• Food Engineering Progress
• /
• v.21 no.1
• /
• pp.42-48
• /
• 2017

The purpose of this study was to verify the drying characteristics of steamed sweet potato and to establish optimal drying conditions for far-infrared drying of steamed sweet potato. 4 kg of steamed sweet potato was sliced to thicknesses of 8 and 10 mm, and dried by a far-infrared dryer until a final moisture content of $25{\pm}0.5%$. The far-infrared dryer conditions were an air velocity of 0.6, 0.8 m/s and drying temperature of 60, 70, and $80^{\circ}C$. The results can be summarized as follows. The drying time tended to be reduced as temperature and air velocity for drying increased. The Lewis and Modified Wang and Singh models were found to be suitable for drying of steamed sweet potato by a far-infrared dryer. The color difference was 35.09 on the following conditions: Thickness of 8 mm, temperature of $80^{\circ}C$, and air velocity of 0.8 m/s. The highest sugar content ($59.11^{\circ}Brix$) was observed on the conditions of a thickness of 8 mm, temperature of 80, and air velocity of 0.8 m/s. Energy consumption decreased on the conditions of higher temperature, slower air velocity, and thinner steamed sweet potato.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.23 no.3
• /
• pp.149-155
• /
• 2021

Early warning systems for weather risk management in the agricultural sector have been developed to predict potential wind damage to crops. These systems take into account the daily maximum wind speed to determine the critical wind speed that causes fruit drops and provide the weather risk information to farmers. In an effort to increase the accuracy of wind risk predictions, an artificial neural network for binary classification was implemented. In the present study, the daily wind speed and other weather data, which were measured at weather stations at sites of interest in Jeollabuk-do and Jeollanam-do as well as Gyeongsangbuk- do and part of Gyeongsangnam- do provinces in 2019, were used for training the neural network. These weather stations include 210 synoptic and automated weather stations operated by the Korean Meteorological Administration (KMA). The wind speed data collected at the same locations between January 1 and December 12, 2020 were used to validate the neural network model. The data collected from December 13, 2020 to February 18, 2021 were used to evaluate the wind risk prediction performance before and after the use of the artificial neural network. The critical wind speed of damage risk was determined to be 11 m/s, which is the wind speed reported to cause fruit drops and damages. Furthermore, the maximum wind speeds were expressed using Weibull distribution probability density function for warning of wind damage. It was found that the accuracy of wind damage risk prediction was improved from 65.36% to 93.62% after re-classification using the artificial neural network. Nevertheless, the error rate also increased from 13.46% to 37.64%, as well. It is likely that the machine learning approach used in the present study would benefit case studies where no prediction by risk warning systems becomes a relatively serious issue.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.175-197
• /
• 2021

