• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.87-92
• /
• 2021

Bias, variance, error and learning are important factors for performance in modeling a big data based recommendation system. The recommendation model in this system must reduce complexity while maintaining the explanatory diagram. In addition, the sparsity of the dataset and the prediction of the system are more likely to be inversely proportional to each other. Therefore, a product recommendation model has been proposed through learning the similarity between products by using a factorization method of the sparsity of the dataset. In this paper, the generalization ability of the model is improved by applying the max-norm regularization as an optimization method for the loss function of this model. The solution is to apply a stochastic projection gradient descent method that projects a gradient. The sparser data became, it was confirmed that the propsed regularization method was relatively effective compared to the existing method through lots of experiment.

• Journal of Information Science Theory and Practice
• /
• v.10 no.spc
• /
• pp.135-142
• /
• 2022

Recently, with the development of data processing technology and the increase of computational power, methods to solving social problems using Artificial Intelligence (AI) are in the spotlight, and AI technologies are replacing and supplementing existing traditional methods in various fields. Meanwhile in Korea, heavy rain is one of the representative factors of natural disasters that cause enormous economic damage and casualties every year. Accurate prediction of heavy rainfall over the Korean peninsula is very difficult due to its geographical features, located between the Eurasian continent and the Pacific Ocean at mid-latitude, and the influence of the summer monsoon. In order to deal with such problems, the Korea Meteorological Administration operates various state-of-the-art observation equipment and a newly developed global atmospheric model system. Nevertheless, for precipitation nowcasting, the use of a separate system based on the extrapolation method is required due to the intrinsic characteristics associated with the operation of numerical weather prediction models. The predictability of existing precipitation nowcasting is reliable in the early stage of forecasting but decreases sharply as forecast lead time increases. At this point, AI technologies to deal with spatio-temporal features of data are expected to greatly contribute to overcoming the limitations of existing precipitation nowcasting systems. Thus, in this project the dataset required to develop, train, and verify deep learning-based precipitation nowcasting models has been constructed in a regularized form. The dataset not only provides various variables obtained from multiple sources, but also coincides with each other in spatio-temporal specifications.

• Journal of The Geomorphological Association of Korea
• /
• v.18 no.4
• /
• pp.79-96
• /
• 2011

Volcanic landforms are classified into the volcanic edifice produced through constructive processes of eruption and the crater generated by destructive processes of eruption. Both landforms are distributed around Korean Peninsula including attaching islands. However, only a few regions such as Mt. Baekdu, Jeju Island, Ulleung Island, and Chugaryeong, which are closely related with the volcanic eruption occurred during the Quaternary, could be considered as a volcanic landform. It results in categorizing the volcanic landform as an unusual topography in Korea. The study of Korean researchers on the volcanic landform were regularized in 1970s on Jeju Island, in 1980s on Ulleung Island, and in 1990s on Mt. Baekdu, respectively. Oreums and lava tubes in Jeju Island have been also examined since 1980s. Compared with other fields of geomorphology, researches as well as researchers on the volcanic landform are very few in Korea. Geomorphologists are expected to perform an active research in that the volcanic landform of Korea have diverse values.

• The Korean Journal of Applied Statistics
• /
• v.36 no.6
• /
• pp.501-514
• /
• 2023

When analyzing high dimensional data such as text data, if we input all the variables as explanatory variables, statistical learning procedures may suffer from over-fitting problems. Furthermore, computational efficiency can deteriorate with a large number of variables. Dimensionality reduction techniques such as feature selection or feature extraction are useful for dealing with these problems. The sparse principal component analysis (SPCA) is one of the regularized least squares methods which employs an elastic net-type objective function. The SPCA can be used to remove insignificant principal components and identify important variables from noisy observations. In this study, we propose a dimension reduction procedure for text data based on the SPCA. Applying the proposed procedure to real data, we find that the reduced feature set maintains sufficient information in text data while the size of the feature set is reduced by removing redundant variables. As a result, the proposed procedure can improve classification accuracy and computational efficiency, especially for some classifiers such as the k-nearest neighbors algorithm.

