• Journal of Nutrition and Health
• /
• v.54 no.5
• /
• pp.534-546
• /
• 2021

Purpose: This study aimed at understanding meal kit purchasing behavior and investigating its relationship with the Nutrition Quotient (NQ) of young adults. Methods: We conducted a survey on adults in their 20s and 30s covering their meal kit purchase experience, satisfaction, recognition, and purchase intention, and examined the relationship between the meal kit purchase and their NQ from February to March 2021. Results: Among the 404 subjects, 37.9% of males and 48.0% of females had experience in purchasing a meal kit (p < 0.001). The highest response indicated that the purchase cost of meal kits was 10,000-20,000 Won at a time, and the frequency of purchase was less than once a month. The convenience of cooking was the main reason for the purchase of meal kits, which were consumed mainly in the evening with family. The satisfaction with the purchase experience of a meal kit was rated 3.6 points for males and 3.7 points for females out of 5 points, and the satisfaction experienced by women was significantly higher than men in terms of freshness of ingredients, packaging design, and adequacy of the quantity of content (p < 0.05). Recognition of the meal kit was rated 3.5 points for males and 3.7 points for females out of 5 points. The purchase intention of the meal kit was rated 3.8 points for those with prior purchase experience, 3.2 points for the non-experienced, 3.3 points for males, and 3.6 points for females out of 5 points each (p < 0.001). The NQ score of dietary behavior in females with experience of meal kit purchases was significantly higher than non-experience (p < 0.05). Conclusions: The dietary behavior of female showed a significant difference by the meal kit purchase experience. It is necessary to understand the consumers' meal kit purchasing behavior to enable the development of various health-oriented meal kit products.

• 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 Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.157-173
• /
• 2021

With the development of information technology, the amount of available information increases daily. However, having access to so much information makes it difficult for users to easily find the information they seek. Users want a visualized system that reduces information retrieval and learning time, saving them from personally reading and judging all available information. As a result, recommendation systems are an increasingly important technologies that are essential to the business. Collaborative filtering is used in various fields with excellent performance because recommendations are made based on similar user interests and preferences. However, limitations do exist. Sparsity occurs when user-item preference information is insufficient, and is the main limitation of collaborative filtering. The evaluation value of the user item matrix may be distorted by the data depending on the popularity of the product, or there may be new users who have not yet evaluated the value. The lack of historical data to identify consumer preferences is referred to as data sparsity, and various methods have been studied to address these problems. However, most attempts to solve the sparsity problem are not optimal because they can only be applied when additional data such as users' personal information, social networks, or characteristics of items are included. Another problem is that real-world score data are mostly biased to high scores, resulting in severe imbalances. One cause of this imbalance distribution is the purchasing bias, in which only users with high product ratings purchase products, so those with low ratings are less likely to purchase products and thus do not leave negative product reviews. Due to these characteristics, unlike most users' actual preferences, reviews by users who purchase products are more likely to be positive. Therefore, the actual rating data is over-learned in many classes with high incidence due to its biased characteristics, distorting the market. Applying collaborative filtering to these imbalanced data leads to poor recommendation performance due to excessive learning of biased classes. Traditional oversampling techniques to address this problem are likely to cause overfitting because they repeat the same data, which acts as noise in learning, reducing recommendation performance. In addition, pre-processing methods for most existing data imbalance problems are designed and used for binary classes. Binary class imbalance techniques are difficult to apply to multi-class problems because they cannot model multi-class problems, such as objects at cross-class boundaries or objects overlapping multiple classes. To solve this problem, research has been conducted to convert and apply multi-class problems to binary class problems. However, simplification of multi-class problems can cause potential classification errors when combined with the results of classifiers learned from other sub-problems, resulting in loss of important information about relationships beyond the selected items. Therefore, it is necessary to develop more effective methods to address multi-class imbalance problems. We propose a collaborative filtering model using CGAN to generate realistic virtual data to populate the empty user-item matrix. Conditional vector y identify distributions for minority classes and generate data reflecting their characteristics. Collaborative filtering then maximizes the performance of the recommendation system via hyperparameter tuning. This process should improve the accuracy of the model by addressing the sparsity problem of collaborative filtering implementations while mitigating data imbalances arising from real data. Our model has superior recommendation performance over existing oversampling techniques and existing real-world data with data sparsity. SMOTE, Borderline SMOTE, SVM-SMOTE, ADASYN, and GAN were used as comparative models and we demonstrate the highest prediction accuracy on the RMSE and MAE evaluation scales. Through this study, oversampling based on deep learning will be able to further refine the performance of recommendation systems using actual data and be used to build business recommendation systems.

