• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.167-183
• /
• 2023

In the recent field of recommendation systems, various studies have been conducted to model sparse data effectively. Among these, GLocal-K(Global and Local Kernels for Recommender Systems) is a research endeavor combining global and local kernels to provide personalized recommendations by considering global data patterns and individual user characteristics. However, due to its utilization of kernel tricks, GLocal-K exhibits diminished performance on highly sparse data and struggles to offer recommendations for new users or items due to the absence of side information. In this paper, to address these limitations of GLocal-K, we propose the GEase-K (Global and EASE kernels for Recommender Systems) model, incorporating the EASE(Embarrassingly Shallow Autoencoders for Sparse Data) model and leveraging side information. Initially, we substitute EASE for the local kernel in GLocal-K to enhance recommendation performance on highly sparse data. EASE, functioning as a simple linear operational structure, is an autoencoder that performs highly on extremely sparse data through regularization and learning item similarity. Additionally, we utilize side information to alleviate the cold-start problem. We enhance the understanding of user-item similarities by employing a conditional autoencoder structure during the training process to incorporate side information. In conclusion, GEase-K demonstrates resilience in highly sparse data and cold-start situations by combining linear and nonlinear structures and utilizing side information. Experimental results show that GEase-K outperforms GLocal-K based on the RMSE and MAE metrics on the highly sparse GoodReads and ModCloth datasets. Furthermore, in cold-start experiments divided into four groups using the GoodReads and ModCloth datasets, GEase-K denotes superior performance compared to GLocal-K.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.185-209
• /
• 2023

With the advent of the 4th industrial revolution, various technologies have received much attention. Technologies related to the 4th industry include the Internet of Things (IoT), big data, artificial intelligence, virtual reality (VR), 3D printers, and robotics, and these technologies are often converged. In particular, the robotics field is combined with technologies such as big data, artificial intelligence, VR, and digital twins. Accordingly, much research using robotics is being conducted, which is applied to distribution, airports, hotels, restaurants, and transportation fields. In the given situation, research on human-robot interaction is attracting attention, but it has not yet reached the level of user satisfaction. However, research on robots capable of perfect communication is steadily being conducted, and it is expected that it will be able to replace human emotional labor. Therefore, it is necessary to discuss whether the current human-robot interaction technology can be applied to business. To this end, this study first examines the trend of human-robot interaction technology. Second, we compare LDA (Latent Dirichlet Allocation) topic modeling and BERTopic topic modeling methods. As a result, we found that the concept of human-robot interaction and basic interaction was discussed in the studies from 1992 to 2002. From 2003 to 2012, many studies on social expression were conducted, and studies related to judgment such as face detection and recognition were conducted. In the studies from 2013 to 2022, service topics such as elderly nursing, education, and autism treatment appeared, and research on social expression continued. However, it seems that it has not yet reached the level that can be applied to business. As a result of comparing LDA (Latent Dirichlet Allocation) topic modeling and the BERTopic topic modeling method, it was confirmed that BERTopic is a superior method to LDA.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.1-17
• /
• 2023

With the recent rapid development of ICT(Information and Communication Technology) and the popularization of digital devices, the size of the online market continues to grow. As a result, we live in a flood of information. Thus, customers are facing information overload problems that require a lot of time and money to select products. Therefore, a personalized recommender system has become an essential methodology to address such issues. Collaborative Filtering(CF) is the most widely used recommender system. Traditional recommender systems mainly utilize quantitative data such as rating values, resulting in poor recommendation accuracy. Quantitative data cannot fully reflect the user's preference. To solve such a problem, studies that reflect qualitative data, such as review contents, are being actively conducted these days. To quantify user review contents, text mining was used in this study. The general CF consists of the following three steps: user-item matrix generation, Top-N neighborhood group search, and Top-K recommendation list generation. In this study, we propose a recommendation algorithm that applies an extended similarity measure, which utilize quantified review contents in addition to user rating values. After calculating review similarity by applying TF-IDF, Word2Vec, and Doc2Vec techniques to review content, extended similarity is created by combining user rating similarity and quantified review contents. To verify this, we used user ratings and review data from the e-commerce site Amazon's "Health and Personal Care". The proposed recommendation model using extended similarity measure showed superior performance to the traditional recommendation model using only user rating value-based similarity measure. In addition, among the various text mining techniques, the similarity obtained using the TF-IDF technique showed the best performance when used in the neighbor group search and recommendation list generation step.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.50 no.2
• /
• pp.168-178
• /
• 2023

