• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.1-17
• /
• 2019

Because stock price forecasting is an important issue both academically and practically, research in stock price prediction has been actively conducted. The stock price forecasting research is classified into using structured data and using unstructured data. With structured data such as historical stock price and financial statements, past studies usually used technical analysis approach and fundamental analysis. In the big data era, the amount of information has rapidly increased, and the artificial intelligence methodology that can find meaning by quantifying string information, which is an unstructured data that takes up a large amount of information, has developed rapidly. With these developments, many attempts with unstructured data are being made to predict stock prices through online news by applying text mining to stock price forecasts. The stock price prediction methodology adopted in many papers is to forecast stock prices with the news of the target companies to be forecasted. However, according to previous research, not only news of a target company affects its stock price, but news of companies that are related to the company can also affect the stock price. However, finding a highly relevant company is not easy because of the market-wide impact and random signs. Thus, existing studies have found highly relevant companies based primarily on pre-determined international industry classification standards. However, according to recent research, global industry classification standard has different homogeneity within the sectors, and it leads to a limitation that forecasting stock prices by taking them all together without considering only relevant companies can adversely affect predictive performance. To overcome the limitation, we first used random matrix theory with text mining for stock prediction. Wherever the dimension of data is large, the classical limit theorems are no longer suitable, because the statistical efficiency will be reduced. Therefore, a simple correlation analysis in the financial market does not mean the true correlation. To solve the issue, we adopt random matrix theory, which is mainly used in econophysics, to remove market-wide effects and random signals and find a true correlation between companies. With the true correlation, we perform cluster analysis to find relevant companies. Also, based on the clustering analysis, we used multiple kernel learning algorithm, which is an ensemble of support vector machine to incorporate the effects of the target firm and its relevant firms simultaneously. Each kernel was assigned to predict stock prices with features of financial news of the target firm and its relevant firms. The results of this study are as follows. The results of this paper are as follows. (1) Following the existing research flow, we confirmed that it is an effective way to forecast stock prices using news from relevant companies. (2) When looking for a relevant company, looking for it in the wrong way can lower AI prediction performance. (3) The proposed approach with random matrix theory shows better performance than previous studies if cluster analysis is performed based on the true correlation by removing market-wide effects and random signals. The contribution of this study is as follows. First, this study shows that random matrix theory, which is used mainly in economic physics, can be combined with artificial intelligence to produce good methodologies. This suggests that it is important not only to develop AI algorithms but also to adopt physics theory. This extends the existing research that presented the methodology by integrating artificial intelligence with complex system theory through transfer entropy. Second, this study stressed that finding the right companies in the stock market is an important issue. This suggests that it is not only important to study artificial intelligence algorithms, but how to theoretically adjust the input values. Third, we confirmed that firms classified as Global Industrial Classification Standard (GICS) might have low relevance and suggested it is necessary to theoretically define the relevance rather than simply finding it in the GICS.

• Journal of the Korean Electrochemical Society
• /
• v.24 no.3
• /
• pp.65-75
• /
• 2021

Coin cell is a basic testing platform for battery research, discovering new materials and concepts, and contributing to fundamental research on next-generation batteries. Li metal batteries (LMBs) are promising since a high energy density (~500 Wh kg^-1) is deliverable far beyond Li-ion. However, Li dendrite-triggered volume fluctuation and high surface cause severe deterioration of performance. Given that such drawbacks are strongly dependent on the cell parameters and structure, such as the amount of electrolyte, Li thickness, and internal pressure, reliable Li metal coin cell testing is challenging. For the LMB-specialized coin cell testing platform, this study suggests the optimal coin cell structure that secures performance and reproducibility of LMBs under stringent conditions, such as lean electrolyte, high mass loading of NMC cathode, and thinner Li use. By controlling the cathode/anode (C/A) area ratio closer to 1.0, the inactive space was minimized, mitigating the cell degradation. The quantification and imaging of inner cell pressure elucidated that the uniformity of the pressure is a crucial matter to improving performance reliability. The LMB coin cells exhibit better cycling retention and reproducibility under higher (0.6 MPa → 2.13 MPa) and uniform (standard deviation: 0.43 → 0.16) stack pressure through the changes in internal parts and introducing a flexible polymer (PDMS) film.