• (The) Research of the performance art and culture
• /
• no.32
• /
• pp.31-70
• /
• 2016

Jongmyo Jeryeak [Royal ancestral shrine music] is a precious cultural heritage, which has been played till now since two great kings Sejong and Sejo who made it. But going through Japanese occupation, Jongmyo Jeryeak has been changed into a music totally different from the music the two kings intended. And the changed Jongmyo Jeryeak is being played these days. The original Jongmyo Jeryeak, which was made by the two kings, remains in old music scores. Therefore there is a need to investigate the differences between the original and current Jongmyo Jeryeak by interpreting the old music scores and recovering the original. This paper recovers the origianl Jongmyo Jeryeak from the music score Daeakhubo and compares it with the current Jongmyo Jeryeak. The results are as follows. The origianl Jongmyo Jeryeak is a set of common songs made with Hyangak and Gochiak to sing the verses which extol royal ancestors' virtues. All of the musical elements are matched with the verses so that the meaning of the verses is transmitted naturally. Jangdan musical time musical structure are matched with the structure of verses, and the musical motif of the songs is matched with the meaning of the verses. The music, which is easy and expresses the meaning of the verses well, demonstrates King Sejong's talent as a musician. The current Jongmyo Jeryeak is a set of special songs in which Sigimsae is emphasized rather than the meaning of the verses. The melodies are broken into pieces inconsistently, the meaningless word 'ae' is added thoughtlessly, and Jangdan musical time musical structure are unrelated to the verses. Therefore the meaning of the verses is not transmitted at all. These changes, which were made during the period of Japanese occupation, seem to desecrate the verses of the original songs. The melodies, which are broken into pieces inconsistently, revive into the current mysterious ritual music through Sigimsae. But in order to be a proper ritual music, some corrections have to be made to convey the meaning of the verses.

• Journal of agriculture & life science
• /
• v.51 no.2
• /
• pp.117-126
• /
• 2017

Hanwoo young bulls are selected based on performance test using the weight at 12 months and pedigree index comprising marbling score. Pedigree index was not based on the progeny tested data but the breeding value of the proven bulls; resulting a lower accuracy. The progeny testing of the young bulls was categorized into testing at farm and at the test station. The farm tested data was difficult to compare with those from test station data. Farm tested bulls had different slaughter ages than those for test station bulls. Therefore, this study had considered a different age at slaughter for respective records on ultrasound traits. Records on body weight at 12 months, ultrasound measures at 12 and 24 months(uIMF, uEMA, uBFT, and uRFT), and carcass traits(CWT, EMA, BFT, and MS) were collected from steers and bulls of Hanwoo national improvement scheme between 2008 and 2013. Fixed effects of batch, test date, test station, personnel for measurement, personnel for judging, and a linear covariate of weight at measurement were fitted in the animal models for ultrasound traits. The ranges of heritability estimates of the ultrasound traits at 12 and 24 months were 0.21-0.43 and 0.32-0.47, respectively. Ultrasound traits at 12 and 24 months between similar carcass traits was genetically correlated at 0.52-0.75 and 0.86-0.89, respectively.

