• Smart Media Journal
• /
• v.12 no.4
• /
• pp.27-40
• /
• 2023

Many researchers have investigated ways to enhance recommender systems by integrating heterogeneous data to address the data sparsity problem. However, only a few studies have successfully integrated heterogeneous data using knowledge graph. Additionally, most of the knowledge graphs built in these studies only incorporate explicit relationships between entities and lack additional information. Therefore, we propose a method for expanding knowledge graphs by using deep learning to model latent relationships between heterogeneous data from multiple knowledge bases. Our extended knowledge graph enhances the quality of entity features and ultimately increases the accuracy of predicted user preferences. Experiments using real music data demonstrate that the expanded knowledge graph leads to an increase in recommendation accuracy when compared to the original knowledge graph.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.65 no.1
• /
• pp.27-39
• /
• 2023

This study was to quantitatively evaluate periodic and maximum energy loads for broiler-standard design models by the Ministry of Agriculture, Food and Rural Affairs (2016). Building energy simulation method was used to compute heating and cooling loads of the designed broiler houses according to regional locations and insulation characteristics of wall and roof. It considered sensible and latent heat generation from broilers, dynamic operation of ventilation system according to environment variations. It was found that variation of periodic heating loads was relatively higher than that of periodic cooling loads according to thickness changes of wall and roof. Assuming that broiler was raised at every even-month, periodic heating and cooling loads were 6 and 18% lower, respectively than odd-month raising condition. When recommendation rules of insulation characteristics (wall and roof thickness) by the Ministry of Land, Infrastructure and Transport was adopted, periodic heating load of Jeju-si was 20.3% higher than national average values. Based on the BES computed periodic and maximum energy loads under the designed experimental condition, these results can contribute to reestablishing standard design of broiler houses, especially for insulation characteristics, and designing management strategies for efficient energy uses.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.2B
• /
• pp.153-160
• /
• 2006

To recommend ecological flow for major tributaries in Han River basin, the Instream Flow Incremental Methodology (IFIM) have been applied. In particular physical habitat simulation using PHABSIM have been selected for microhabitat variables and QUAL2E model have been used to implement macrohabitat simulation. Habitat Suitability Criteria (HSC) for different life stages in accordance with different hydraulic variables (depth and velocity) have been presented by the field surveying data. We review IFIM procedures and discuss limitations of habitat simulation with specific reference to Han River basin. The results of this research can be used as reference flow for estimation of instream flow in Han River.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.4
• /
• pp.154-160
• /
• 2023

Recently, with the activation of eBook services, books are being published simultaneously as physical books and digitized eBooks. Paper books are more expensive than e-books due to printing and distribution costs, so demand for relatively inexpensive e-books is increasing. There are cases where previously published physical books cannot be digitized due to the circumstances of the publisher or author, so there is a movement among individual users to digitize books that have been published for a long time. However, existing research has only studied the advancement of the pre-processing process that can improve text recognition before applying OCR technology, and there are limitations to digitization depending on the condition of the book. Therefore, support for book digitization services depending on the condition of the physical book is needed. need. In this paper, we propose a method to support digitalization services according to the status of physical books held by book owners. Create images by scanning books and extract text information from the images through OCR. We propose a method to recover text that cannot be extracted depending on the state of the book using BERT, a natural language processing deep learning model. As a result, it was confirmed that the recovery method using BERT is superior when compared to RNN, which is widely used in recommendation technology.

