Yong-Hyuk Kim’s research while affiliated with Kwangwoon University and other places

... Our study presents two assignment strategies: one minimizing work hours based on South Korea's regulations and another minimizing skimmer usage. To determine these assignments, we employ a genetic algorithm (GA) [4,5], which has been representative among swarm and evolutionary computation methods [6], and compare the results with current practices in South Korea through simulations. However, the complexity and timeintensiveness of simulation-based evaluations necessitate alternative approaches. ...

Reference:

Evolutionary Approach to Optimal Oil Skimmer Assignment for Oil Spill Response: A Case Study

... Therefore, in our second study, we propose a surrogate model that employs a deep neural network (DNN) [7,8] to expedite resource allocation optimization. Building upon prior research by Shin and Kim [9], we aim to substitute simulation-based evaluations with an efficient surrogate model, comparing its performance and optimization results to precedent studies. ...

Reference:

Evolutionary Approach to Optimal Oil Skimmer Assignment for Oil Spill Response: A Case Study

... 1. Heuristic Measure using Multiple Surrogates: The advantage of using ensembles that employ multiple surrogates is that it is possible to address the degradation problem of a single surrogate, and the predicted performance variance helps avoid false optima [28]. In this study, we designed multiple homogeneous surrogates that combine classifiers and regressors designed using the same training data. ...

Reference:

Optimal Agent Search Using Surrogate-Assisted Genetic Algorithms

... In the field of Microarray data classification, a diverse range of methods have been employed for the purpose of selecting pertinent attributes and ascertaining the ideal number of features from the extensive initial dataset [ Prajapati and Gourisaria (2023)]. Microarray data has never been easy to categorize due to its huge dimensionality [Kim and Yoon (2022)]. These days, researchers can look at a single experiment done with over the genes 1000 because of the advancement of microarray data technology. ...

Reference:

FEATURE SELECTION AND CLASSIFICATION FOR MICROARRAY DATA USING UPGRADE CHI-SQUARE TEST

... When they incorporate a simple neighborhood search into each of the five metaheuristics, in addition to improved solution quality, there was no statistically significant difference among the results for these five metaheuristics. Yang et al. (2021) developed a memetic (a genetic algorithm combined with local search) algorithm designed specifically to solve MMKPs that have few feasible solutions. Experimental results were based on modified multi-dimensional knapsack problems (MKPs) from Beasley's OR-Library. ...

Reference:

Simple strategies that generate bounded solutions for the multiple-choice multi-dimensional knapsack problem: a guide for OR practitioners

... Researchers have sought to establish more accurate price prediction models to enhance price prediction accuracy. Methodologies range from mathematical models such as linear regression and statistical analysis (Ma et al., 2016;Bhargava et al., 2017;Chandrasekaran et al., 2021;Amik et al., 2021;Cho et al., 2021;Zheng and He, 2021) to alternative techniques employing artificial intelligence, such as expert systems (Kim and Kim, 2010;Kim et al., 2005)), neural networks (Hsu, 2011;Kim et al., 2005;Bhargava et al., 2017;Juszczyk, 2017;Odeck, 2019;Hao et al., 2020;Amik et al., 2021;Zhang and Ma, 2021), and deep learning Wang and Wang (2020); Haq et al. (2021); Zheng and He (2021); Critien et al. (2022).Hence, forecasting prices is essential for every stakeholder in the business to mitigate risks and make well-informed decisions. ...

Reference:

A novel decision fusion approach for sale price prediction using Elastic Net and MOPSO

... Filter-based feature selection [31][32][33] has the advantage of deriving feature subsets by identifying correlations between features within a relatively short time; however, it has the disadvantage that it may be difficult to quantify relevance and redundancy between selected features. In this study, a new fitness function was devised to emphasize the advantages and make up for the disadvantages. ...

Reference:

Genetic Feature Selection Applied to KOSPI and Cryptocurrency Price Prediction

... Due to uncertainties in high mobility environments like maritime situations [15]- [18], a number of research projects focus on maximizing sensors' coverage area or on planning paths for sensors to accurately sense targets. Specifically, [19] attempts to maximize sensor coverage in wireless sensor networks (WSN) using a local search-based algorithm. In addition, the authors mathematically determine the upper and lower bounds of coverage deployment. ...

Reference:

Sensor Allocation and Online-Learning-Based Path Planning for Maritime Situational Awareness Enhancement: A Multi-Agent Approach

... For Lagrangian applications in the ocean, deep learningbased surrogate models have been used to mimic, for example, the evolution of virtual particle density maps at the ocean surface (Jenkins et al., 2023) and oil slick transport (Janati et al., 2020). DLMs have been applied not only to replicate outputs from numerical simulations but also to emulate observations, like in applications dedicated to learning drifter trajectories (Nam et al., 2020;Aksamit et al., 2020;Shen et al., 2022;Botvynko et al., 2023). In all these cases, currents and other potentially relevant variables were used to feed their DLMs resulting in remarkably good results. ...

Reference:

Efficient deep learning surrogate method for predicting the transport of particle patches in coastal environments

... Our TLCA method considers the time lags present in atmospheric time-series data, which have been noted in prior research (Kawale et al., 2012). Although there have been limited studies on the impact of time-lagged effects on feature selection (Colabone et al., 2015;Kim et al., 2020), identifying time-lagged linkages is a critical task. Therefore, examining time-lagged relationships could greatly enhance the accuracy of sea fog forecasting. ...

Reference:

Intelligent model for offshore China sea fog forecasting