• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2023.05a
• /
• pp.100-103
• /
• 2023

Recently, development of maritime autonomoust surface ships and eco-friendly ships, production and evaluation research considering various marine environments is needed in the field of optimal routes as the demand for accurate and detailed real-time marine environment prediction information expands. An algorithm that can calculate the optimal route while reducing the risk of the marine environment and uncertainty in energy consumption in smart ships was developed in 2 stages. In the first stage, a profile was created by combining marine environmental information with ship location and status information within the Automatic Ship Identification System(AIS). In the second stage, a model was developed that could define the marine environment energy map using the configured profile results, A regression equation was generated by applying Random Forest among machine learning techniques to reflect about 600,000 data. The Random Forest coefficient of determination (R²) was 0.89, showing very high reliability. The Dijikstra shortest path algorithm was applied to the marine environment prediction at June 1 to 3, 2021, and to calculate the optimal safety route and express it on the map. The route calculated by the random forest regression model was streamlined, and the route was derived considering the state of the marine environment prediction information. The concept of route calculation based on real-time marine environment prediction information in this study is expected to be able to calculate a realistic and safe route that reflects the movement tendency of ships, and to be expanded to a range of economic, safety, and eco-friendliness evaluation models in the future.

• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• v.3 no.2
• /
• pp.84-96
• /
• 2022

Most of the islands of Korea are distributed in the South and West Sea, and it consists of independent small stream. As a result, the fish community that inhabits the island's stream is isolated from the mainland and other island. This study utilized a Self-Organizing Map (SOM) and a random forest model to analyze the relationship between environmental variables and fish communities inhabiting islands in South Korea. Through the SOM analysis, the fish communities were divided into three clusters, and there were differences in biotic and abiotic factors between these groups. Cluster I consisted of sites with relatively larger island areas and a higher number of species and population. It was found that 15 out of 16 indicator species were included. Meanwhile, the remaining clusters had fewer species and populations. Cluster II, especially, showed the lowest impact from physical variables such as water width and depth. As a result of predicting the species richness using the random forest model, physical variables in habitats, such as stream width and water depth, had a relatively higher importance on species richness. On the other hand, forest area was the most important variables for predicting Shannon diversity, followed by maximum water depth, and gravel. The results suggest that this study can be used as basic data for establishing a stream ecosystem management strategy in terms of conservation and protection of biological resources in streams of islands.

• Proceedings of The Korean Society of Agricultural and Forest Meteorology Conference
• /
• 2017.11a
• /
• pp.43-44
• /
• 2017

• The Pure and Applied Mathematics
• /
• v.24 no.4
• /
• pp.211-226
• /
• 2017

In this paper, machine learning models employed in various fields are discussed and applied to KOSPI200 stock index return forecasting. The results of hyperparameter analysis of the machine learning models are also reported and practical methods for each model are presented. As a result of the analysis, Support Vector Machine and Artificial Neural Network showed a better performance than k-Nearest Neighbor and Random Forest.

• Proceedings of the Korean Society of Disaster Information Conference
• /
• 2023.11a
• /
• pp.257-258
• /
• 2023

본 연구는 2019년부터 2021년의 고속도로 교통사고 위치 좌표를 콘존 데이터와 결합한 후 파이캐럿을 활용하여 고속도로 교통사고 심각도에 영향을 끼치는 요인을 분석할 수 있는 최적 모델을 선정하고 채택된 Random Forest 기법으로 고속도로 교통사고 심각도에 영향을 끼치는 요인을 분석하고자 하였으며, 향후 전국 고속도로 교통사고에 영향을 주는 요인으로 확대하여 분석하고 사고 심각도 개선을 위한 대안 방안 마련이 가능할 것으로 판단된다.

