• 지질공학
• /
• 제34권1호
• /
• pp.107-115
• /
• 2024

This study assesses forest fire susceptibility in Gangwon-do, South Korea, which hosts the largest forested area in the nation and constitutes ~21% of the country's forested land. With 81% of its terrain forested, Gangwon-do is particularly susceptible to wildfires, as evidenced by the fact that seven out of the ten most extensive wildfires in Korea have occurred in this region, with significant ecological and economic implications. Here, we analyze 480 historical wildfire occurrences in Gangwon-do between 2003 and 2019 using 17 predictor variables of wildfire occurrence. We utilized three machine learning algorithms—random forest, logistic regression, and support vector machine—to construct wildfire susceptibility prediction models and identify the best-performing model for Gangwon-do. Forest and soil map data were integrated as important indicators of wildfire susceptibility and enhanced the precision of the three models in identifying areas at high risk of wildfires. Of the three models examined, the random forest model showed the best predictive performance, with an area-under-the-curve value of 0.936. The findings of this study, especially the maps generated by the models, are expected to offer important guidance to local governments in formulating effective management and conservation strategies. These strategies aim to ensure the sustainable preservation of forest resources and to enhance the well-being of communities situated in areas adjacent to forests. Furthermore, the outcomes of this study are anticipated to contribute to the safeguarding of forest resources and biodiversity and to the development of comprehensive plans for forest resource protection, biodiversity conservation, and environmental management.

• 한국물환경학회지
• /
• 제35권5호
• /
• pp.425-431
• /
• 2019

Recently, people have been attracting attention to the good quality of water resources as well as water welfare. to improve the quality of life. This study is a papers on the prediction of benthic macroinvertebrate index (BMI), which is a aquatic ecological health, using the machine learning based CFS (Correlation-based Feature Selection) method and the random forest model to compare the measured and predicted values of the BMI. The data collected from the Han River's branch for 10 years are extracted and utilized in 1312 data. Through the utilized data, Pearson correlation analysis showed a lack of correlation between single factor and BMI. The CFS method for multiple regression analysis was introduced. This study calculated 10 factors(water temperature, DO, electrical conductivity, turbidity, BOD, $NH_3-N$, T-N, $PO_4-P$, T-P, Average flow rate) that are considered to be related to the BMI. The random forest model was used based on the ten factors. In order to prove the validity of the model, $R^2$, %Difference, NSE (Nash-Sutcliffe Efficiency) and RMSE (Root Mean Square Error) were used. Each factor was 0.9438, -0.997, and 0,992, and accuracy rate was 71.6% level. As a result, These results can suggest the future direction of water resource management and Pre-review function for water ecological prediction.

• 응용통계연구
• /
• 제31권3호
• /
• pp.367-382
• /
• 2018

회사채 신용 등급 예측 모형에 대한 연구는 신용 평가 기관이 회사채 신용 등급 평가에 사용될 것이라 예상 되는 여러 재무적 특성 변수들을 기반으로 진행되었으며 선형 회귀 모형(linear regression), 순위 로짓(ordered logit), 순위 프로빗(ordered probit), 서포트 벡터 기계(support vector machine), 랜덤 포레스트(random forest) 등 다양한 모형들을 적용하여 개발되었다. 하지만 기존 연구들에서 고려한 회사채 신용 등급은 연구에 따라 5등급에서 20등급까지 다른 등급 구간을 적용하였으며 분석에 이용된 표본 자료의 기간 및 대상도 상이하여 예측 성능의 공정한 비교에 어려움이 있다. 따라서 본 연구에서는 2013년부터 2017년까지의 회사채 신용 등급 자료와 기존 연구들에서 사용된 재무 지표들을 통합하여 기존에 발표된 예측 모형들을 동일한 자료에 적용하고 예측 성능을 비교하였다. 추가적으로 Elastic-net 벌점화 회귀 모형 및 순위 로짓, 순위 프로빗 모형을 적합하여 LASSO 벌점이 선택됨을 확인하였으며 LASSO 벌점을 고려한 예측 모형이 대응하는 기존의 예측 모형들보다 향상된 성능을 보임을 확인하였다. 본 연구의 수행 결과, 랜덤 포레스트를 이용한 예측 모형이 15등급 기준 검증 자료에서 정확한 등급 예측률이 69.6%로 다른 모형과 비교하여 높은 예측 성능을 나타내었다.

