• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.323-338
• /
• 2023

TV advertising with deep analysis of watching pattern of audiences is important to set-top box audience targeting. Applying session-based recommendation model(SBR) to internet commercial, or recommendation based on searching history of user showed its effectiveness in previous studies, but applying SBR to the TV advertising was difficult in South Korea due to data unavailabilities. Also, traditional SBR has limitations for dealing with user preferences, especially in data with user identification information. To tackle with these problems, we first obtain set-top box data from three major broadcasting companies in South Korea(SKB, KT, LGU+) through collaboration with Korea Broadcast Advertising Corporation(KOBACO), and this data contains of watching sequence of 4,847 anonymized users for 6 month respectively. Second, we develop personalized session-based recommendation model to deal with hierarchical data of user-session-item. Experiments conducted on set-top box audience dataset and two other public dataset for validation. In result, our proposed model outperformed baseline model in some criteria.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.3B
• /
• pp.311-319
• /
• 2006

A bed level change model(SED-FLUX) is introduced based on the realistic sediment transport process including bed load and suspended load behaviours at the bottom boundary layer. The model SED-FLUX includes wave module, hydrodynamic module and sediment transport and diffusion module that calculate suspended sediment concentration, net sediment erosion flux($Q_s$) and bed load flux. Bed load transport rate is evaluated by the van Rijn's TRANSPOR program which has been verified in wave-current fields. The net sediment erosion flux($Q_s$) at the bottom is evaluated as a source/sink term in the numerical sediment diffusion model where the suspended sediment concentration becomes a verification parameter of the $Q_s$. Bed level change module calculates a bed level change amount(${\Delta}h_{i,j}$) and updates a bed level. For the model verification the limit depth of the bed load transport is compared with the field experiment data and some formula on the threshold depth for the bed load movement by waves and currents. This model is applied to the beach profile changes by waves, then the model shows a clear erosion and accumulation profile according to the incident wave characteristics. Finally the beach evolution by waves and wave-induced currents behind the offshore breakwater is calculated, where the model shows a tombolo formation in the landward area of the breakwater.

• 한국노년학
• /
• v.31 no.4
• /
• pp.1119-1134
• /
• 2011

The purpose of this study is firstly to understand socio-demographic and health characteristics and economic characteristics of elderly single households aged over 60 and secondly to examine the relations between these factors and level of life satisfaction, especially differentiated relations by income level. We used the data drawn from third KREIS (Korean Retirement and Income Study) surveyed by National Pension Research Institute. The statistical methods used for the analyses were t-test, X², multiple regression analysis. For the whole sample, the results showed that the life satisfaction is positively related to higher income, better physical and emotional health status and having a religion. When we conducted the regression on two groups, the religion and income level were no longer significant factors. On the other hand, being a woman and enjoying good health contributed to life satisfaction for lower-income group while having a job and enjoying good health played a positive role in life satisfaction for higher-income group.

• 한국노년학
• /
• v.27 no.2
• /
• pp.371-392
• /
• 2007

Marital status has drawn much attention as previous studies have been pointed it out as a key factor of health. Nonetheless, systematic studies on elderly marital status and health have been quite limited, for most researches have neglected the varying effects of marital status on health over the lifespan and mainly focused on midlife. This study, using nationwide survey on elderly population, attempts to discover the health differentials between widowed and married elderly and explain the differences though the structure and function of their social network. The results reveal that the magnitudes of marital status effects were differed by dimensions of health. The widowed were more likely to be unhappy than married, even after controlling the socio-demographic characteristics. In physical health, however, the health differences between widowed and married were less noticeable or disappeared, as sex, age and other structure factors were considered. Furthermore, the strength of social network factors affecting the physical and mental health of elderly turned out to be different between married and widowed: widowed were more likely than married to be affected by the contact with children and less likely to be affected by contact with friends/relatives. Such results had both positive and negative impact on physical and mental health of widowed. This may imply the difference in pathways of regulating health among the married and widowed elderly.

• 한국노년학
• /
• v.39 no.3
• /
• pp.635-652
• /
• 2019

A few studies have examined the relationship between living arrangements and mental health status among older Korean immigrants in the United States and older Koreans in South Korea. This study attempts to fill this gap by conducting a comparative study to understand the relationship between living arrangements and psychological distress. Survey data from older Koreans between the ages of 60 and 79 from the two countries (N= 480) was analyzed descriptively and in hierarchical multiple regressions. This study found that among older Korean immigrants in the United States 26.4 % of those living alone and 7.3 % of those living with a spouse only reported 'severe' psychological distress while their counterparts in South Korea 20.0 % of those living alone and 20.6 % of those living with a spouse only reported 'severe' psychological distress. The hierarchical multivariate analysis reveled that older Koreans living alone are not significant predictor of psychological distress in both countries. Interestingly, older Korean immigrants living with a spouse only and living with others are significant predictors of psychological distress. This study also contributes to the existing literature by searching for diverse conditions that lead to mental health problems among older Koreans in both countries.

