• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.26 no.1
• /
• pp.23-48
• /
• 2022

In this paper, we provide predictive models for the market price of fruits, and analyze the performance of each fruit price predictive model. The data used to create the predictive models are fruit price data, weather data, and Korea composite stock price index (KOSPI) data. We collect these data through Open-API for 10 years period from year 2011 to year 2020. Six types of fruit price predictive models are constructed using the LSTM algorithm, a special form of deep learning RNN algorithm, and the performance is measured using the root mean square error. For each model, the data from year 2011 to year 2018 are trained to predict the fruit price in year 2019, and the data from year 2011 to year 2019 are trained to predict the fruit price in year 2020. By comparing the fruit price predictive models of year 2019 and those models of year 2020, the model with excellent efficiency is identified and the best model to provide the service is selected. The model we made will be available in other countries and regions as well.

• International journal of advanced smart convergence
• /
• v.13 no.2
• /
• pp.130-135
• /
• 2024

The Stable diffsuion AI tool is popular among designers because of its flexible and powerful image generation capabilities. However, due to the diversity of its AI models, it needs to spend a lot of time testing different AI models in the face of different design plans, so choosing a suitable general AI model has become a big problem at present. In this paper, by comparing the AI images generated by two different Stable diffsuion models, the advantages and disadvantages of each model are analyzed from the aspects of the matching degree of the AI image and the prompt, the color composition and light composition of the image, and the general AI model that the generated AI image has an aesthetic sense is analyzed, and the designer does not need to take cumbersome steps. A satisfactory AI image can be obtained. The results show that Playground V2.5 model can be used as a general AI model, which has both aesthetic and design sense in various style design requirements. As a result, content designers can focus more on creative content development, and expect more groundbreaking technologies to merge generative AI with content design.

• ETRI Journal
• /
• v.27 no.1
• /
• pp.64-74
• /
• 2005

Traditional Internet pricing schemes are coming under continual pressure to adapt to, and encourage, a changing mix of Internet applications and consumer usage patterns. Much research effort over the last decade has been focused on developing more efficient and attractive charging schemes. However, none of the proposed models has been widely deployed. This raises questions regarding the inhibiting factors and missing pieces that make pricing the Internet such a challenge. In this paper, we discuss the problems with current Internet pricing schemes, review the history of Internet pricing research over the last ten years, and summarize the key features and motivations of the most significant models. We develop a novel visual approach to comparing and evaluating such schemes using a three-dimensional (3D) metric encompassing technical efficiency, economic efficiency, and social impact. We address and discuss the important factors that have inhibited the deployment of the reviewed models and suggest productive areas of focus for future Internet pricing research.

• Computers and Concrete
• /
• v.14 no.4
• /
• pp.387-403
• /
• 2014

The aim of this study is to assess the structural performance of deteriorated reinforced concrete bridge piers, and to provide method for developing improved evaluation method. For a deteriorated bridge piers, once the cover spalls off and bond between the reinforcement and concrete has been lost, compressed reinforcements are likely to buckle. By using a sophisticated nonlinear finite element analysis program, the accuracy and objectivity of the assessment process can be enhanced. A computer program, RCAHEST (Reinforced Concrete Analysis in Higher Evaluation System Technology), is used to analyze reinforced concrete structures. Material nonlinearity is taken into account by comprising tensile, compressive and shear models of cracked concrete and a model of reinforcing steel. Advanced deteriorated material models are developed to predict behaviors of deteriorated reinforced concrete. The proposed numerical method for the structural performance assessment of deteriorated reinforced concrete bridge piers is verified by comparing it with reliable experimental results. Additionally, the studies and discussions presented in this investigation provide an insight into the key behavioral aspects of deteriorated reinforced concrete bridge piers.