• 한국사회정책
• /
• v.24 no.1
• /
• pp.153-185
• /
• 2017

The purpose of this study is to investigate the actual conditions and causes of the problem behaviors of Korean Chinese'left-behind'children in China in order to propose a support system to prevent problem behaviors of them. For this purpose, a questionnaire survey was conducted on 399 children who attend at three Korean Chines schools in Yonbian in China. The questionnaire consisted of general characteristics, internalizing problem behavior, social support, self-esteem, and self-resilience. This paper analysed the survey data by employing one-way ANOVA and a structural equation modeling. It verified if there is significant difference in internalizing problem behaviour, self-esteem, self-resilience, and social support between left-behind children's group and non left-behind children's group. It also identified a structural causal relationship and direct or indirect effects among problematic behaviour, self-esteem, self-resilience, and social support. The results of the analysis are as follows. First, there was a statistically significant difference in the social withdrawal and depression of internalizing problem behaviors between left-behind children's group and non left-behind children's group. Second, the left-behind children's group showed no significant difference in self-resilience and social support compared to non left-behind children's group, but showed a significant difference in self-esteem. In the positive self- esteem factor, non left-behind children's group showed much higher score whereas left-behind children's group was higher in the negative self-esteem factor. Third, social support for left-behind children's group has a statistically significant direct negative effect on internalizing problem behaviors, and indirectly negative effects on problem behavior through self-resilience. These results suggest the necessity of establishing a social support system for mitigating and preventing problem behaviors and the necessity of preparing measures to improve self-resilience. Based on the results of the study, we discussed how to establish a social support system in China to mitigate internalizing problem behaviors of Korean Chinese left-behind children.

• Journal of agriculture & life science
• /
• v.51 no.6
• /
• pp.89-105
• /
• 2017

This study utilized 32,312 records of 17 linear type and 10 conformation traits(including final scores) of Hanwoo cows in the KAIA(Korea Animal Improvement Association) ('09~'10), with 60,556 animals in the pedigree file. Traits included stature, body length, strength, body depth, angularity, shank thickness, rump angle, rump length, pin bone width, thigh thickness, udder volume, teat length, teat placement, foot angle, hock angle, rear leg back view, body balance, breed characteristic, head development, forequarter quality, back line, rump, thigh development, udder development, leg line, and final score. Genetic and residual(co) variances were estimated using bi-trait pairwise analyses with EM-REML algorithm. Herd-year-classifier, year at classification, and calving stage were considered as fixed effects with classification months as a covariate. The heritability estimates ranged from 0.03(teat placement) to 0.42(body length). Rump length had the highest positive genetic correlation with pin bone width(0.96). Moreover, stature, body length, strength, and body depth had the highest positive genetic correlations with rump length, pin bone width, and thigh thickness(0.81-0.94). Stature, body length, strength, body depth, rump length, pin bone width, and thigh thickness traits also had high positive genetic correlations.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.49 no.3
• /
• pp.1-10
• /
• 2021

This study aimed to evaluate preferences of Bukhansan dulegil using sentiment analysis, a natural language processing technique, to derive preferred and non-preferred factors. Therefore, we collected blog articles written in 2019 and produced sentimental scores by the derivation of positive and negative words in the texts for 21 dulegil courses. Then, content analysis was conducted to determine which factors led visitors to prefer or dislike each course. In blogs written about Bukhansan dulegil, positive words appeared in approximately 73% of the content, and the percentage of positive documents was significantly higher than that of negative documents for each course. Through this, it can be seen that visitors generally had positive sentiments toward Bukhansan dulegil. Nevertheless, according to the sentiment score analysis, all 21 dulegil courses belonged to both the preferred and non-preferred courses. Among courses, visitors preferred less difficult courses, in which they could walk without a burden, and in which various landscape elements (visual, auditory, olfactory, etc.) were harmonious yet distinct. Furthermore, they preferred courses with various landscapes and landscape sequences. Additionally, visitors appreciated the presence of viewpoints, such as observation decks, as a significant factor and preferred courses with excellent accessibility and information provisions, such as information boards. Conversely, the dissatisfaction with the dulegil courses was due to noise caused by adjacent roads, excessive urban areas, and the inequality or difficulty of the course which was primarily attributed to insufficient information on the landscape or section of the course. The results of this study can serve not only serve as a guide in national parks but also in the management of nearby forest green areas to formulate a plan to repair and improve dulegil. Further, the sentiment analysis used in this study is meaningful in that it can continuously monitor actual users' responses towards natural areas. However, since it was evaluated based on a predefined sentiment dictionary, continuous updates are needed. Additionally, since there is a tendency to share positive content rather than negative views due to the nature of social media, it is necessary to compare and review the results of analysis, such as with on-site surveys.