• 한국노년학
• /
• v.39 no.1
• /
• pp.145-167
• /
• 2019

The purpose of this study is to examine the solidarity types between parents-children and to verify whether the types of solidarity relationships are directly related to life satisfaction of older adults. To achieve this, 2,072 Korean elderlies from proportional stratified sample were participated, and the data was divided into urban and rural area where respondents live in. The Latent Profile Analysis(LPA), multiple regression analysis were mainly used for data analysis. The results of the study were as follows: The major findings are following. Patterns of parental bond among urban and rural elderly were classified into three(urban) and two(rural) patterns. For the effect of each parental bonding pattern on life satisfaction, positive effect of parental bond was found only in urban dwellers. In other words, for older people in rural areas, parental bond did not significantly affected on life satisfaction. However, elderly in rural area showed non-familial factor based relationship, such as social capital(community trust, social cohesion, and social support), was influenced their life satisfaction rather than relationship with children. Finding from the study highlight political and practical implications for improving life satisfaction for the elderly.

• Journal of the Korean Society for information Management
• /
• v.40 no.4
• /
• pp.307-327
• /
• 2023

To analyze the impact of word embedding on book titles, this study utilized word embedding models (Word2vec, GloVe, fastText) to generate embedding vectors from book titles. These vectors were then used as classification features for automatic classification. The classifier utilized the k-nearest neighbors (kNN) algorithm, with the categories for automatic classification based on the DDC (Dewey Decimal Classification) main class 300 assigned by libraries to books. In the automatic classification experiment applying word embeddings to book titles, the Skip-gram architectures of Word2vec and fastText showed better results in the automatic classification performance of the kNN classifier compared to the TF-IDF features. In the optimization of various hyperparameters across the three models, the Skip-gram architecture of the fastText model demonstrated overall good performance. Specifically, better performance was observed when using hierarchical softmax and larger embedding dimensions as hyperparameters in this model. From a performance perspective, fastText can generate embeddings for substrings or subwords using the n-gram method, which has been shown to increase recall. The Skip-gram architecture of the Word2vec model generally showed good performance at low dimensions(size 300) and with small sizes of negative sampling (3 or 5).

• Journal of Navigation and Port Research
• /
• v.47 no.6
• /
• pp.430-436
• /
• 2023

Study on Changes in Vessel Traffic Services Due to Introduction of Maritime Autonomous Surface ShipsThe development of technologies related to Maritime Autonomous Surface Ships (MASS) has been actively progressing since the mid-2010s, focusing on themes such as collision avoidance, route planning, digital twin, and communication technologies. On the other hand, research on land-based infrastructure connected with MASS, such as logistics systems, port facilities, and vessel traffic services, has relatively received less attention. This study analyzed impact of emergence of MASS on existing vessel traffic service operations and proposed changes in control operations to prepare for its impact. To do this, current vessel traffic service operations were analyzed and elements of MASS technology that could affect vessel traffic control were identified. A survey was conducted among vessel traffic controllers to identify items related to the control of MASS. Results analyzed using the AHP method showed that preparation for emergency response and communication methods with MASS were the most important. Based on this, we were able to derive detailed plans for basic MASS control procedures and emergency response procedures based on data communication within maritime traffic control areas. MASS control procedures proposed in this study are expected to be used as a solution to resolve issues related to traffic safety of MASS in coastal areas.

• Journal of the Korean Regional Science Association
• /
• v.40 no.2
• /
• pp.107-130
• /
• 2024

Urban land surface temperature (LST) change is a major environmental factor that affects the thermal comfort, energy consumption, and health of urban residents. Most studies that explored the relationship between LST and urban built-environment form analyzed only midday LST. This study explores the diurnal variation of summertime LST in Seoul using ECOSTRESS data, which observes LST at various times of the day and analyzes whether the LST variation differs by built environment type. Launched in 2018, ECOSTRESS operates in a non-sun-synchronous orbit, observing LST with a high resolution of 70 meters. This study collected data from early morning (6:25) to evening (17:26) from 2019 to 2022 to build time-series LST. Based on greenery, water bodies, and building form data, eight types of Seoul's built environment were derived by hierarchical clustering, and the LST fluctuation characteristics of each cluster were compared. The results showed that the spatial disparity in LST increased after dawn, peaked at noon, and decreased again, highlighting areas with rapid versus stable LST changes. Low-rise and high-rise compact districts experienced fast, high temperature increases and high variability, while low-density apartments experienced moderate LST increases and low variability. These results suggest urban forms that can mitigate rapid daytime heating.

• Information Systems Review
• /
• v.21 no.4
• /
• pp.47-68
• /
• 2019

As cloud services and deployment models become diverse, there are a growing number of cloud computing selection options. Therefore, financial companies need a methodology to select the appropriated cloud for each financial computing system. This study adopted the Balanced Scorecard (BSC) framework to classify factors for the introduction of cloud computing in financial companies. Using Analytic Hierarchy Process (AHP), the evaluation items are layered into the performance perspective and the cloud consideration factor and a comprehensive decision model is proposed. To verify the proposed research model, a system of financial company is divided into three: account, information, and channel system, and the result of decision making by both financial business experts and technology experts from two financial companies were collected. The result shows that some common factors are important in all systems, but most of the factors considered are very different from system to system. We expect that our methodology contributes to the spread of cloud computing adoption.