• Genomics & Informatics
• /
• v.14 no.4
• /
• pp.149-159
• /
• 2016

With the success of the genome-wide association studies (GWASs), many candidate loci for complex human diseases have been reported in the GWAS catalog. Recently, many disease prediction models based on penalized regression or statistical learning methods were proposed using candidate causal variants from significant single-nucleotide polymorphisms of GWASs. However, there have been only a few systematic studies comparing existing methods. In this study, we first constructed risk prediction models, such as stepwise linear regression (SLR), least absolute shrinkage and selection operator (LASSO), and Elastic-Net (EN), using a GWAS chip and GWAS catalog. We then compared the prediction accuracy by calculating the mean square error (MSE) value on data from the Korea Association Resource (KARE) with body mass index. Our results show that SLR provides a smaller MSE value than the other methods, while the numbers of selected variables in each model were similar.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.354-359
• /
• 2006

The analysis of large-scale water resources systems is often complicated by the presence of multiple reservoirs and diversions, the uncertainty of unregulated inflows and demands, and conflicting objectives. Reinforcement learning is presented herein as a new approach to solving the challenging problem of stochastic optimization of multi-reservoir systems. The Q-Learning method, one of the reinforcement learning algorithms, is used for generating integrated monthly operation rules for the Keum River basin in Korea. The Q-Learning model is evaluated by comparing with implicit stochastic dynamic programming and sampling stochastic dynamic programming approaches. Evaluation of the stochastic basin-wide operational models considered several options relating to the choice of hydrologic state and discount factors as well as various stochastic dynamic programming models. The performance of Q-Learning model outperforms the other models in handling of uncertainty of inflows.

• New & Renewable Energy
• /
• v.2 no.2 s.6
• /
• pp.37-43
• /
• 2006

In this paper, the first forecasting system of wind power generation, KIER Forecaster is presented. KIER Forecaster has been constructed based on statistical models and was trained with wind speed data observed at Gosan Weather Station nearby Walryong Site. Due to short period of measurements at Walryong Site for training the model, Gosan wind data were substituted and transplanted to Walryong Site by using Measure-Correlate-Predict(MCP) technique. The results of One to Three-hour advanced forecasting models are consistent with the measurement at Walryong site. In particular, the multiple regression model by classification of wind speed pattern, which has been developed in this work, shows the best performance comparing with neural network and auto-regressive models.

• International Journal of Concrete Structures and Materials
• /
• v.6 no.4
• /
• pp.221-230
• /
• 2012

Cracking of repair material due to restraint of shrinkage could hinder the intended extension of serviceability of repaired concrete structure. The availability of model to predict shrinkage stress under restraint condition will be useful to assess whether repair material with particular deformation properties is resistance to cracking or not. The accuracy in the prediction will depend upon reliability of the model, input parameters, testing methods used to characterize the input parameters, etc. This paper reviews a variety of models to predict shrinkage stress in patch repair system. Effect of creep and composite action to release shrinkage stress in the patch repair system are quantified and discussed. Accuracy of the models is examined by comparing predicted and measured shrinkage stress. Simplified model to estimate shrinkage stress is proposed which requires only shrinkage property of repair material as an input parameter.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.11
• /
• pp.1005-1010
• /
• 2012

We propose a novel approach to classify fingerprints using the extracted adaptive core block for improving classification performance of incomplete fingerprints in this paper. We compute representative directions from fingerprint images by the block unit and learn horizontal and vertical Markov models by deciding the center position of a fingerprint image based on the expert knowledge. The center block of a test image is the block has the highest probability after comparing the Markov model with $11{\times}11$ blocks. The proposed approach can effectively classify incomplete fingerprints using the optimal center block.

• Korean Journal of Computational Design and Engineering
• /
• v.17 no.4
• /
• pp.294-302
• /
• 2012

Process mining is a useful methodology that can be used for extracting user patterns in log files in order to discover efficient or inefficient processes in organizations. In general, it is used to find and reduce differences between pre-defined processes and actually executed processes in an organization. In this paper, we propose a method to improve processes in PDM/PLM systems based on process mining. In order to improve and detect the inefficient processes, we gathered event logs from PDM/PLM systems and derived process models using several process mining techniques such as ${\alpha}$-algorithm mining, heuristics mining, and fuzzy miner. By comparing original process models with process mining results, it is possible to detect differences between predefined processes and real ones; thereby we can build improved process models for future application.