• Asia pacific journal of information systems
• /
• v.29 no.4
• /
• pp.856-874
• /
• 2019

The performance of recommender systems (RS) has been measured mainly in terms of accuracy. However, there are other aspects of performance that are difficult to understand in terms of accuracy, such as coverage, serendipity, and satisfaction with recommended results. Moreover, particularly with RSs that suggest multiple items at a time, such as YouTube, user satisfaction with recommended results may vary not only depending on their accuracy, but also on their configuration, content, and design displayed to the user. This is true when classifying an RS as a single RS with one recommended result and as a multiple RS with diverse results. No empirical analysis has been conducted on the influence of the content and distribution of recommendation items on user satisfaction. In this study, we propose a research model representing the content and distribution of recommended items and how they affect user satisfaction with the RS. We focus on RSs that recommend multiple items. We performed an empirical analysis involving 149 YouTube users. The results suggest that user satisfaction with recommended results is significantly affected according to the HHI (Herfindahl-Hirschman Index). In addition, satisfaction significantly increased when the recommended item on the top of the list was the same category in terms of content that users were currently watching. Particularly when the purpose of using RS is hedonic, not utilitarian, the results showed greater satisfaction when the number of views of the recommended items was evenly distributed. However, other characteristics of selected content, such as view count and playback time, had relatively less impact on satisfaction with recommended items. To the best of our knowledge, this study is the first to show that the category concentration of items impacts user satisfaction on websites recommending diverse items in different categories using a content-based filtering system, such as YouTube. In addition, our use of the HHI index, which has been extensively used in economics research, to show the distributional characteristics of recommended items, is also unique. The HHI for categories of recommended items was useful in explaining user satisfaction.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.14 no.3
• /
• pp.153-168
• /
• 2019

Augmented Reality(AR) has been one of the important technologies of the 4th industrial revolution. Consumer acceptance of new technologies is substantial issue for market expansion, but there have been few empirical studies on factors that affect the acceptance or use intention of AR. In this study, we have explored and analyzed the factors influencing technology acceptance based on the extended unified theory of acceptance and use of technology(UTAUT2) model in the AR business and have discussed it with comparison with existing research based on this analysis. The results of this study suggest that the main variables of the existing UTAUT1 model had significant positive effect on the intention to use, such as performance expectancy, effort expectancy, facilitating conditions and hedonic motivation, habits of UTAUT2. In addition, perceived risk introduced in this study had a negative effect on intention to use. Furthermore, the impact between these two factors have been effort expectancy(${\beta}=.294$)>habits(${\beta}=.268$)>hedonic motivation(${\beta}=.266$)>performance expectancy,(${\beta}=.263$)>facilitating conditions(${\beta}=.233$)>perceived risk(${\beta}=-.094$). The impact of social influence did not have a significant effect on intention to use. The intention to use was analyzed to have a significant positive effect on the actual use and recommendation intention. On the other hand, the hypothesis that the age and gender has played a moderating role between independent variables and the intention of use were investigated. Age was found out to play a role as a moderator between social influence, facilitating conditions, hedonic motivation, habits and intention to use. In the same way, gender has been shown to play a moderating role between facilitating conditions, perceived risk and intention to use. Academic and practical implications are suggested based on the results of this study.

• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.105-129
• /
• 2020

This study uses corporate data from 2012 to 2018 when K-IFRS was applied in earnest to predict default risks. The data used in the analysis totaled 10,545 rows, consisting of 160 columns including 38 in the statement of financial position, 26 in the statement of comprehensive income, 11 in the statement of cash flows, and 76 in the index of financial ratios. Unlike most previous prior studies used the default event as the basis for learning about default risk, this study calculated default risk using the market capitalization and stock price volatility of each company based on the Merton model. Through this, it was able to solve the problem of data imbalance due to the scarcity of default events, which had been pointed out as the limitation of the existing methodology, and the problem of reflecting the difference in default risk that exists within ordinary companies. Because learning was conducted only by using corporate information available to unlisted companies, default risks of unlisted companies without stock price information can be appropriately derived. Through this, it can provide stable default risk assessment services to unlisted companies that are difficult to determine proper default risk with traditional credit rating models such as small and medium-sized companies and startups. Although there has been an active study of predicting corporate default risks using machine learning recently, model bias issues exist because most studies are making predictions based on a single model. Stable and reliable valuation methodology is required for the calculation of default risk, given that the entity's default risk information is very widely utilized in the market and the sensitivity to the difference in default risk is high. Also, Strict standards are also required for methods of calculation. The credit rating method stipulated by the Financial Services Commission in the Financial Investment Regulations calls for the preparation of evaluation methods, including verification of the adequacy of evaluation methods, in consideration of past statistical data and experiences on credit ratings and changes in future market conditions. This study allowed the reduction of individual models' bias by utilizing stacking ensemble techniques that synthesize various machine learning models. This allows us to capture complex nonlinear relationships between default risk and various corporate information and maximize the advantages of machine learning-based default risk prediction models that take less time to calculate. To calculate forecasts by sub model to be used as input data for the Stacking Ensemble model, training data were divided into seven pieces, and sub-models were trained in a divided set to produce forecasts. To compare the predictive power of the Stacking Ensemble model, Random Forest, MLP, and CNN models were trained with full training data, then the predictive power of each model was verified on the test set. The analysis showed that the Stacking Ensemble model exceeded the predictive power of the Random Forest model, which had the best performance on a single model. Next, to check for statistically significant differences between the Stacking Ensemble model and the forecasts for each individual model, the Pair between the Stacking Ensemble model and each individual model was constructed. Because the results of the Shapiro-wilk normality test also showed that all Pair did not follow normality, Using the nonparametric method wilcoxon rank sum test, we checked whether the two model forecasts that make up the Pair showed statistically significant differences. The analysis showed that the forecasts of the Staging Ensemble model showed statistically significant differences from those of the MLP model and CNN model. In addition, this study can provide a methodology that allows existing credit rating agencies to apply machine learning-based bankruptcy risk prediction methodologies, given that traditional credit rating models can also be reflected as sub-models to calculate the final default probability. Also, the Stacking Ensemble techniques proposed in this study can help design to meet the requirements of the Financial Investment Business Regulations through the combination of various sub-models. We hope that this research will be used as a resource to increase practical use by overcoming and improving the limitations of existing machine learning-based models.

• Journal of Legislation Research
• /
• no.44
• /
• pp.343-386
• /
• 2013

"Financial ADR" system in South Korea can be represented by so-called "Financial Dispute Resolution System", in which Financial Supervisory Service (FSS) and Financial Dispute Resolution Committee are the principal actors in operation of the system, and this is discussed as an "Administrative Financial ADR System". The system has over 10-year history since it was introduced in around 1999. Nonetheless, it was not until when financial consumer protection began to be highlighted after the 2008 financial crisis that Financial ADR system actually started to draw attention in Korea. This was because interest has been rising in "Alternative Dispute Resolution (ADR)" as an institutional measure to protect financial consumers damaged via financial transactions. However, the current discussion on the domestic Financial ADR system shows an aspect that it is confined to who is to be a principal actor for the operation of Financial ADR institution with main regards to reorganization of supervisory system. This article aims to embody these facts in an institutional model by recognizing them as a problem and analyzing the features of the Financial ADR system, thereby clarifying problems of the system and presenting the direction of improvement. The Korean Financial ADR system can be judged as "administrative model integrated model consensual model quasi-judicial model non-prepositive Internal Dispute Resolution (IDR) model". However, at the same time, it is confronted with a task to overcome the two problems; the system is not equipped with institutional basis for securing its validity in spite of the adopted quasi-judicial effect model; and a burden of operating an integrated ADR system is considerable. From this perspective, the article suggests improvement plans for security of validity in the current system and for expansion of industry-control ADR system, in particular, a system of prepositive IDR model. Amongst them, it suggests further plans for securing the validity of the system as follows; promotion to expand the number of internal persons and to differentiate mediation procedures and effect; a plan to keep a financial institution from filing a lawsuit before an agreement recommendation or a mediation proposal is advised; and a plan to grant suspension of extinctive prescription as well as that of procedures of the lawsuit.

