• Journal of the Korea Society of Computer and Information
• /
• v.28 no.8
• /
• pp.67-75
• /
• 2023

Stock price prediction is an important topic extensively discussed in the financial market, but it is considered a challenging subject due to numerous factors that can influence it. In this research, performance was compared and analyzed by applying time series prediction models (LSTM, GRU) and non-time series prediction models (RF, SVR, KNN, LGBM) that do not take into account the temporal dependence of data into stock price prediction. In addition, various data such as stock price data, technical indicators, financial statements indicators, buy sell indicators, short selling, and foreign indicators were combined to find optimal predictors and analyze major factors affecting stock price prediction by industry. Through the hyperparameter optimization process, the process of improving the prediction performance for each algorithm was also conducted to analyze the factors affecting the performance. As a result of feature selection and hyperparameter optimization, it was found that the forecast accuracy of the time series prediction algorithm GRU and LSTM+GRU was the highest.

• Biomedical Science Letters
• /
• v.23 no.2
• /
• pp.118-127
• /
• 2017

Elevated serum uric acid and resting heart rate are risk factors and predictors of metabolic syndrome. However, few studies have examined the optimal cutoff value for serum uric acid and resting heart rate to predict metabolic syndrome in Korean adults. Subjects for this study were 22,302 adults (average age 45 years old), who underwent health screening examination from January 2010 to December 2012 at the Health Promotion Center of one hospital in Gyeonggi-do for general health check-up. The uric acid and resting heart rate cutoff values were calculated by ROC analysis for metabolic syndrome. Elevated serum uric acid and resting heart rate were associated with an increased prevalence of metabolic syndrome in Korean adults. The optimal cutoff value for uric acid level to predict metabolic syndrome in adults was 4.95 mg/dL (male 6.35, female 4.55) and optimal cutoff value for resting heart rate to predict metabolic syndrome was 68 beats per minute (male 66, female 68). However, serum uric acid and resting heart rate were found to have limitations for the diagnosis of metabolic syndrome.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.2
• /
• pp.193-200
• /
• 2022

We conducted a case study to determine the optimal model parameters and predictors of Long Short-Term Memory (LSTM) for the reproduction of dissolved oxygen (DO) concentration in Jinhae Bay. The model parameter case study indicated the lowest accuracy when the Hidden node=10, Epoch=100. This was caused by underfitting of machine learning. The accuracy increased as the Hidden node and Epoch increased. The accuracy was the highest when the Hidden node=80 and Epoch=100 with R²=0.99. In the bottom DO reproduction of Step 1 of the predictors case study, accuracy was highest when the water temperature was used as a predictor with R²=0.81. In Step 2, The R² value increased up to 0.92 when the water temperature and SiO₂ were used as a predictor. This was caused by a high correlation between the bottom DO and SiO₂ concentrations. Consequently, we determined the optimal model parameters and predictors of LSTM for the reproduction of DO concentration in Jinhae Bay.

• Journal of Preventive Medicine and Public Health
• /
• v.38 no.2
• /
• pp.203-207
• /
• 2005

Objectives: Recently, the American Diabetes Association (ADA) redefined the criteria of prediabetes, which has lowered the diagnostic level of fasting plasma glucose (FPG) from 110 to 125 mg/dl, down to levels between 100 to 125mg/dl. The purpose of this study was to determine the predictive cutoff level of FPG as a risk for the development of diabetes mellitus in Korean men. Methods: A retrospective cohort study was conducted on 11,423 (64.5%) out of 17,696 males $\leq$30 years of age, and who met the FPG of $\leq$125 mg/dl and hemoglobin A1c of $\leq$ 6.4% criteria, without a history of diabetes, and who were enrolled at the screening center of a certain university hospital between January and December 1999. The subjects were followed from January 1999 to December 2002 (mean follow-up duration; 2.3(${\pm}0.7$) years). They were classified as normal (FPG <100mg/dl), high glucose (FPG $\geq$100mg/dl and <110mg/dl) and impaired fasting glucose (FPG $\geq$110mg/dl and $\leq$125mg/dl) on the basis of their fasting plasma glucose level measured in 1999. We compared the incidence of diabetes between the 3 groups by performing Cox proportional hazards model and used receiver operating characteristic analyses of the FPG level, in order to estimate the optimal cut-off values as predictors of incident diabetes. Results: At the baseline, most of the study subjects were in age in their 30s to 40s (mean age, 41.8(${\pm}7.1$) year). The incidence of diabetes mellitus in this study was 1.19 per 1,000 person-years (95% CI=0.68-1.79), which was much lower than the results of a community-based study that was 5.01 per 1,000 person-years. The relative risks of incident diabetes in the high glucose and impaired fasting glucose groups, compared with the normal glucose group, were 10.3 (95% CI=2.58-41.2) and 95.2 (95% CI= 29.3-309.1), respectively. After adjustment for age, body mass index, and log triglyceride, a FPG greater than 100mg/dl remained significant predictors of incident diabetes. Using the receiver operating characteristic (ROC) curve, the optimal cutoff level of FPG as a predictor of incident diabetes was 97.5 mg/dl, with a sensitivity and a specificity of 81.0% and 86.0%, respectively. Conclusion: These results suggest that lowering the criteria of impaired fasting glucose is needed in Korean male adults. Future studies on community-based populations, including women, will be required to determine the optimal cutoff level of FPG as a predictor of incident diabetes.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.23 no.6
• /
• pp.1482-1489
• /
• 1998