Recently, with the development of deep learning technology, research on unstructured data analysis is being actively conducted, and it is showing remarkable results in various fields such as classification, summary, and generation. Among various text analysis fields, text classification is the most widely used technology in academia and industry. Text classification includes binary class classification with one label among two classes, multi-class classification with one label among several classes, and multi-label classification with multiple labels among several classes. In particular, multi-label classification requires a different training method from binary class classification and multi-class classification because of the characteristic of having multiple labels. In addition, since the number of labels to be predicted increases as the number of labels and classes increases, there is a limitation in that performance improvement is difficult due to an increase in prediction difficulty. To overcome these limitations, (i) compressing the initially given high-dimensional label space into a low-dimensional latent label space, (ii) after performing training to predict the compressed label, (iii) restoring the predicted label to the high-dimensional original label space, research on label embedding is being actively conducted. Typical label embedding techniques include Principal Label Space Transformation (PLST), Multi-Label Classification via Boolean Matrix Decomposition (MLC-BMaD), and Bayesian Multi-Label Compressed Sensing (BML-CS). However, since these techniques consider only the linear relationship between labels or compress the labels by random transformation, it is difficult to understand the non-linear relationship between labels, so there is a limitation in that it is not possible to create a latent label space sufficiently containing the information of the original label. Recently, there have been increasing attempts to improve performance by applying deep learning technology to label embedding. Label embedding using an autoencoder, a deep learning model that is effective for data compression and restoration, is representative. However, the traditional autoencoder-based label embedding has a limitation in that a large amount of information loss occurs when compressing a high-dimensional label space having a myriad of classes into a low-dimensional latent label space. This can be found in the gradient loss problem that occurs in the backpropagation process of learning. To solve this problem, skip connection was devised, and by adding the input of the layer to the output to prevent gradient loss during backpropagation, efficient learning is possible even when the layer is deep. Skip connection is mainly used for image feature extraction in convolutional neural networks, but studies using skip connection in autoencoder or label embedding process are still lacking. Therefore, in this study, we propose an autoencoder-based label embedding methodology in which skip connections are added to each of the encoder and decoder to form a low-dimensional latent label space that reflects the information of the high-dimensional label space well. In addition, the proposed methodology was applied to actual paper keywords to derive the high-dimensional keyword label space and the low-dimensional latent label space. Using this, we conducted an experiment to predict the compressed keyword vector existing in the latent label space from the paper abstract and to evaluate the multi-label classification by restoring the predicted keyword vector back to the original label space. As a result, the accuracy, precision, recall, and F1 score used as performance indicators showed far superior performance in multi-label classification based on the proposed methodology compared to traditional multi-label classification methods. This can be seen that the low-dimensional latent label space derived through the proposed methodology well reflected the information of the high-dimensional label space, which ultimately led to the improvement of the performance of the multi-label classification itself. In addition, the utility of the proposed methodology was identified by comparing the performance of the proposed methodology according to the domain characteristics and the number of dimensions of the latent label space.

• Journal of Environmental Impact Assessment
• /
• v.30 no.2
• /
• pp.117-131
• /
• 2021

Urban development is continuously being carried out to stabilize housing supply. The importance of ecosystem services assessments considering the various urban spatial structures is emerging as land use change in the wake of urban development has impact on the provision of ecosystem services. However, few studies are available as to the effects of land use transition in agricultural land due to urban development on ecosystem services. The purpose of this study is to examine the applicability of decision-making in the urban planning process by analyzing the impact of land use change on ecosystem services due to urban development. Therefore, the scenario was set on before and after the city development, targeting Namyangju Wangsuk Zone 1 and InVEST model was used to compare and analyze changes in value of ecosystem services. Analysis results of ecosystem services before the urban development, it turned out that habitat quality and water yield increased overtime but carbon storage and crop production decreased. Analysis results of ecosystem services after the urban development indicated that all items in ecosystem services by scenario decreased more than in 2018. Among the scenarios of urban development, compact city had the lowest value of water resource supply but had the highest value in terms of habitat quality, carbon storage amount, and crop production. The study results demonstrate that the compact city has positive effects on ecosystem services and is expected to be used as the basic data for supporting policy decision-making in the establishment of future urban development and management plans.

• Journal of Technology Innovation
• /
• v.28 no.3
• /
• pp.1-28
• /
• 2020

This study examines whether the firm with high innovation efficiency realizes high operating performance. We measured innovation efficiency by the ratio of patent applications for R&D expenditure or R&D stock and measured operating performance by the ratio of operating income or operating cash flow to total assets for the following year. The sample consists of 1,880 manufacturing firm-years, which listed on the Korean Exchange between 2014 and 2017. We analyze the effect of innovation efficiency on operating performance using a model of Hirshleifer et al. (2013) results show that both innovation efficiency variables have a significantly positive relationship with the total asset operating margin. Besides, the following year's performance, measured by the total asset operating cash flow ratio, also shows a positive relationship with the two innovation efficiency variables at the 5% and 1% significance levels, respectively. The results indicate that high innovation efficiency firms that link the outcomes of R&D to more patent applications realize higher operating performance. Also, we divided the R&D-intensive and non-R&D-intensive industries and performed the same analysis. As a result, the innovation efficiency has a significant positive effect on operating margin in both industries. However, the effect of innovation efficiency on the operating cash flow is only significant in R&D-intensive industries. This study suggests that the effects of innovation efficiency are more consistent in the R&D-intensive industry. Additionally, we divided the high patent application and low patent applications industries and performed the same analysis. As a result, the innovation efficiency has a significant positive effect on operating margin in both industries. This study suggests that the effects of innovation efficiency are more consistent in the high patent application industry. We show that a firm's innovation efficiency is a critical factor for a firm's performance, while prior studies on the R&D performance have not considered the innovation efficiency of each firm. The evidence suggests that firms not only consider R&D expenditures but also improve the performance of companies by increasing innovation efficiency. Investors need to consider their innovation efficiency when evaluating the value of firms.