• Korean Journal of Psychosomatic Medicine
• /
• v.29 no.1
• /
• pp.67-76
• /
• 2021

Objectives : The aim of this study was to compare activities of daily living (ADLs) according to degenerative changes in brain [i.e., medial temporal lobe atrophy (MTA), white matter hyperintensities] and to examine the association between neurocognitive functions and ADLs in Korean patients with dementia due to Alzheimer's disease (AD) and mild cognitive impairment (MCI). Methods : Participants were 111 elderly subjects diagnosed with AD or MCI in this cross-sectional study. MTA in brain MRI was rated with standardized visual rating scales (Scheltens scale) and the subjects were divided into two groups according to Scheltens scale. ADLs was evaluated with the Korean version of Blessed Dementia Scale-Activity of daily living (BDS-ADL). Neurocognitive function was evaluated with the Korean version of the Consortium to Establish a Registry for Alzheimer's Disease assessment packet (CERAD-K). Independent t-test was performed to compare ADLs with the degree of MTA. Pearson correlation and hierarchical multiple regression analyses were performed to analyze the relationship between ADLs and neurocognitive functions. Results : The group with high severity of the MTA showed significantly higher BDS-ADL scores (p<0.05). The BDS-ADL score showed the strongest correlation with the word list recognition test among sub-items of the CERAD-K test (r=-0.568). Findings from the hierarchical multiple regression analysis revealed that the scores of MMSE-K and word list recognition test were factors that predict ADLs (F=44.611, p<0.001). Conclusions : ADLs of AD and MCI patients had significant association with MTA. Our study, which identifies factors correlated with ADLs can provide useful information in clinical settings. Further evaluation is needed to confirm the association between certain brain structures and ADLs.

• Sleep Medicine and Psychophysiology
• /
• v.27 no.2
• /
• pp.56-66
• /
• 2020

Objectives: In this study, the clinical characteristics of OSA patients and the quality of life before and after CPAP use were compared to determine the degree of improvement in quality of life according to CPAP use. Methods: Age, sex, height, weight, body mass index, Epworth Sleepiness Scale, Modified Mallampatti Score, Montreal Cognitive Assessment-Korean, and Pittsburgh Sleep Quality Index were compared between men and women through medical records. To understand the degree of improvement in quality of life resulting from use of CPAP, a personal telephone call was made to compare the VAS scores for quality of life before and after CPAP use. Results: In height (HT) (Z = -4.525, p < 0.001), weight (BW) (Z = -2.844, p < 0.05), sleep quality (PSQI) (Z = -2.671, p < 0.05), and arousal index (AI) (Z = -2.105, p < 0.05), there was a difference between men and women (p < 0.05). There was no difference in the remaining variables. Cross-analysis (Chi-square test) confirmed a difference between severity and sex of OSA. It has been found that there is no statistically significant order in size according to level-specific severity of OSA for PreCPAP QOL, PostCPAP QOL, CPAPUse Months, and CPAP4Hr/d (%) (p > 0.05). The difference between AHI before and after CPAP was 36.48 ± 21.54 (t = 11.609, p < 0.001) and the difference between QOL before and after CPAP was -25.43 ± 22.06 (t = -7.901, p < 0.001), both of which were significant (p < 0.001). Conclusion: Among OSA patients, there were differences in height (HT), weight (BW), sleep quality (PSQI), arousal index (AI), and severity of OSA between men and women, but the quality of life before and after CPAP was different. However, there was no difference between men and women in quality of life before and after CPAP. In addition, quality of life in OSA patients improved after using CPAP.