• (The) Research of the performance art and culture
• /
• no.33
• /
• pp.329-354
• /
• 2016

Hyupryulrang was the position that announced the start and end of music in royal ceremonies. It appeared when the royal etiquette was categorized and implemented due to the five etiquette system, which was formed by the influence of Confucianism. Confucianism valued etiquette and music and this aspect was reflected in royal five etiquette, making music involved in royal ceremonies. So there was a need to have a mediator who will announce the insertion of music according to the process of royal ceremonies. For harmonious realization of royal ceremonies and music, hyupryulrang was indispensible. In Korea hyupryulrang appeared in Goryo era and lasted until Joseon era. Hyupryulrang during Joseon was handled by bongsanshi and once was taken by jeonak( 典樂) temporarily but finally was managed by officials in jang-akwon(掌樂院). Among the officials in jang-akwon, jang-akwon jeong(正) mainly served the role but jang-akwon chumjeong(僉正) and jang-akwon juboo(主簿) were sometimes recruited for the role according to circumstances. What was common among jang-akwon jeong, chumjeong, and juboo was that they were all danghakwan(堂下官). Danghakwan was an official who had the fundamental limitation of not being able to participate in policy making so was in a lower position compared to dangsangkwan. Meanwhile, according to circumstances of ceremonial process or the characteristics of ceremonies, gyeraseonjeonkwan(啓螺宣傳官), mushingyungseonjeonkwan(武臣兼宣傳官), and yeojipsa(女執事) were recruited as hyupryulrang instead of officials of jang-akwon, so that there would be no problems in ceremonies and performance of music. The activities of hyupryulrang can be summarized as setting up or laying down hui in most ceremonies that involved band. At night, however, as hui(麾) was invisible, jochok(照燭) or sometimes geumgogi(金鼓旗) was used. As for the term that referred to hyupryulrang, in case of royal banquet, the names of the ceremonial tools were borrowed such as geohuichabi(擧麾差備) and jochokchabi(照燭差備). The location of hyupryulrang was in the west on top of seogye(西階) facing toward the east, which was a position where hyupryulrang could watch the ceremonial process easily and be close to the band. That is, it was a position where one can see the space of ceremony and the space of music at the same time. Also, hyupryulrang was involved in musical parts related to ceremonies such as rehearsals, arrangement of the band, controlling the speed of music, and prevention of missing any musical pieces, and was in charge of such tasks. Hyupryulrang, who had to take charge of music in accordance with ceremonial procedure, was a mediator between royal ceremonies and music.

• Journal of the Korean Geotechnical Society
• /
• v.38 no.7
• /
• pp.5-24
• /
• 2022

The current performance evaluation of slope anchors qualitatively determines the physical bonding between the anchor head and ground as well as cracks or breakage of the anchor head. However, such performance evaluation does not measure these primary factors quantitatively. Therefore, the time-dependent management of the anchors is almost impossible. This study is an evaluation of the 3D numerical model by SfM which combines UAS images with terrestrial LiDAR to collect numerical data on the damage factors. It also utilizes the data for the quantitative maintenance of the anchor system once it is installed on slopes. The UAS 3D model, which often shows relatively low precision in the z-coordinate for vertical objects such as slopes, is combined with terrestrial LiDAR scan data to improve the accuracy of the z-coordinate measurement. After validating the system, a field test is conducted with ten anchors installed on a slope with arbitrarily damaged heads. The damages (such as cracks, breakages, and rotational displacements) are detected and numerically evaluated through the orthogonal projection of the measurement system. The results show that the introduced system at the resolution of 8K can detect cracks less than 0.3 mm in any aperture with an error range of 0.05 mm. Also, the system can successfully detect the volume of the damaged part, showing that the maximum damage area of the anchor head was within 3% of the original design guideline. Originally, the ground adhesion to the anchor head, where the z-coordinate is highly relevant, was almost impossible to measure with the UAS 3D numerical model alone because of its blind spots. However, by applying the combined system, elevation differences between the anchor bottom and the irregular ground surface was identified so that the average value at 20 various locations was calculated for the ground adhesion. Additionally, rotation angle and displacement of the anchor head less than 1" were detected. From the observations, the validity of the 3D numerical model can obtain quantitative data on anchor damage. Such data collection can potentially create a database that could be used as a fundamental resource for quantitative anchor damage evaluation in the future.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.249-263
• /
• 2023

