Poisson, deep-learning based long short-term memory models (LSTM) and tree based gradient boosting machine (GBM). After prediction, model performance comparison was evelauated using relative mean squared errors (RMSE) for two sets of train (January 20, 2020-December 31, 2020 and January 20, 2020-January 31, 2021) and testing data (January 1, 2021-February 28, 2021 and February 1, 2021-February 28, 2021) . Except for segmented Poisson model, the other models predicted a decline in the daily confirmed cases in the country for the coming future. RMSE values' comparison showed that LLR, GBM, SEIR, NB, and LSTM respectively, performed well in the forecasting of the pandemic situation of the country. A good understanding of the epidemic dynamics would greatly enhance the control and prevention of COVID-19 and other infectious diseases. Therefore, with increasing daily confirmed cases since this year, these results could help in the pandemic response by informing decisions about planning, resource allocation, and decision concerning social distancing policies.

• 한국수자원학회논문집
• /
• 제36권6호
• /
• pp.1047-1058
• /
• 2003

겨울철에 금강하류에서는 암모니아성 질소(NH3-N) 농도가 주기적으로 높게 검출되어, 부여지점에서 취수하는 정수장의 수처리 공정에 큰 장애가 되고 있다. 질소농도 저하와 소독부산물 생성 억제를 위해 종종 대청댐의 추가 방류가 검토되고 있으나, 방류량과 직소농도 관계의 정량적 분석에 어려움이 있었다. 본 연구에서는 8년간의 일별 수질자료와 댐 방류량 자료를 이용하여 겨울철(12월∼3월) 동안 일별 NH3-N 농도를 예측할 수 있는 다중회귀모형을 개발하고, 최근 2년간의 자료를 이용하여 모형을 검증하였다. 모형 개발과정에서 모의값은 실측값과의 결정계수와 모형효율이 모두 0.95 이상으로 높게 나타났다. 검증과정에서는 각각 0.84∼0.94와 0.77∼0.93으로써 신뢰도가 약간 떨어졌지만, 방류량과 하류의 NH3-N 농도 관계를 분석하는데 충분히 활용가능 한 것으로 평가되었다. 모형은 갈수기 동안 댐 방류량이 NH3-N 농도 저감에 미친 효과를 분석하는데 사용되었다. 1∼3월 동안 방류량을 5 에서 50cms까지 증가시켜 파며 모의한 결과, NH3-N 농도는 평균 0.332∼0.583 mg/L 감소하였으며 2월에 가장 큰 저감효과가 나타났다. 개발된 다중회귀 수질모의기법은 충분한 실측자료가 확보된 경우에 수치모형이 요구하는 광범위한 경계조건 자료 없이도 댐 방류량과 하천수질의 인과관계를 분석하는데 유용하게 활용가능 할 것으로 기대된다

• 한국환경보건학회지
• /
• 제47권6호
• /
• pp.548-557
• /
• 2021

Background: Due to the emergence of new diseases such as COVID-19, an increasing number of people are struggling with stress and depression. Interest is growing in forest-based recreation for physical and mental relief. Objectives: A prediction model equation using meteorological factors and data was developed to predict the quantities of medicinal substances generated in forests (monoterpenes) in real-time. Methods: The concentration of phytoncide and meteorological factors in the forests near Chuncheon in South Korea were measured for nearly two years. Meteorological factors affecting the observation data were acquired through a multiple regression analysis. A model equation was developed by applying a linear regression equation with the main factors. Results: The linear regression analysis revealed a high explanatory power for the coefficients of determination of temperature and humidity in the coniferous forest (R²=0.7028 and R²=0.5859). With a temperature increase of 1℃, the phytoncide concentration increased by 31.7 ng/Sm³. A humidity increase of 1% led to an increase in the coniferous forest by 21.9 ng/Sm³. In the deciduous forest, the coefficients of determination of temperature and humidity had approximately 60% explanatory power (R²=0.6611 and R²=0.5893). A temperature increase of 1℃ led to an increase of approximately 9.6 ng/Sm³, and 1% humidity resulted in a change of approximately 6.9 ng/Sm³. A prediction model equation was suggested based on such meteorological factors and related equations that showed a 30% error with statistical verification. Conclusions: Follow-up research is required to reduce the prediction error. In addition, phytoncide data for each region can be acquired by applying actual regional phytoncide data and the prediction technique proposed in this study.