• Journal of Korea Technology Innovation Society
• /
• v.16 no.1
• /
• pp.323-345
• /
• 2013

In this paper, problems with the Korean system regulating the use of university royalties are identified and investigated in order to suggest measures to improve the system in a way that provides a better R&D environment at universities. The Delphi technique was used to gather data from royalty specialists at universities and government ministries. The first Delphi survey conducted used open questions to identify problems in the use of university royalties. Then, closed questions were used for the second Delphi survey. The number of responses and the frequency of answers were analyzed after the first survey, and validity, stability, and reliability analyses were conducted for the second survey. The measures suggested to improve the system regulating the use of university royalties are as follows: First, bonuses for researchers, which are currently 50% or more of collected royalties, need to be decreased, as they are rather high compared to similar bonuses in developed countries, which are around 30% of collected royalties. The guideline for limiting the bonuses, which is explained as XX% or less of collected royalties, is suggested to prevent the excessive use of royalties. Second, rewards for those who contribute to technology transfer and commercialization should be increased. It is also important to build a consensus around the need to reward those who contribute to technology transfer and commercialization. Third, the scale of re-investment into R&D needs to increase. Regulations on royalties should be meaningfully applied to create a positive feedback structure for R&D, which can be described as the process of research, R&D outcomes, technology transfer, collecting royalties, rewarding researchers, and re-investing in R&D. To build a university's R&D capability, re-investment into R&D needs to be regularized as XX% or more of royalties. Fourth, regulations on the royalties of ministries and universities need to be unified. Each category for using royalties needs to be regularized, with detailed matters such as the guideline, process and method for using royalties specified. Also, universities need to make their own specific regulations. Fifth, specific priorities on the use of royalties need to be suggested. Regulation is necessary for the categories that do not have guideline and priorities for the use of royalties. It is hoped that the findings of this research will contribute to reinforcing the R&D capability of universities.

• THE INTERNATIONAL COMMERCE & LAW REVIEW
• /
• v.38
• /
• pp.327-376
• /
• 2008

China is marking 9.4% annual growth rate in average since 1978. GDP reached $1090 in 2003 as the first time and China ranked at 4th with their economy size in 2006. One of the remarkable change in China is the extension of foreign open-door policy. China joined WTO in the end of 2001 and it strengthen the foundation of Chinese market economy structure and encouraged the inflow of foreign capital. While 400 of the 500 global corporations advanced into China, the economy trade has been rapidly increasing between Korea and China. The economy trade in both countries has been regularized since 1992 and the annual trade is tending upwards in last 15 years. Korean trade toward China reached 134,400 million which is increased 27 times compared with the total of 1982. In this period, Korean trade toward China marked 24.5% in Export increasing rate and 16.7% in import increasing rate. China became the 2nd biggest export country of Korea in 2001 and became the top in 2003. As the China foreign direct investment has been increasing rapidly, the number of Korean companies advanced into China has been remarkably increasing. By focusing on a thorough review of the nationally published documents of Korean-Chinese business management research during more than two decades (1981-2004), the present paper has been systematically classified and analyzed the current status of Korean-Chinese business management research. The paper raised some important issues regarding Korean-Chinese business management research and predominantly, its future prospects are outlined. In the paper, the documents which are registered in the Korean Academic Processing Foundation registration of journals and candidate registration of journals have been classified by: research purpose, main subject, research method and the results. Careful analysis among the research clarified the active and inactive business management research fields. This clarification enables us to get a better understanding of the current research of Korean-Chinese business management, and more importantly, it pointed out to the direction of future development of research. In addition, the systematic classification made by this paper may contribute to the decision making of subject index of Korean-Chinese business management research since there has been no classification standard of it until now.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.5
• /
• pp.1153-1165
• /
• 2017

Structural equation modeling (SEM) is a basic tool for testing theories in a variety of disciplines. A maximum likelihood (ML) method for parameter estimation is by far the most widely used in SEM. Alternatively, two-stage least squares (2SLS) estimator has been proposed as a more robust procedure to address model misspecification. A regularized extension of 2SLS, two-stage ridge least squares (2SRLS) has recently been introduced as an alternative to ML to effectively handle the small-sample-size issue. However, it is unclear whether and when misspecification and small sample sizes may pose problems in theory testing with 2SLS, 2SRLS, and ML. The purpose of this article is to evaluate the three estimation methods in terms of inferences errors as well as parameter recovery under two experimental conditions. We find that: 1) when the model is misspecified, 2SRLS tends to recover parameters better than the other two estimation methods; 2) Regardless of specification errors, 2SRLS produces small or relatively acceptable Type II error rates for the small sample sizes.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1244-1244
• /
• 2001

