• Knowledge Management Research
• /
• v.20 no.3
• /
• pp.155-171
• /
• 2019

In the entertainment industry which has great uncertainty, it is essential to predict public preference first. Thanks to various mass media channels such as cable TV and internet-based streaming services, the reality audition program has been getting big attention every day and it is being used as a new window to new entertainers' debut. This phenomenon means that it is changing from a closed selection process to an open selection process, which delegates selection rights to the public. This is characterized by the popularity of the public being reflected in the selection process. Therefore, this study aims to implement a machine learning model which predicts the winner of , which has recently been popular in South Korea. By doing so, this study is to extend the research method in the cultural industry and to suggest practical implications. We collected the data of winners from the 1st, 2nd, and 3rd seasons of the Produce 101 and implemented the predictive model through the machine learning method with the accumulated data. We tried to develop the best predictive model that can predict winners of by using four machine learning methods such as Random Forest, Decision Tree, Support Vector Machine (SVM), and Neural Network. This study found that the audience voting and the amount of internet news articles on each participant were the main variables for predicting the winner and extended the discussion by analyzing the precision of prediction.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.6
• /
• pp.1119-1127
• /
• 2009

In PGA golf, total prize money and average score are good response variable related to golf skills such as driving distance, green in regulation and putts per green in regulation. But it's not easy to predict the winner of coming tournament. Thus I applied Neural Networks which has pretty good advantages for non-linear complex modeling to binary data. In neural network architectures, I applied NRBF and MLP architecture model for binary data which represent who had a win or not.

• Communications for Statistical Applications and Methods
• /
• v.28 no.4
• /
• pp.329-338
• /
• 2021

Deep learning methods have been developed, used in various fields, and they have shown outstanding performances in many cases. Many studies predicted a daily stock return, a classic example of time-series data, using deep learning methods. We also tried to apply deep learning methods to Korea's stock market data. We used Korea's stock market index (KOSPI) and several individual stocks to forecast daily returns and directions. We compared several deep learning models with other machine learning methods, including random forest and XGBoost. In regression, long short term memory (LSTM) and gated recurrent unit (GRU) models are better than other prediction models. For the classification applications, there is no clear winner. However, even the best deep learning models cannot predict significantly better than the simple base model. We believe that it is challenging to predict daily stock return data even if we use the latest deep learning methods.

• The Journal of the Korea Contents Association
• /
• v.22 no.5
• /
• pp.48-56
• /
• 2022

The importance of data analysis in baseball has been increasing after the success of MLB's Oakland which applied Billy Beane's money ball theory, and the 2020 KBO winner NC Dinos. Various studies using data in baseball has been conducted not only in the United States but also in Korea, In particular, the models using deep learning and machine learning has been suggested. However, in the previous studies using deep learning and machine learning, the focus is only on predicting the win or loss of the game, and there is a limitation in that it is difficult to interpret the results of which factors have an important influence on the game. In this paper, to investigate which factors is important by position, the prediction model for the Golden Glove award which is given for the best player by position is developed. To develop the prediction model, XGBoost which is one of boosting method is used, which also provide the feature importance which can be used to interpret the factors for prediction results. From the analysis, the important factors by position are identified.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.23 no.4
• /
• pp.213-224
• /
• 1998

Since the late 1980s, an Increasing number of neural network models have been studied in the areas of financial prediction and analysis. The purpose of this study is to Investigate the possibility of building a neural network model that is able to construct a profitable trading strategy in the Korean Stock Market. This study classifies stocks into the future market winners and losers from the publicly available accounting information and builds portfolios based on this information. The performances of the winner portfolios and the loser portfolios are compared with each other and against the market index. The empirical result of this research is consistent with the traditional fundamental analysis where it is claimed that the financial statements contain firm values that may not be fully reflected In stock prices without delay. Despite the supporting empirical evidence. It is somewhat Inconclusive as to whether or not the abnormal return in excess of market return is the result of the extra knowledge obtained in the neural network models derived from the historical accounting data. This research attempts to open another avenue using neural network models for searching for evidence against market efficiency where statistics and intuition have played a major role.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.77-84
• /
• 2020

In this study, we conducted a study on the win-loss predicton analysis of korean professional baseball by artificial intelligence models. Based on the model, we predicted the winner as well as each team's final rank in the league. Additionally, we developed a website for viewers' understanding. In each game's first, third, and fifth inning, we analyze to select the best model that performs the highest accuracy and minimizes errors. Based on the result, we generate the rankings. We used the predicted data started from May 5, the season's opening day, to August 30, 2020 to generate the rankings. In the games which Kia Tigers did not play, however, we used actual games' results in the data. KNN and AdaBoost selected the most optimized machine learning model. As a result, we observe a decreasing trend of the predicted results' ranking error as the season progresses. The deep learning model recorded 89% of the model accuracy. It provides the same result of decreasing ranking error trends of the predicted results that we observe in the machine learning model. We estimate that this study's result applies to future KBO predictions as well as other fields. We expect broadcasting enhancements by posting the predicted winning percentage per inning which is generated by AI algorism. We expect this will bring new interest to the KBO fans. Furthermore, the prediction generated at each inning would provide insights to teams so that they can analyze data and come up with successful strategies.

• Genomics & Informatics
• /
• v.17 no.4
• /
• pp.48.1-48.6
• /
• 2019

Over the last decade, genome-wide association studies (GWASs) have provided an unprecedented amount of genetic variations that are associated with various phenotypes. However, previous GWAS were mostly conducted in European populations, and these biased results for non-Europeans may result in a significant reduction in risk prediction for non-Europeans. An issue with the early GWAS was the winner's curse problem, which led to misleading results when constructing the polygenic risk scores (PRS). Therefore, more non-European population-based studies are needed to validate reported variants and improve genetic risk assessment across diverse populations. In this study, we validated 422 variants independently associated with glycemic indexes, liver enzymes, and type 2 diabetes in 125,872 samples from a Korean population, and further validated the results by assessing publicly available summary statistics from European GWAS (n = 898,130). Among the 422 independently associated variants, 284, 320, and 361 variants were replicated in Koreans, Europeans, and either one of the two populations. In addition, the effect sizes for Koreans and Europeans were moderately correlated (r = 0.33-0.68). However, 61 variants were not replicated in both Koreans and Europeans. Our findings provide valuable information on effect sizes and statistical significance, which is essential to improve the assessment of disease risk using PRS analysis.

• Journal of Life Science
• /
• v.22 no.11
• /
• pp.1435-1443
• /
• 2012

The changes of plant communities at Angol valley and Baetgol valley in Geoje-do were studied. Both ${\alpha}$-diversity and ${\gamma}$-diversity decreased with the sizes and degrees of habitat fragmentations. The mean number of winner species decreased with habitat fragmentation. All Quercus species, including Quercus acutissima, significantly declined in importance, while the many introduced species, including Trifolium pretense, all increased in importance. As the proportional change in adult survival rate increased, the absolute value of the bias in the elasticity prediction also increased from 2003 to 2011 at undisturbed forests. However, the bias was low for decreases in disturbed populations. Moran's I values showed overall decreases for habitat fragmentation and for the periods of habitat conversion. Eventually, plant communities, due to urbanization in Goeje-do, might have led to decreased chances of common species when the environment was disturbed.