Collaborative Filtering, a representative recommendation system methodology, consists of two approaches: neighbor methods and latent factor models. Among these, the latent factor model using matrix factorization decomposes the user-item interaction matrix into two lower-dimensional rectangular matrices, predicting the item's rating through the product of these matrices. Due to the factor vectors inferred from rating patterns capturing user and item characteristics, this method is superior in scalability, accuracy, and flexibility compared to neighbor-based methods. However, it has a fundamental drawback: the need to reflect the diversity of preferences of different individuals for items with no ratings. This limitation leads to repetitive and inaccurate recommendations. The Adaptive Deep Latent Factor Model (ADLFM) was developed to address this issue. This model adaptively learns the preferences for each item by using the item description, which provides a detailed summary and explanation of the item. ADLFM takes in item description as input, calculates latent vectors of the user and item, and presents a method that can reflect personal diversity using an attention score. However, due to the requirement of a dataset that includes item descriptions, the domain that can apply ADLFM is limited, resulting in generalization limitations. This study proposes a Generalized Adaptive Deep Latent Factor Recommendation Model, G-ADLFRM, to improve the limitations of ADLFM. Firstly, we use item ID, commonly used in recommendation systems, as input instead of the item description. Additionally, we apply improved deep learning model structures such as Self-Attention, Multi-head Attention, and Multi-Conv1D. We conducted experiments on various datasets with input and model structure changes. The results showed that when only the input was changed, MAE increased slightly compared to ADLFM due to accompanying information loss, resulting in decreased recommendation performance. However, the average learning speed per epoch significantly improved as the amount of information to be processed decreased. When both the input and the model structure were changed, the best-performing Multi-Conv1d structure showed similar performance to ADLFM, sufficiently counteracting the information loss caused by the input change. We conclude that G-ADLFRM is a new, lightweight, and generalizable model that maintains the performance of the existing ADLFM while enabling fast learning and inference.

• Asia Marketing Journal
• /
• v.13 no.4
• /
• pp.105-137
• /
• 2012

After relationship marketing emerged as a new approach in the marketing field in the 1980s, it has been widely studied in the United States, Europe and Asia. Rapid environmental changes and global competition has made it inevitable for companies to consider their relationships with the environment more closely. Under these circumstances, relationship marketing has held a position as a pivotal paradigm in the field of strategy as well as in marketing. In addition, relationship marketing has overcome the limitations of a traditional marketing research while providing richer implications in company's marketing activities. The paradigm shift to relationship marketing has brought fundamental changes in a marketing point of view. First, in philosophical aspects, unlike past research which focused solely on customer satisfaction, organizational relationship parameters which focuses on trust and commitment has become key elements of successful relationship marketing while shifts in thoughts naturally take place from adaptive marketing to strategic marketing. Second, in structural aspects, the relational mechanism of governance such as network structure with a variety of relational partners has emerged as a new marketing organization from the previous simple structure focusing on the micro-economic, marketbased trading between seller and customer. Third, in behavioral aspects, it proposed the strategic course of the action of gaining an advantage over the competition on the individual firm level by focusing on building long-term relationships and considering partnership with the components in the entire marketing system, rather than with one-time transaction-centric action between a seller and a customer. Fourth, in the aspects of marketing performance, marketing performance was sought through the long-term and cooperative relationship with various stakeholders, including customers in the marketing system, focusing on the overall competitive advantage based on relationship rather than individual performance of individual companies' marketing activities, such as market share and customer satisfaction. However, studies of relationship marketing were mostly centered in interorganizational relationships focusing on the relational structure and properties of commercial sector in the marketing system. Paradoxically, the circumstance of the consumer's side that must be considered is evolving again in relationship marketing. In structural aspects, a community, as the new relationship governance structure in the digital environment, and in behavioral aspects, the changing role of consumer participation demanding big changes in the digital environment engaged in the marketing system. The possibility of building a relationship marketing community for common value creation is presented in terms of organization of consumers with the focus on changing marketing environment and marketing system according to the new realities of the 21st century- the popularity of digital environments and the diffusion of customer participation. Therefore, future research of relationship marketing must seek for a truly integrated model including all of the existing structure and properties of the research oriented relationship from both the commercial and consumer sector.