In linear discriminant analysis there are two important properties concerning the effectiveness of discriminant function modeling. The first is the separability of the discriminant function for different classes. The separability reaches its optimum by maximizing the ratio of between-class to within-class variance. The second is the stability of the discriminant function against noises present in the measurement variables. One can optimize the stability by exploring the discriminant variates in a principal variation subspace, i. e., the directions that account for a majority of the total variation of the data. An unstable discriminant function will exhibit inflated variance in the prediction of future unclassified objects, exposed to a significantly increased risk of erroneous prediction. Therefore, an ideal discriminant function should not only separate different classes with a minimum misclassification rate for the training set, but also possess a good stability such that the prediction variance for unclassified objects can be as small as possible. In other words, an optimal classifier should find a balance between the separability and the stability. This is of special significance for multivariate spectroscopy-based classification where multicollinearity always leads to discriminant directions located in low-spread subspaces. A new regularized discriminant analysis technique, the principal discriminant variate (PDV) method, has been developed for handling effectively multicollinear data commonly encountered in multivariate spectroscopy-based classification. The motivation behind this method is to seek a sequence of discriminant directions that not only optimize the separability between different classes, but also account for a maximized variation present in the data. Three different formulations for the PDV methods are suggested, and an effective computing procedure is proposed for a PDV method. Near-infrared (NIR) spectra of blood plasma samples from mastitic and healthy cows have been used to evaluate the behavior of the PDV method in comparison with principal component analysis (PCA), discriminant partial least squares (DPLS), soft independent modeling of class analogies (SIMCA) and Fisher linear discriminant analysis (FLDA). Results obtained demonstrate that the PDV method exhibits improved stability in prediction without significant loss of separability. The NIR spectra of blood plasma samples from mastitic and healthy cows are clearly discriminated between by the PDV method. Moreover, the proposed method provides superior performance to PCA, DPLS, SIMCA and FLDA, indicating that PDV is a promising tool in discriminant analysis of spectra-characterized samples with only small compositional difference, thereby providing a useful means for spectroscopy-based clinic applications.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.1042-1042
• /
• 2001

In linear discriminant analysis there are two important properties concerning the effectiveness of discriminant function modeling. The first is the separability of the discriminant function for different classes. The separability reaches its optimum by maximizing the ratio of between-class to within-class variance. The second is the stability of the discriminant function against noises present in the measurement variables. One can optimize the stability by exploring the discriminant variates in a principal variation subspace, i. e., the directions that account for a majority of the total variation of the data. An unstable discriminant function will exhibit inflated variance in the prediction of future unclassified objects, exposed to a significantly increased risk of erroneous prediction. Therefore, an ideal discriminant function should not only separate different classes with a minimum misclassification rate for the training set, but also possess a good stability such that the prediction variance for unclassified objects can be as small as possible. In other words, an optimal classifier should find a balance between the separability and the stability. This is of special significance for multivariate spectroscopy-based classification where multicollinearity always leads to discriminant directions located in low-spread subspaces. A new regularized discriminant analysis technique, the principal discriminant variate (PDV) method, has been developed for handling effectively multicollinear data commonly encountered in multivariate spectroscopy-based classification. The motivation behind this method is to seek a sequence of discriminant directions that not only optimize the separability between different classes, but also account for a maximized variation present in the data. Three different formulations for the PDV methods are suggested, and an effective computing procedure is proposed for a PDV method. Near-infrared (NIR) spectra of blood plasma samples from daily monitoring of two Japanese cows have been used to evaluate the behavior of the PDV method in comparison with principal component analysis (PCA), discriminant partial least squares (DPLS), soft independent modeling of class analogies (SIMCA) and Fisher linear discriminant analysis (FLDA). Results obtained demonstrate that the PDV method exhibits improved stability in prediction without significant loss of separability. The NIR spectra of blood plasma samples from two cows are clearly discriminated between by the PDV method. Moreover, the proposed method provides superior performance to PCA, DPLS, SIMCA md FLDA, indicating that PDV is a promising tool in discriminant analysis of spectra-characterized samples with only small compositional difference.

• Cartoon and Animation Studies
• /
• s.51
• /
• pp.225-254
• /
• 2018

For the domestic web-toon industry, a new industrial environment has been created by the growth various domestic and international factors, such as technology development, the increase of platforms, charged web-toon, and globalization. Although the role of the agency has not been emphasized at the formation and introduction time of the web-toon, the role growth of the agency has been regularized with the expanding of the web-toon industry. The expending web-toon industry is due to the introduction of smart-phone, by which people can consume the web-toon without any limit of time and place, so the web-toon industry has been growing rapidly. Thereafter the production of the second contents using the web-toon IP has been actively carried out and the activity area of the web-toon artist has been diversified, so the necessity of the specialized web-toon agency has begun to be brought in earnest. In this study, the web-toon industry has been, firstly, classified as transition period of cartoon industry, foothold formation period of web-toon, introduction period of web-toon industry, expansion period, and maturity period according to the industrial life cycle theory. As a result, the web-toon agency has been emerged from the expansion period of web-toon industry, and the role of agency, to prevent web-toon industry entering into declining period from its maturity period, has been researched in this study.