Objectives: This study aimed to compare the physical properties and antibacterial effectiveness of iodoform based root filling pastes, Vitapex^® and Metapex^®, with sodium iodide root filling paste (NaI paste) for primary teeth. Materials and Methods: The physical properties (flowability, film thickness, radiopacity) of the pastes were evaluated according to ISO 6876:2012. The antibacterial activity against Enterococcus faecalis strain (ATCC 6538) was evaluated using a direct contact test. Results: There was no significant statistical difference (p > 0.05) observed in the flow and film thickness of NaI paste when compared to the currently available root canal filling materials. The average flow capacities for Vitapex^®, Metapex^®, and NaI paste were 15.40 mm, 21.25 mm, and 20.01 mm, respectively. The average film thickness for Vitapex^®, Metapex^®, and NaI paste were 33.3 ㎕, 22.6 ㎕, and 25.0 ㎕, respectively. However, NaI paste showed lower radiopacity than the existing materials, and this difference was statistically significant (p < 0.05) NaI paste demonstrated higher antimicrobial activity than the available materials, and this difference was also statistically significant (p < 0.05). Conclusion: Compared to the existing commercialized root canal filling materials, NaI paste exhibited similar performance in terms of flow and film thickness, and superior antimicrobial activity against E. faecalis. Hence, NaI paste could be a promising root filling material for primary teeth and may be a potential alternative to existing materials.

• Economic Analysis
• /
• v.26 no.1
• /
• pp.84-132
• /
• 2020

A central bank infers market expectations of future yields based on yield curves. The central bank needs to precisely understand the changes in market expectations of future yields in order to have a more effective monetary policy. This need explains why a range of models have attempted to produce yield curves and market expectations that are as accurate as possible. Alongside the development of bond markets, the interconnectedness between them and macroeconomic factors has deepened, and this has rendered understanding of what macroeconomic variables affect yield curves even more important. However, the existence of various theories about determinants of yields inevitably means that previous studies have applied different macroeconomics variables when estimating yield curves. This indicates model uncertainties and naturally poses a question: Which model better estimates yield curves? Put differently, which variables should be applied to better estimate yield curves? This study employs the Dynamic Nelson-Siegel Model and takes the Bayesian approach to variable selection in order to ensure precision in estimating yield curves and market expectations of future yields. Bayesian variable selection may be an effective estimation method because it is expected to alleviate problems arising from a priori selection of the key variables comprising a model, and because it is a comprehensive approach that efficiently reflects model uncertainties in estimations. A comparison of Bayesian variable selection with the models of previous studies finds that the question of which macroeconomic variables are applied to a model has considerable impact on market expectations of future yields. This shows that model uncertainties exert great influence on the resultant estimates, and that it is reasonable to reflect model uncertainties in the estimation. Those implications are underscored by the superior forecasting performance of Bayesian variable selection models over those models used in previous studies. Therefore, the use of a Bayesian variable selection model is advisable in estimating yield curves and market expectations of yield curves with greater exactitude in consideration of the impact of model uncertainties on the estimation.

