• Journal of Environmental Impact Assessment
• /
• v.33 no.2
• /
• pp.74-83
• /
• 2024

This study aims to estimate roadkill occurrences and investigate influential factors in Chungcheongnam-do, contributing to the establishment of roadkill prevention measures. By comprehensively considering weather, road, and environmental information, machine learning was utilized to estimate roadkill incidents and analyze the importance of each variable, deriving primary influencing factors. The Gradient Boosting Machine (GBM) exhibited the best performance, achieving an accuracy of 92.0%, a recall of 84.6%, an F1-score of 89.2%, and an AUC of 0.907. The key factors affecting roadkill included average local atmospheric pressure (hPa), average ground temperature (℃), month, average dew point temperature (℃), presence of median barriers, and average wind speed (m/s). These findings are anticipated to contribute to roadkill prevention strategies and enhance traffic safety, playing a crucial role in maintaining a balance between ecosystems and road development.

• Computers and Concrete
• /
• v.33 no.1
• /
• pp.77-90
• /
• 2024

Waste foundry sand (WFS) is the waste product that cause environmental hazards. WFS can be used as a partial replacement of cement or fine aggregates in concrete. A database comprising 234 compressive strength tests of concrete fabricated with WFS is used. To construct the machine learning-based prediction models, the water-to-cement ratio, WFS replacement percentage, WFS-to-cement content ratio, and fineness modulus of WFS were considered as the model's inputs, and the compressive strength of concrete is set as the model's output. A base extreme gradient boosting (XGBoost) model together with two hybrid XGBoost models mixed with the tunicate swarm algorithm (TSA) and the salp swarm algorithm (SSA) were applied. The role of TSA and SSA is to identify the optimum values of XGBoost hyperparameters to obtain the higher performance. The results of these hybrid techniques were compared with the results of the base XGBoost model in order to investigate and justify the implementation of optimisation algorithms. The results showed that the hybrid XGBoost models are faster and more accurate compared to the base XGBoost technique. The XGBoost-SSA model shows superior performance compared to previously published works in the literature, offering a reduced system error rate. Although the WFS-to-cement ratio is significant, the WFS replacement percentage has a smaller influence on the compressive strength of concrete. To improve the compressive strength of concrete fabricated with WFS, the simultaneous consideration of the water-to-cement ratio and fineness modulus of WFS is recommended.

• Geomechanics and Engineering
• /
• v.36 no.3
• /
• pp.259-276
• /
• 2024

The undrained shear strength is widely acknowledged as a fundamental mechanical property of soil and is considered a critical engineering parameter. In recent years, researchers have employed various methodologies to evaluate the shear strength of soil under undrained conditions. These methods encompass both numerical analyses and empirical techniques, such as the cone penetration test (CPT), to gain insights into the properties and behavior of soil. However, several of these methods rely on correlation assumptions, which can lead to inconsistent accuracy and precision. The study involved the development of innovative methods using extreme gradient boosting (XGB) to predict the pile set-up component "A" based on two distinct data sets. The first data set includes average modified cone point bearing capacity (q_t), average wall friction (f_s), and effective vertical stress (σ_vo), while the second data set comprises plasticity index (PI), soil undrained shear cohesion (S_u), and the over consolidation ratio (OCR). These data sets were utilized to develop XGBoost-based methods for predicting the pile set-up component "A". To optimize the internal hyperparameters of the XGBoost model, four optimization algorithms were employed: Particle Swarm Optimization (PSO), Social Spider Optimization (SSO), Arithmetic Optimization Algorithm (AOA), and Sine Cosine Optimization Algorithm (SCOA). The results from the first data set indicate that the XGBoost model optimized using the Arithmetic Optimization Algorithm (XGB - AOA) achieved the highest accuracy, with R2 values of 0.9962 for the training part and 0.9807 for the testing part. The performance of the developed models was further evaluated using the RMSE, MAE, and VAF indices. The results revealed that the XGBoost model optimized using XGBoost - AOA outperformed other models in terms of accuracy, with RMSE, MAE, and VAF values of 0.0078, 0.0015, and 99.6189 for the training part and 0.0141, 0.0112, and 98.0394 for the testing part, respectively. These findings suggest that XGBoost - AOA is the most accurate model for predicting the pile set-up component.