• Journal of Korea Water Resources Association
• /
• v.42 no.12
• /
• pp.1029-1038
• /
• 2009

Recently, it is expected that small hydropower (SHP) could potentially provide sufficient amounts of alternative energy in Korea where there is an abundance of potential sites and where social efforts are being made to reduce the emissions of green house gases. In the past, the resources survey for SHP development has been carried out using onsite surveys and paper maps, which incurred a great deal of time and cost. Furthermore, the tools for decision making such as determining development priorities or evaluating feasibility have been considered only economic aspect and focused on the performance characteristics of power generation. However, as the concept of sustainable development has been being advanced in recent years, especially focused on human-social, environmental and ecological in addition to economical sector; the consideration of these multiple criteria has become essential for sustainable SHP development. This study aims to propose the spatial multi-criteria decision making (MCDM) methodology for determining priorities among a number of locations on the planning stage of SHP development using AHP and GIS. The proposed methodology is applied for determining development priorities among the SHP locations in Cho River basin and this study presents the detailed spatial information data and the results of development priorities. As a fundamental work, this study will be beneficial to the future activation of SHP development and will help the decision making in evaluating the feasibility of SHP development.

• Journal of Energy Engineering
• /
• v.25 no.1
• /
• pp.100-107
• /
• 2016

In this study confirmed the reduction effect of pollutant by applying Fi-EGR and FPI-EGR to hybrid combustion system realizing premixed flame and non-premxied flame at once. The results showed that NOx emission index decreased significantly in case of adopting EGR. Additionally, the hybrid combustion system with EGR resulted in a better performance compared to usual non-premixed combustion system such that it can reduce $NO_x$ emission at equivalent EGR ratios. Especially, in the case of 25% of FI-EGR ratio at hybrid combustion system that the ratio of non-premixed and premixed is 50 : 50, NOx emission index reduction rate was about 59% compared to $NO_x$ emission of non-premixed combustion system without EGR and in the case of 15% of FPI-EGR ratio at hybrid combustion system that the ratio of non-premixed and premixed is 70 : 30, $NO_x$ emission index reduction rate was about 48% compared to $NO_x$ emission of hybrid combustion system without EGR.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.28 no.1
• /
• pp.19-27
• /
• 2015

Recently, the most of domestic high-rise residential complex buildings are constructed with reinforced concrete structures, which may bring structural problems during construction. This study is aimed to analyze structural safety and lateral load-resisting performance of RC high-rise residential complex building under construction. The tower-typed building with 60 floors is selected as a sample model, and numerical analyses are performed. The structural performances of building structures at construction stages, which are resulted form the analyses of numerical models completed up to 10th, 20th, 30th, 40th, 50th, or 60th floor, are compared to those of the completed building structure. For the comparisons of structural performances, modal shapes and fundamental periods of building structures, lateral load-resisting performances, and structural design performances of structural members are considered. The lateral displacement and story drift ratio are analyzed for lateral load-resisting performances, and comparisons of design ratios at construction and design stages are performed for structural design performances of structural members. The guideline of design loads and structural analysis schemes for checking the safety of RC high-rise building under construction is presented.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.18 no.5
• /
• pp.389-397
• /
• 1998

For an effective and efficient management of large amounts of preservice/inservice inspection(PSI/ISI) data in nuclear power plants, an intellegent Windows 95-based data management program was developed. This program enables the prompt extraction of previously conducted PSI/ISI conditions and results so that the time-consuming data management, painstaking data processing and analysis in the past are avoided. The program extracts, and the associated remedies. Furthermore, additional inspection data and comments can be easily added or deleted for subsequent PSI/ISI operation. Although the initial version of the program was applied to Kori nuclear power plant, this program can be equally applied to other nuclear power plant. And also this program can be used to offer the fundamental data for application of evaluation data related to fracture mechanics analysis(FMA), probabilistic reliability assessment(PRA) of PSI/ISI results, performance demonstration initiative(PDI) and risk-informed ISI based on probability of detection(POD) information of ultrasonic examination. Besides, the program can be further developed as a unique PSI/ISI data management expert system that can be apart of PSI/ISI data management expert system that can be a part of PSI/ISI Total Support System(TSS) for Korean nuclear power plants.