• Economic Analysis
• /
• v.27 no.4
• /
• pp.70-119
• /
• 2021

This paper analyzes the changes in financial reform during the Kim Jong-un era based on North Korean literature. We find that North Korea has systematically and functionally separated the central bank from commercial banks since the Kim Jong-un era began. In addition, enterprises have been allowed to withdraw cash from bank accounts and make inter-enterprise cash payments. In other words, nowadays non-cash currencies with passive money can partially serve as active money with purchasing power. With the systematic and functional separation of the central bank and the commercial bank, the issuance of the central bank changed to a money supply method through the commercial bank, and changes in the currency distribution structure have allowed commercial bank's credit creation function to be implemented. This means that the banking system and the monetary·payment system of the socialist planned economy are changing in the way of the market economy. Reforms in the financial sector are believed to have been necessary to support changes in the economic system and to restore the function of the public financial sector. These changes have progressed in terms of the level of reform, but they are still considered similar to the period of the former Soviet Union's Perestroika or to the early period of China's reform and opening. Although North Korea's financial reform is superior in terms of enacting the banking law, it is insufficient in terms of realizing the functions of commercial banks. In addition, it is assessed that institutional constraints such as maintaining a planned economy, and the lack of confidence in public finances limit the effectiveness and development of the financial system. It should be noted that these results are based on literature published in North Korea. In other words, there is a limit in the fact that such recent changes have been carried out on a trial basis in some areas, or have been carried out in a full-scale manner with a blueprint, since Kim Jong-un's inauguration.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.49 no.3
• /
• pp.247-258
• /
• 2023

This study aims to investigate the enhancement of water resistance and improvement in adhesion to the skin by combining dextrin palmitate and isopropyl titanium triisostearate coating materials with titanium dioxide. Due to the recent increase in consumers who enjoy outdoor activities, the demand for sunscreen with excellent water resistance is increasing. Prior research was conducted with O/W, Pickering, and W/O/W multiple formulations, but there was a limit to water resistance. The purpose of this study is to develop a complex inorganic powder that can improve water resistance and increase adhesion to the skin to solve this problem. First, we combined dextrin palmitate and isopropyl titanium triisostearate coating materials to form a composite with titanium dioxide. The coating of the inorganic powder was confirmed using FE-SEM and FT-IR analysis. The composite exhibited significantly higher in vitro water resistance compared to other formulations. The hydrophobicity of the coated inorganic powder was compared by measuring the contact angles. When the coated inorganic powder was applied to the W/O sunscreen formulation and the non-coated inorganic powder was applied to the W/O sunscreen formulation as a control, the SPF of the sunscreen containing the coated inorganic powder was higher. These results were the same when observed with a UV camera. Finally the adhesion of the coated inorganic powder to the skin was assessed by applying it to a foundation product. In vivo study, it was observed that the product formulated with the coated powder exhibited less smudging compared to the foundation product formulated with the non-coated powder. The developed inorganic powder in this study demonstrated excellent adhesion to the skin, providing a superior sensory experience, as well as enhanced hydrophobicity and remarkable water resistance effects. In the future, the result of this study is expected to help develop various sunscreen products to improve water resistance.

• Korean Chemical Engineering Research
• /
• v.61 no.2
• /
• pp.302-311
• /
• 2023

This study was conducted to evaluate the applicability of castor oil (CSO) as a natural wood preservative. CSO was treated into wood blocks prepared with domestic and imported wood species using a vacuum-pressure method, and then treatability, leachability and decay resistance of the CSO-treated wood blocks were examined. Although CSO was penetrated effectively into wood blocks of all wood species, the CSO-treatability was the highest in Western hemlock, followed by Japanese larch (LA), soft maple and Mongolian oak due to the difference of its anatomical structure. Except for LA, the more retained, the more leached during a saline water-immersing process for 48h. The use of ethanol added to reduce the viscosity of CSO affected negatively the treatability and leachability of wood blocks. Decay resistance, which was evaluated by the weight loss of wood blocks exposed against Fomitopsis palustris (FOP) and Trametes versicolor, of the CSO-treated/leached wood blocks was superior to that of control. Especially, most of wood blocks treated with preserving solution composed of only CSO (CSO-2) did not decayed and showed a very low weight loss against FOP. The decay resistance results from CSO retained in wood blocks after leaching. The retention of CSO could identify using the observation of X-ray microscope. Length of wood strips, which were treated with CSO-2 and then immersed in saline water for 2 weeks, hardly changed in all cutting directions. In addition, weight gain and length-swelling rate of the wood strips were extremely low compared to those of control. These results indicate that moisture resistance of the wood strips was improved by the CSO treatment. It is concluded that the treatment of CSO using a vacuum-pressure method provides the decay resistance and dimensional stability of wood, and thus CSO can be used as a natural wood preservative on various indoor and outdoor circumstances.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.329-352
• /
• 2022