• Korean Journal of Remote Sensing
• /
• v.37 no.2
• /
• pp.321-335
• /
• 2021

Particulate matter (PM10 and PM2.5 with a diameter less than 10 and 2.5 ㎛, respectively) can be absorbed by the human body and adversely affect human health. Although most of the PM monitoring are based on ground-based observations, they are limited to point-based measurement sites, which leads to uncertainty in PM estimation for regions without observation sites. It is possible to overcome their spatial limitation by using satellite data. In this study, we developed machine learning-based retrieval algorithm for ground-level PM10 and PM2.5 concentrations using aerosol parameters from Geostationary Ocean Color Imager (GOCI) satellite and various meteorological parameters from a numerical weather prediction model during January to December of 2019. Gradient Boosted Regression Trees (GBRT) and Light Gradient Boosting Machine (LightGBM) were used to estimate PM concentrations. The model performances were examined for two types of feature sets-all input parameters (Feature set 1) and a subset of input parameters without meteorological and land-cover parameters (Feature set 2). Both models showed higher accuracy (about 10 % higher in R2) by using the Feature set 1 than the Feature set 2. The GBRT model using Feature set 1 was chosen as the final model for further analysis(PM10: R2 = 0.82, nRMSE = 34.9 %, PM2.5: R2 = 0.75, nRMSE = 35.6 %). The spatial distribution of the seasonal and annual-averaged PM concentrations was similar with in-situ observations, except for the northeastern part of China with bright surface reflectance. Their spatial distribution and seasonal changes were well matched with in-situ measurements.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.147-170
• /
• 2022

Although the number of patent which is one of the core outputs of technological innovation continues to increase, the number of low-value patents also hugely increased. Therefore, efficient evaluation of patents has become important. Estimation of patent lifespan which represents private value of a patent, has been studied for a long time, but in most cases it relied on a linear model. Even if machine learning methods were used, interpretation or explanation of the relationship between explanatory variables and patent lifespan was insufficient. In this study, patent lifespan (number of renewals) is predicted based on the idea that patent lifespan represents the value of the patent. For the research, 4,033,414 patents applied between 1996 and 2017 and finally granted were collected from USPTO (US Patent and Trademark Office). To predict the patent lifespan, we use variables that can reflect the characteristics of the patent, the patent owner's characteristics, and the inventor's characteristics. We build four different models (Ridge Regression, Random Forest, Feed Forward Neural Network, Gradient Boosting Models) and perform hyperparameter tuning through 5-fold Cross Validation. Then, the performance of the generated models are evaluated, and the relative importance of predictors is also presented. In addition, based on the Gradient Boosting Model which have excellent performance, Accumulated Local Effects Plot is presented to visualize the relationship between predictors and patent lifespan. Finally, we apply Kernal SHAP (SHapley Additive exPlanations) to present the evaluation reason of individual patents, and discuss applicability to the patent evaluation system. This study has academic significance in that it cumulatively contributes to the existing patent life estimation research and supplements the limitations of existing patent life estimation studies based on linearity. It is academically meaningful that this study contributes cumulatively to the existing studies which estimate patent lifespan, and that it supplements the limitations of linear models. Also, it is practically meaningful to suggest a method for deriving the evaluation basis for individual patent value and examine the applicability to patent evaluation systems.

• Asia-Pacific Journal of Business
• /
• v.13 no.2
• /
• pp.107-122
• /
• 2022

Purpose - This study aims to develop machine learning models to predict administrative issue designation in KOSDAQ Market using financial data. Design/methodology/approach - Employing four classification techniques including logistic regression, support vector machine, random forest, and gradient boosting to a matched sample of five hundred and thirty-six firms over an eight-year period, the authors develop prediction models and explore the practicality of the models. Findings - The resulting four binary selection models reveal overall satisfactory classification performance in terms of various measures including AUC (area under the receiver operating characteristic curve), accuracy, F1-score, and top quartile lift, while the ensemble models (random forest and gradienct boosting) outperform the others in terms of most measures. Research implications or Originality - Although the assessment of administrative issue potential of firms is critical information to investors and financial institutions, detailed empirical investigation has lagged behind. The current research fills this gap in the literature by proposing parsimonious prediction models based on a few financial variables and validating the applicability of the models.

• Structural Engineering and Mechanics
• /
• v.84 no.5
• /
• pp.605-618
• /
• 2022

Glass fiber reinforced polymer (GFRP) elastic gridshells consist of long continuous GFRP tubes that form elastic deformations. In this paper, a method for the form-finding of gridshell structures is presented based on the interpretable machine learning (ML) approaches. A comparative study is conducted on several ML algorithms, including support vector regression (SVR), K-nearest neighbors (KNN), decision tree (DT), random forest (RF), AdaBoost, XGBoost, category boosting (CatBoost), and light gradient boosting machine (LightGBM). A numerical example is presented using a standard double-hump gridshell considering two characteristics of deformation as objective functions. The combination of the grid search approach and k-fold cross-validation (CV) is implemented for fine-tuning the parameters of ML models. The results of the comparative study indicate that the LightGBM model presents the highest prediction accuracy. Finally, interpretable ML approaches, including Shapely additive explanations (SHAP), partial dependence plot (PDP), and accumulated local effects (ALE), are applied to explain the predictions of the ML model since it is essential to understand the effect of various values of input parameters on objective functions. As a result of interpretability approaches, an optimum gridshell structure is obtained and new opportunities are verified for form-finding investigation of GFRP elastic gridshells during lifting construction.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.217-217
• /
• 2023