• (The)Study of the Eastern Classic
• /
• no.49
• /
• pp.357-392
• /
• 2012

An author is bound to reflect his or her own thinking and inclination in his or her works. The previous studies on Changam(蒼巖), however, mostly discussed the aesthetics in the forms of his introductions and works, hardly addressing his thinking reflected in his works. Recognizing that he had the "reactionism tendency" unlike the Bukhak-School(北學派), which was the cultural mainstream of the days, this study examined the specific patterns of the reactionism calligraphy style in his learning and calligraphy processes and works. He loved to write xing-cao-shu(行草書) with a focus on the materials written in one's own calligraphy, but he also emphasized that one should obtain the force of his or her calligraphy style by mastering kai shu before calligraphy xing cao shu. He thus left a lot of works in the xiao kai(小楷) of the Wang Xzhi(王羲之) calligraphy style throughout his life, which is attributed to the influences of the calligraphers of dong-guk-jin-che(東國眞體) in the latter half of Joseon(朝鮮) and those of Lee Gwang-sa(李匡師), his master in spirit. He is distinguished from the other calligraphers of the times in that he made lifelong efforts to compensate for the lacking stroke of the pen in the model calligraphy of Wang Xzhi. In the calligraphy theory, he put importance on the traditional method of Han-Wei(漢魏) and took Cai Yong(蔡邕) and Zhong Yao(鍾繇) as the fundamentals. For da kai(大楷), he constantly practiced the with the stroke of the pen by added to it, the letters of Wei(魏) Wudi(武帝), by Yan Zhenqing(顔眞卿), and letters of Kim Saeng(金生). His late works using the intended conception of and , in particular, present his unique calligraphy style that added the crooked forms of to the shapes of characters of that were in the kai-shu(楷書) style. It is a limitation that a considerable number of calligraphy materials Changam studied or consulted were either reprint copy or block book rather than original rubbing edition due to time and space restrictions. However, it is also true that those restrictions made an important contribution to his creation of his unique calligraphy style with deep local colors at the result of his constant efforts.

• The Journal of Korean Academy of Prosthodontics
• /
• v.59 no.1
• /
• pp.18-26
• /
• 2021

Purpose: Generally, patients are noticed to store denture in water when removed from the mouth. However, few studies have reported the advantage of volumetric change in underwater storage over dry storage. To be a reference in defining the proper denture storage method, this study aims to evaluate the volumetric change and dimensional deformation in case of underwater and dry storage. Materials and methods: Definitive casts were scanned by a model scanner, and denture bases were designed with computer-aided design (CAD) software. Twelve denture bases (upper 6, lower 6) were printed with 3D printer. Printed denture bases were invested and flasked with heat-curing method. 6 upper and 6 lower dentures were divided into group A and B, and each group contains 3 upper and 3 lower dentures. Group A was stored dry at room temperature, group B was stored underwater. Group B was scanned at every 24 hours for 28 days and scanned data was saved as stereolithography (SLA) file. These SLA files were analyzed to measure the difference in volumetric change of a month and Kruskal-Wallis test were used for statistical analysis. Best-fit algorithm was used to overlap and 3-dimensional color-coded map was used to observe the changing pattern of impression surface. Results: No significant difference was found in volumetric changes regardless of the storage methods. In dry-stored denture base, significant changes were found in the palate of upper jaw and posterior lingual border of lower jaw in direction away from the underlying tissue, maxillary tuberosity of upper jaw and retromolar pad area of lower jaw in direction towards the underlying tissue. Conclusion: Storing the denture underwater shows less volumetric change of impression surface than storing in the dry air.