It is reported that particulate matter(PM) penetrates the lungs and blood vessels and causes various heart diseases and respiratory diseases such as lung cancer. The subway is a means of transportation used by an average of 10 million people a day, and although it is important to create a clean and comfortable environment, the level of particulate matter pollution is shown to be high. It is because the subways run through an underground tunnel and the particulate matter trapped in the tunnel moves to the underground station due to the train wind. The Ministry of Environment and the Seoul Metropolitan Government are making various efforts to reduce PM concentration by establishing measures to improve air quality at underground stations. The smart air quality management system is a system that manages air quality in advance by collecting air quality data, analyzing and predicting the PM concentration. The prediction model of the PM concentration is an important component of this system. Various studies on time series data prediction are being conducted, but in relation to the PM prediction in subway stations, it is limited to statistical or recurrent neural network-based deep learning model researches. Therefore, in this study, we propose four transformer-based models including spatiotemporal transformers. As a result of performing PM concentration prediction experiments in the waiting rooms of subway stations in Seoul, it was confirmed that the performance of the transformer-based models was superior to that of the existing ARIMA, LSTM, and Seq2Seq models. Among the transformer-based models, the performance of the spatiotemporal transformers was the best. The smart air quality management system operated through data-based prediction becomes more effective and energy efficient as the accuracy of PM prediction improves. The results of this study are expected to contribute to the efficient operation of the smart air quality management system.

• Journal of the Korean Orthopaedic Association
• /
• v.54 no.4
• /
• pp.343-352
• /
• 2019

Purpose: The purpose of this study was to document the sonographic morphology of the subscapularis footprint, particularly the 1st facet, of the non-pathologic subscapularis tendon and footprint, and analyze the correlation between the size of the 1st facet and the demographic variables. Materials and Methods: Between March 2015 and December 2017, retrospectively data analysis was performed for the ultrasound (US) scans of 115 consecutive shoulder (mean age 53.4 years, range 23-74 years) with non-pathologic subscapularis tendon and footprint. The sonographic findings of the 1st facet of the subscapularis footprint was a very unique, flat, broad, and plane angle in the upward direction, which were distinguished from the other facets. On US, the transverse (medio-lateral) and longitudinal (superior-inferior) length of the 1st facet on axis of the humerus shaft were recorded. The demographic variables, including age, site, body height, weight, body mass index (BMI), and arm length, were reviewed. Results: On US, the mean transverse length of the 1st facet was 12.75 mm (range 10.54-14.50 mm, standard deviation [SD] 0.712) and the mean longitudinal length was 12.22 mm (range 9.20-13.30 mm, SD 0.888). The transverse and longitudinal length of the size of the 1st facet were significantly greater in males than in females (p<0.001, p=0.001). Of the demographic data (body height, weight, BMI, arm length) that showed a significant positive linear correlation, the correlation with body height (transverse r=0.749, p<0.001; longitudinal r=0.642, p<0.001) showed the strongest relationship, and the correlation with the BMI was weakly related. The relationships between the size of the 1st facet to site/age were not statistically significant or appeared to have no linear correlation. Conclusion: The structural and morphologic features of the 1st facet of the subscapularis footprint on the US were identified. This will provide anatomic knowledge of an US examination for subscapularis tendon pathology.