• 한국항해항만학회지
• /
• 제42권6호
• /
• pp.539-545
• /
• 2018

운임시장의 심한 변동성과 시계열 데이터의 불안정성으로 해운시황 예측에 대한 연구가 큰 성과를 내지 못하고 있지만 최근 대표적인 비선형 모델인 기계학습모델을 적용한 연구들이 활발히 진행되고 있다. 대부분의 기존 연구가 계량모델의 설계단계에서 입력변수에 해당하는 요인들을 기존 문헌연구와 연구자의 직관에 의존하여 선정했기 때문에 요인선정에 대한 체계적인 연구가 필요하다. 본 연구에서는 케이프선 운임을 대상으로 단계적 회귀모형과 랜덤포레스트모델을 이용하여 중요 영향요인을 분석하였다. 해운시장에서 비교적 단순한 수급구조를 가져 요인파악이 용이한 케이프선 운임을 대상으로 하였으며 총 16개의 수급요인들을 사전 추출하였다. 요인간의 상호관련성을 파악하여 단계적 회귀는 8개 요인, 랜덤포레스트는 10개 요인을 분석대상으로 선정하였으며 선정된 변수를 입력변수로 하여 예측한 결과를 비교하였다. 랜덤포레스트의 예측성능이 아주 우수하였는데 수요요인이 주로 선정된 단계적 회귀분석과는 달리 공급요인이 비중 있게 선정되었기 때문인 것으로 판단된다. 본 연구는 운임예측 연구에 있어 운임결정요인에 대한 과학적인 근거를 마련하였으며 이를 위해 기계학습 기반의 모델을 활용하였다는데 연구적 의의가 있다. 또한 시장정보의 분석에 있어 실무자들이 어떤 변수에 중점을 두어야 하는지에 대해 합리적 근거를 제시한 측면에서 해운기업의 의사결정에 실질적 도움이 될 것으로 기대된다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2019년도 추계학술발표대회
• /
• pp.585-588
• /
• 2019

Predictions of Energy Consumption for Industries gain an important place in energy management and control system, as there are dynamic and seasonal changes in the demand and supply of energy. This paper presents and discusses the predictive models for energy consumption of the steel industry. Data used includes lagging and leading current reactive power, lagging and leading current power factor, carbon dioxide (tCO2) emission and load type. In the test set, four statistical models are trained and evaluated: (a) Linear regression (LR), (b) Support Vector Machine with radial kernel (SVM RBF), (c) Gradient Boosting Machine (GBM), (d) random forest (RF). Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) are used to measure the prediction efficiency of regression designs. When using all the predictors, the best model RF can provide RMSE value 7.33 in the test set.

• Communications for Statistical Applications and Methods
• /
• 제29권2호
• /
• pp.177-191
• /
• 2022

This paper addresses the use of machine learning methods for causal estimation of treatment effects from observational data. Even though conducting randomized experimental trials is a gold standard to reveal potential causal relationships, observational study is another rich source for investigation of exposure effects, for example, in the research of comparative effectiveness and safety of treatments, where the causal effect can be identified if covariates contain all confounding variables. In this context, statistical regression models for the expected outcome and the probability of treatment are often imposed, which can be combined in a clever way to yield more efficient and robust causal estimators. Recently, targeted maximum likelihood estimation and causal random forest is proposed and extensively studied for the use of data-adaptive regression in estimation of causal inference parameters. Machine learning methods are a natural choice in these settings to improve the quality of the final estimate of the treatment effect. We explore how we can adapt the design and training of several machine learning algorithms for causal inference and study their finite-sample performance through simulation experiments under various scenarios. Application to the percutaneous coronary intervention (PCI) data shows that these adaptations can improve simple linear regression-based methods.

• Educational Technology International
• /
• 제15권2호
• /
• pp.71-88
• /
• 2014

A variety of studies to predict students' performance have been conducted since educational data such as web-log files traced from Learning Management System (LMS) are increasingly used to analyze students' learning behaviors. However, it is still challenging to predict students' learning achievement in blended learning environment where online and offline learning are combined. In higher education, diverse cases of blended learning can be formed from simple use of LMS for administrative purposes to full usages of functions in LMS for online distance learning class. As a result, a generalized model to predict students' academic success does not fulfill diverse cases of blended learning. This study compares two blended learning classes with each prediction model. The first blended class which involves online discussion-based learning revealed a linear regression model, which explained 70% of the variance in total score through six variables including total log-in time, log-in frequencies, log-in regularities, visits on boards, visits on repositories, and the number of postings. However, the second case, a lecture-based class providing regular basis online lecture notes in Moodle show weaker results from the same linear regression model mainly due to non-linearity of variables. To investigate the non-linear relations between online activities and total score, RF (Random Forest) was utilized. The results indicate that there are different set of important variables for the two distinctive types of blended learning cases. Results suggest that the prediction models and data-mining technique should be based on the considerations of diverse pedagogical characteristics of blended learning classes.