상수도관망은 사용자에게 고품질의 물을 안정적으로 공급하는 것을 목적으로 하며, 이를 평가하기 위한 지표 중 하나로 압력을 활용한다. 최근 스마트 센서의 설치가 확장됨에 따라 기계학습기법을 이용한 실시간 데이터 기반의 분석이 활발하다. 따라서 어디에서 데이터를 수집하느냐에 대한 센서 위치 결정이 중요하다. 본 연구는 eXtreme Gradient Boosting(XGBoost) 모델을 활용하여 대규모 상수도관망 내 센서 위치를 최적화하는 방법론을 제안한다. XGBoost 모델은 여러 의사결정 나무(decision tree)를 활용하는 앙상블(ensemble) 모델이며, 오차에 따른 가중치를 부여하여 성능을 향상시키는 부스팅(boosting) 방식을 이용한다. 이는 분산 및 병렬 처리가 가능해 메모리리소스를 최적으로 사용하고, 학습 속도가 빠르며 결측치에 대한 전처리 과정을 모델 내에 포함하고 있다는 장점이 있다. 모델 구현을 위한 독립 변수 결정을 위해 압력 데이터의 변동성 및 평균압력 값을 고려하여 상수도관망을 대표하는 중요 절점(critical node)를 선정한다. 중요 절점의 압력 값을 예측하는 XGBoost 모델을 구축하고 모델의 성능과 요인 중요도(feature importance) 값을 고려하여 센서의 최적 위치를 선정한다. 이러한 방법론을 기반으로 상수도관망의 특성에 따른 경향성을 파악하기 위해 다양한 형태(예를 들어, 망형, 가지형)와 구성 절점의 수를 변화시키며 결과를 분석한다. 본 연구에서 구축한 XGBoost 모델은 추가적인 전처리 과정을 최소화하며 대규모 관망에 간편하게 사용할 수 있어 추후 다양한 입출력 데이터의 조합을 통해 센서 위치 외에도 상수도관망에서의 성능 최적화에 활용할 수 있을 것으로 기대한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.163-163
• /
• 2021

하천에서의 화학사고는 자연적 및 인위적인 원인으로 인해 발생할 수 있으며, 이러한 화학사고가 발생하게 되면 수환경 변화를 야기해 생태계나 인간에게 악영향을 발생시킬 수 있어 신속한 초기대응이 필요하다. 하천으로 유입된 화학물질의 평가에 대한 연구는 활발하게 진행되고 있지만, 화학사고 초기대응을 위한 연구는 미비한 실정이다. 초기대응을 위해서는 현장에서 측정이 용이한 지표를 활용해야하며, 이 지표를 이용해 유출된 화학물질에 대한 정보를 취득 할 수 있어야 한다. 하천의 주요 지점에는 pH 및 EC 등을 실시간으로 측정하는 자동측정망을 운영하고 있는데, 이러한 측정항목들을 지표로 활용한다면 하천 화학사고 대응을 위한 중요한 기초자료로 활용될 수 있을 것이다. 또한 측정된 데이터를 머신러닝 기법을 적용한다면 화학사고 발생 시 초기대응을 위한 기초자료로 활용될 수 있을 것이다. 본 연구에서는 분석한 유해화학물질은 총 26종이며, pH 및 EC를 화학물질들의 특성을 파악하기 위한 대체지표로 선정하였다. 화학물질의 농도변화에 따른 대체지표 변화를 측정하였으며, 실험결과를 바탕으로 성질이 유사한 화학물질들을 Group별로 분류하여 데이터베이스를 구축하였다. 구축된 데이터베이스를 바탕으로 머신러닝 기법인 Decision Tree, Random Forest, Gradient Boosting, XG Boosting에 적용해 각 알고리즘에 대한 성능 평가를 진행하여 가장 우수한 성능의 머신러닝 기법을 선정한다. 본 연구 결과를 바탕으로 선정된 머신러닝 기법을 활용한다면 향수 수환경 화학사고 발생 시 유출된 유해화학물질에 대한 정보를 제공할 수 있으며 그에 따른 신속한 대응의 기초자료로 활용될 수 있을 것으로 판단된다.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.30 no.9
• /
• pp.589-595
• /
• 2017

MOS-FET structured gas sensors were manufactured using MWCNTs for application as NOx gas sensors. As the gas sensors need to be heated to facilitate desorption of the gas molecules, heat dispersion plays a key role in boosting the degree of uniformity of molecular desorption. We report the desorption of gas molecules from the sensor at $150^{\circ}C$ for different sensor electrode gaps (30, 60, and $90{\mu}m$). The COMSOL analysis program was used to verify the process of heat dispersion. For heat analysis, structure of FET gas sensor modeling was proceeded. In addition, a property value of the material was used for two-dimensional modeling. To ascertain the degree of heat dispersion by FEM, the governing equations were presented as partial differential equations. The heat analysis revealed that although a large electrode gap is advantageous for effective gas adsorption, consideration of the heat dispersion gradient indicated that the optimal electrode gap for the sensor is $60{\mu}m$.