• Journal of Intelligence and Information Systems
• /
• v.27 no.2
• /
• pp.1-15
• /
• 2021

If it is difficult for us to make decisions, we ask for advice from friends or people around us. When we decide to buy products online, we read anonymous reviews and buy them. With the advent of the Data-driven era, IT technology's development is spilling out many data from individuals to objects. Companies or individuals have accumulated, processed, and analyzed such a large amount of data that they can now make decisions or execute directly using data that used to depend on experts. Nowadays, the recommender system plays a vital role in determining the user's preferences to purchase goods and uses a recommender system to induce clicks on web services (Facebook, Amazon, Netflix, Youtube). For example, Youtube's recommender system, which is used by 1 billion people worldwide every month, includes videos that users like, "like" and videos they watched. Recommended system research is deeply linked to practical business. Therefore, many researchers are interested in building better solutions. Recommender systems use the information obtained from their users to generate recommendations because the development of the provided recommender systems requires information on items that are likely to be preferred by the user. We began to trust patterns and rules derived from data rather than empirical intuition through the recommender systems. The capacity and development of data have led machine learning to develop deep learning. However, such recommender systems are not all solutions. Proceeding with the recommender systems, there should be no scarcity in all data and a sufficient amount. Also, it requires detailed information about the individual. The recommender systems work correctly when these conditions operate. The recommender systems become a complex problem for both consumers and sellers when the interaction log is insufficient. Because the seller's perspective needs to make recommendations at a personal level to the consumer and receive appropriate recommendations with reliable data from the consumer's perspective. In this paper, to improve the accuracy problem for "appropriate recommendation" to consumers, the recommender systems are proposed in combination with context-based deep learning. This research is to combine user-based data to create hybrid Recommender Systems. The hybrid approach developed is not a collaborative type of Recommender Systems, but a collaborative extension that integrates user data with deep learning. Customer review data were used for the data set. Consumers buy products in online shopping malls and then evaluate product reviews. Rating reviews are based on reviews from buyers who have already purchased, giving users confidence before purchasing the product. However, the recommendation system mainly uses scores or ratings rather than reviews to suggest items purchased by many users. In fact, consumer reviews include product opinions and user sentiment that will be spent on evaluation. By incorporating these parts into the study, this paper aims to improve the recommendation system. This study is an algorithm used when individuals have difficulty in selecting an item. Consumer reviews and record patterns made it possible to rely on recommendations appropriately. The algorithm implements a recommendation system through collaborative filtering. This study's predictive accuracy is measured by Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). Netflix is strategically using the referral system in its programs through competitions that reduce RMSE every year, making fair use of predictive accuracy. Research on hybrid recommender systems combining the NLP approach for personalization recommender systems, deep learning base, etc. has been increasing. Among NLP studies, sentiment analysis began to take shape in the mid-2000s as user review data increased. Sentiment analysis is a text classification task based on machine learning. The machine learning-based sentiment analysis has a disadvantage in that it is difficult to identify the review's information expression because it is challenging to consider the text's characteristics. In this study, we propose a deep learning recommender system that utilizes BERT's sentiment analysis by minimizing the disadvantages of machine learning. This study offers a deep learning recommender system that uses BERT's sentiment analysis by reducing the disadvantages of machine learning. The comparison model was performed through a recommender system based on Naive-CF(collaborative filtering), SVD(singular value decomposition)-CF, MF(matrix factorization)-CF, BPR-MF(Bayesian personalized ranking matrix factorization)-CF, LSTM, CNN-LSTM, GRU(Gated Recurrent Units). As a result of the experiment, the recommender system based on BERT was the best.