• Geomechanics and Engineering
• /
• 제36권4호
• /
• pp.381-390
• /
• 2024

The introduction of bio-based materials has been recommended in the geotechnical engineering field to reduce environmental pollutants such as heavy metals and greenhouse gases. However, bio-treated soil methods face limitations in field application due to short research periods and insufficient verification of engineering performance, especially when compared to conventional materials like cement. Therefore, this study aimed to develop a machine learning model for predicting the unconfined compressive strength, a representative soil property, of biopolymer-based soil treatment (BPST). Four machine learning algorithms were compared to determine a suitable model, including linear regression (LR), support vector regression (SVR), random forest (RF), and neural network (NN). Except for LR, the SVR, RF, and NN algorithms exhibited high predictive performance with an R² value of 0.98 or higher. The permutation feature importance technique was used to identify the main factors affecting the strength enhancement of BPST. The results indicated that the unconfined compressive strength of BPST is affected by mean particle size, followed by biopolymer content and water content. With a reliable prediction model, the proposed model can present guidelines prior to laboratory testing and field application, thereby saving a significant amount of time and money.

• 해양환경안전학회지
• /
• 제27권1호
• /
• pp.13-21
• /
• 2021

본 연구의 목적은 Neural Network Regression 모델을 활용하여 선박의 가치평가 모델을 개발하는 것이다. 가치평가의 대상은 중고 VLCC선이며, 선행연구를 통해 선박의 가치 변화를 유발하는 주요 요인들을 선별하여 변수를 설정하고, 2000년 1월부터 2020년 8월까지의 해당 데이터를 확보하였다. 변수의 안정성을 판단하기 위해 다중 공선성 검사를 수행하여 최종적으로 6개의 독립변수와 1개의 종속변수를 선정하고 연구 구조를 설계하였다. 이를 바탕으로 Linear Regression, Neural Network Regression, Random Forest Algorithm을 활용하여 총 9개의 시뮬레이션 모델을 설계하였다. 또한 각 모델간의 비교검증을 통해 평가결과의 정확성을 제고시켰다. 평가 결과, VLCC실제값과의 비교를 통해 2층으로 구성된 Hidden Layer의 Neural Network Regression 모델이 가장 정확도가 높은 것으로 나타났다. 본 연구의 시사점은 첫째, 기존 정형화된 평가기법에서 벗어나 기계학습기반 모델을 선박가치평가에 적용하였다는 점이다. 둘째, 해운시장 변화요인을 동태적 관점에서 분석하고 예측함으로써 연구결과의 객관성을 제고시켰다고 할 수 있다.

• 대한원격탐사학회지
• /
• 제34권1호
• /
• pp.141-150
• /
• 2018

Crop type classification is essential for supporting agricultural decisions and resource monitoring. Remote sensing techniques, especially using hyperspectral imagery, have been effective in agricultural applications. Hyperspectral imagery acquires contiguous and narrow spectral bands in a wide range. However, large dimensionality results in unreliable estimates of classifiers and high computational burdens. Therefore, reducing the dimensionality of hyperspectral imagery is necessary. In this study, the Random Forest (RF) classifier was utilized for dimensionality reduction as well as classification purpose. RF is an ensemble-learning algorithm created based on the Classification and Regression Tree (CART), which has gained attention due to its high classification accuracy and fast processing speed. The RF performance for crop classification with airborne hyperspectral imagery was assessed. The study area was the cultivated area in Chogye-myeon, Habcheon-gun, Gyeongsangnam-do, South Korea, where the main crops are garlic, onion, and wheat. Parameter optimization was conducted to maximize the classification accuracy. Then, the dimensionality reduction was conducted based on RF variable importance. The result shows that using the selected bands presents an excellent classification accuracy without using whole datasets. Moreover, a majority of selected bands are concentrated on visible (VIS) region, especially region related to chlorophyll content. Therefore, it can be inferred that the phenological status after the mature stage influences red-edge spectral reflectance.