• Journal of Intelligence and Information Systems
• /
• v.26 no.3
• /
• pp.149-169
• /
• 2020

"The Urban Renewal New Deal project", one of the government's major national projects, is about developing underdeveloped areas by investing 50 trillion won in 100 locations on the first year and 500 over the next four years. This project is drawing keen attention from the media and local governments. However, the project model which fails to reflect the original characteristics of the area as it divides project area into five categories: "Our Neighborhood Restoration, Housing Maintenance Support Type, General Neighborhood Type, Central Urban Type, and Economic Base Type," According to keywords for successful urban regeneration in Korea, "resident participation," "regional specialization," "ministerial cooperation" and "public-private cooperation", when local governments propose urban regeneration projects to the government, they can see that it is most important to accurately understand the characteristics of the city and push ahead with the projects in a way that suits the characteristics of the city with the help of local residents and private companies. In addition, considering the gentrification problem, which is one of the side effects of urban regeneration projects, it is important to select and implement urban regeneration types suitable for the characteristics of the area. In order to supplement the limitations of the 'Urban Regeneration New Deal Project' methodology, this study aims to propose a system that recommends urban regeneration types suitable for urban regeneration sites by utilizing various machine learning algorithms, referring to the urban regeneration types of the '2025 Seoul Metropolitan Government Urban Regeneration Strategy Plan' promoted based on regional characteristics. There are four types of urban regeneration in Seoul: "Low-use Low-Level Development, Abandonment, Deteriorated Housing, and Specialization of Historical and Cultural Resources" (Shon and Park, 2017). In order to identify regional characteristics, approximately 100,000 text data were collected for 22 regions where the project was carried out for a total of four types of urban regeneration. Using the collected data, we drew key keywords for each region according to the type of urban regeneration and conducted topic modeling to explore whether there were differences between types. As a result, it was confirmed that a number of topics related to real estate and economy appeared in old residential areas, and in the case of declining and underdeveloped areas, topics reflecting the characteristics of areas where industrial activities were active in the past appeared. In the case of the historical and cultural resource area, since it is an area that contains traces of the past, many keywords related to the government appeared. Therefore, it was possible to confirm political topics and cultural topics resulting from various events. Finally, in the case of low-use and under-developed areas, many topics on real estate and accessibility are emerging, so accessibility is good. It mainly had the characteristics of a region where development is planned or is likely to be developed. Furthermore, a model was implemented that proposes urban regeneration types tailored to regional characteristics for regions other than Seoul. Machine learning technology was used to implement the model, and training data and test data were randomly extracted at an 8:2 ratio and used. In order to compare the performance between various models, the input variables are set in two ways: Count Vector and TF-IDF Vector, and as Classifier, there are 5 types of SVM (Support Vector Machine), Decision Tree, Random Forest, Logistic Regression, and Gradient Boosting. By applying it, performance comparison for a total of 10 models was conducted. The model with the highest performance was the Gradient Boosting method using TF-IDF Vector input data, and the accuracy was 97%. Therefore, the recommendation system proposed in this study is expected to recommend urban regeneration types based on the regional characteristics of new business sites in the process of carrying out urban regeneration projects."

• The Journal of the Korea Contents Association
• /
• v.17 no.10
• /
• pp.548-558
• /
• 2017

The purpose of this study is to investigate how the types of perceived brand image related to the main concept building brand equity affect 'brand preference' and 'behavioral intentions'. The perceived brand image is set as the brand image type perceived by the consumer from the image pursued by the corporate brand, while in addition to brand preference, behavioral intentions are set as purchase intention and recommendation intention for the result variables. The result shows that the types of perceived brand image were extracted as 'factor 1(challenge spirit)' and factor 2(reliability) and through the cluster analysis 3 groups under each type were identified. Also, a significant difference between the influence of each type of perceived brand image on 'brand preference', 'purchase intention' and 'recommend intention' was indicated. In addition, the differences of perceived brand image types were found to be higher in order of 'challenge spirit type', 'reliability type', 'integrated type'. The empirical implementation of this study lies in the fact that it classifies the concept of brand image not as a broad theoretical model, but as a model directly related to real consumer perception, and that it gives practical suggestion for brand image management related to advertising.