• Maritime Security
• /
• v.1 no.1
• /
• pp.273-310
• /
• 2020

This study aims at empirically analyzing the overall mechanism of the "Gray Zone Strategy", which has begun to be used as one of Chinese major maritime security strategies in maritime conflicts surrounding the South China Sea and East China Sea since early 2010, and comparing the resulting conflict patterns in those reg ions. To this end, I made the following two hypotheses about Chinese gray zone strategy. The hypotheses that I have argued in this study are the first, "The marine gray zone strategy used by China shows different structures of implementation in the South China Sea and the East China Sea, which are major conflict areas.", the second, "Therefore, the patterns of disputes in the South China Sea and the East China Sea also show a difference." In order to examine this, I will classify Chinese gray zone strategy mechanisms multi-dimensionally in large order, 1) conflict trends and frequency of strategy execution, 2) types and strengths of strategy, 3) actors of strategy execution, and 4) response methods of counterparts. So, I tried to collect data related to this based on quantitative modeling to test these. After that, about 10 years of data pertaining to this topic were processed, and a research model was designed with a new categorization and operational definition of gray zone strategies. Based on this, I was able to successfully test all the hypotheses by successfully comparing the comprehensive mechanisms of the gray zone strategy used by China and the conflict patterns between the South China Sea and the East China Sea. In the conclusion, the verified results were rementioned with emphasizing the need to overcome the security vulnerabilities in East Asia that could be caused by China's marine gray zone strategy. This study, which has never been attempted so far, is of great significance in that it clarified the intrinsic structure in which China's gray zone strategy was implemented using empirical case studies, and the correlation between this and maritime conflict patterns was investigated.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.16 no.1
• /
• pp.71-85
• /
• 2021

Representative domestic start-up support organizations include the Business Incubator(BI), Korea Institute of Startup & Entrepreneurship Development(KISED), Techno Park(TP), and Center of Creative Economy Innovation(CCEI), and there are about 260 Business incubator nationwide. The Business incubator is operated by universities, research institutes, and private foundations or associations. The organization consists of the center director and the incubating professionals (hereinafter referred to as "manager"), etc., and performs tasks such as center operation management and incubation support services for tenant companies. Until now, research on the operation of Business Incubator has been mainly focused on the performance of tenant companies. Studies on whether the manager's competency characteristics directly or indirectly affect the performance of the tenant companies through psychological mediators such as self-efficacy and organizational commitment were very scarce. The purpose of this study is to explore various factors influencing organizational commitment and job performance by the competence characteristics of Business incubator managers, and to explain the causal relationship among those factors. In particular, the difference in perception was investigated by a manager's survey that influences organizational commitment and work performance at the Business incubator. Through this, we intend to present practical implications for the role of managers in the operation of Business incubators. This study is an exploratory study, and the subject of the study was a survey of about 600 managers working at Business incubator nationwide, of which 116 responses were analyzed. Data analysis included descriptive statistics, exploratory factor analysis, and reliability. Structural equation model analysis was performed for hypothesis tests. As a result of the analysis, it was found that the cognitive characteristics of the Business incubator manager, communication, and situational response as the behavioral characteristics had a positive effect on the manager's self-efficacy, and the behavioral characteristics had a greater effect on the self-efficacy. It was also found that the manager's cognitive and behavioral characteristics, and self-efficacy had a positive effect on organizational commitment and work performance. In particular, a manager's self-efficacy has a positive effect on organizational commitment and work performance. This result showed that the manager's competency characteristics increase the manager's self-efficacy as a mediating factor rather than directly affecting organizational commitment and work performance. This study explains that the manager's competency characteristics are transferred to organizational commitment and work performance. The results of the study are expected to reflect the job standard of the National Competency Standards (NCS) and basic vocational competency to the job competency of managers, and it also provides a guideline for the effective business incubator operation in terms of human resource management. In practice, it is expected that the results of the study can reflect the vocational basic skills of the Business Incubator manager's job competency in the National Competency Standards(NCS) section, and suggest directions for the operation of the Business Incubator and the manager's education and training.