This paper aims to extract characteristics of the spectrum of fingerprint image and to apply them to image enhancement techniques in spatial frequency domain. Based on 1$\times$64 window as a processing unit and the different record lengths(32, 16, 8), the estimate of power spectrum density for each length was made. Each acquired spectrum characteristics was applied to the re-synthesis process of the fingerprint image, an improved gray scale image was obtained. In order to select an optimal predictor and the Huffman table for the fingerprint iamge, the lossless JPEG algorithm was used. Experiments were performed for extracting distribution characteristics for the each of 7 predictors from the fingerprint image and modeling processes, and the result was applied to the data compression algorithm and the selection of the optimal predictor.

• Biomedical Science Letters
• /
• v.25 no.3
• /
• pp.256-266
• /
• 2019

In this study, obese adults were compared for their ability to predict obesity and lipid related variables and their optimal cutoff values to predict metabolic syndrome and insulin resistance. In this study, 9,256 adults aged 20 years or older and less than 80 years old, who were in the Gyeonggi region from January 2014 to December 2016 and who were examined at a general hospital, were enrolled. The diagnostic criteria for obesity were WHO (World Health Organization), and BMI $25kg/m^2$ or more presented in the Asia-Pacific region. Metabolic syndrome was diagnosed based on the criteria of American Heart Association / National Heart, Lung, and Blood Institute (AHA / NHLBI). According to the results of receiver operating characteristic curve (ROC) analysis, Triglyceride / HDL-cholesterol (TG / HDL-C), Triglyceride and Glucose (TyG) index, lipid accumulation product (LAP) and visceral adiposity index (VAI) showed high predictive power for diagnosing metabolic syndrome. The diagnostic accuracy of LAP (AUC: 0.854) for males and VAI (0.888) for females was the highest. The optimal cutoff value of LAP was 42.71 for male and 35.44 for female, and the cutoff value of VAI was 1.92 for male and 2.15 for female. In addition, WHtR (waist to height ratio), TyG index, and LAP were used as predictors of insulin resistance in obese adults. Therefore, LAP and VAI were superior to other indicators in predicting metabolic syndrome in obese adults.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.27 no.2
• /
• pp.1-9
• /
• 2004

This study aims to examine the relationship between total quality management(TQM) practices and operational performance perceived by employees who are in different functional organization types, R&D function and Non-R&D function organization in manufacturing companies. Operational performance is defined as the degree of operation efficiency in the perspective of quality, delivery, and cost. Our study showed that the significant elements of TQM practice were different depending on functional organization types. In case of R&D organization, the categories of customer focus and process management were the strongest significant predictors of operational performance. Therefore, the optimal TQM practices for R&D organization can be obtained in consideration of customer focus and process management.

• Journal of Advanced Marine Engineering and Technology
• /
• v.22 no.2
• /
• pp.174-180
• /
• 1998

Robust stability of discrete-time regulators which utilize state predictors to compensate computation delays is considered. Novel expressions for the return difference matrices and the complementary sensitivity matrices at the input and the output of the regulator are found to obtain simple bounds for unstructured perturbations. Robust stability for pertubations of the system matrix and /or the gain matrix is also considered. under certain restriction on the nominal system simple bounds for the pertubations are obtained directly from the characteristic equation. It is shown that as far as the effect of the computation delays concerns these bounds have explicit relation to those for the unstructured pertubations.

• Journal of Gastric Cancer
• /
• v.23 no.1
• /
• pp.182-193
• /
• 2023

Presently, surgery is the only treatment approach for gastric cancer and improving the prognosis of locally advanced gastric cancer is one of the key factors in promoting gastric cancer survival benefit. The MAGIC study was the first to demonstrate the efficacy of neoadjuvant chemotherapy (NAC) in European countries. In recent years, several clinical trials have provided evidence for the use of NAC in Asian patients with locally advanced gastric cancer. However, clinical practice guidelines vary between Asian and non-Asian populations. Optimal NAC regimens, proper target populations, and predictors of NAC outcomes in Asian patients are still under investigation. Herein, we summarized the current progress in the administration of NAC in Asian patients with gastric cancer.

• Communications for Statistical Applications and Methods
• /
• v.29 no.6
• /
• pp.629-640
• /
• 2022

Variable selection is one of the most crucial tasks in supervised learning, such as regression and classification. The best subset selection is straightforward and optimal but not practically applicable unless the number of predictors is small. In this article, we propose directly solving the best subset selection via the genetic algorithm (GA), a popular stochastic optimization algorithm based on the principle of Darwinian evolution. To further improve the variable selection performance, we propose to run multiple GA to solve the best subset selection and then synthesize the results, which we call ensemble GA (EGA). The EGA significantly improves variable selection performance. In addition, the proposed method is essentially the best subset selection and hence applicable to a variety of models with different selection criteria. We compare the proposed EGA to existing variable selection methods under various models, including linear regression, Poisson regression, and Cox regression for survival data. Both simulation and real data analysis demonstrate the promising performance of the proposed method.