• Journal of Korean Society of Forest Science
• /
• v.18 no.1
• /
• pp.1-7
• /
• 1973

This survey was estimated under the ratio estimate such as single class method, simple random sampling method, compound ratio sampling method, separate ratio sampling method and average tree sampling method artificial forest pitch pine volume. The following results were realized by the ratio estimates. At the above table simple random sampling method and compound ratio sampling method are the only ones which is included the actual stand volume in the ratio estimatedstand volume. It is thought that the sampling was in a such good result was because of stand structual stands were simple forest. The most simple measurement and calcuation on the stand volume estimates, in order, would be (1) single class method, (2) simple random sampling method (3) average tree method (4) separate ratio sampling method and compound ratio sampling method, and at the planted evenaged forest the method has realized the best results in obtaining good accuracy and the measure stand volume with least time, expenses and labor in considerably.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.43 no.3
• /
• pp.397-411
• /
• 2023

The number of deteriorated bridges with a service period of more than 30 years has been rapidly increasing in Korea. Accordingly, the importance of advanced maintenance technologies through the predictions of age-induced deterioration degree, condition, and performance of bridges is more and more noticed. The prediction method of the safety grade of bridges was proposed in this study using the classification models of the Decision Tree and the Random Forest based on machine learning. As a result of analyzing these models for the 8,850 bridges located in national roads with various evaluation indexes such as confusion matrix, balanced accuracy, recall, ROC curve, and AUC, the Random Forest largely showed better predictive performance than that of the Decision Tree. In particular, random under-sampling in the Random Forest showed higher predictive performance than that of other sampling techniques for the C and D grade bridges, with the recall of 83.4%, which need more attention to maintenance because of the significant deterioration degree. The proposed model can be usefully applied to rapidly identify the safety grade and to establish an efficient and economical maintenance plan of bridges that have not recently been inspected.

• Journal of Korean Society of Forest Science
• /
• v.109 no.3
• /
• pp.291-299
• /
• 2020

The purpose of this study was to propose the most cost-effective sampling method, by analyzing the cost of forest resource investigation per sampling method for the planned harvesting area of in Chunyang-myeon, Byeonghwa-gun, Gyeongsangbuk-do, Korea. For this study, three sampling methods were selected: random sampling method, systematic sampling method, and line transect method. For each method, sample size, hourly wage, number of sample points, survey time, travel time, the sample error rate of the estimated average volume, and the desired sampling error rate were used to calculate the cost of forest resource inventories. Thus, 10 sampling points were extracted for each sampling method, and the factors required for cost analysis were calculated via a field survey. As a result, the field survey cost per ha using the random sampling method was found to be have the lowest cost, regardless of the desired sampling error rate, followed by the systematic sampling method, and the line transect method.

• Journal of Biomedical Engineering Research
• /
• v.37 no.2
• /
• pp.61-67
• /
• 2016

This paper presented a method for random forest based the arrhythmia classification using both heart rate (HR) and heart rate variability (HRV) features. We analyzed the MIT-BIH arrhythmia database which contains half-hour ECG recorded from 48 subjects. This study included not only the linear features but also non-linear features for the improvement of classification performance. We classified abnormal ECG using mean_NN (mean of heart rate), SD1/SD2 (geometrical feature of poincare HRV plot), SE (spectral entropy), pNN100 (percentage of a heart rate longer than 100 ms) affecting accurate classification among combined of linear and nonlinear features. We compared our proposed method with Neural Networks to evaluate the accuracy of the algorithm. When we used the features extracted from the HRV as an input variable for classifier, random forest used only the most contributed variable for classification unlike the neural networks. The characteristics of random forest enable the dimensionality reduction of the input variables, increase a efficiency of classifier and can be obtained faster, 11.1% higher accuracy than the neural networks.

• Journal of IKEEE
• /
• v.24 no.1
• /
• pp.9-18
• /
• 2020

The pattern recognition or fuzzy inference, which is mainly used for the development of the automatic walking mode change of the above knee prosthesis, has a disadvantage in that it is difficult to estimate with the immediate change of the walking environment. In order to solve a disadvantage, this paper developed an algorithm that automatically converts the walking mode of the next step by estimating the walking environment at a specific gait phase. Since the proposed algorithm should be implanted and operated in the microcontroller, it is developed using the random forest base in consideration of calculation amount and estimated time. The developed random forest based gait and environmental estimation model were implanted in the microcontroller and evaluated for validity.