• KIISE Transactions on Computing Practices
• /
• v.24 no.3
• /
• pp.138-144
• /
• 2018

A video is a vivid medium similar to human's visual-linguistic experiences, since it can inculcate a sequence of situations, actions or dialogues that can be told as a story. In this study, we propose story learning/regeneration frameworks from videos with successive event order supervision for contextual coherence. The supervision induces each episode to have a form of trajectory in the latent space, which constructs a composite representation of ordering and semantics. In this study, we incorporated the use of kids videos as a training data. Some of the advantages associated with the kids videos include omnibus style, simple/explicit storyline in short, chronological narrative order, and relatively limited number of characters and spatial environments. We build the encoder-decoder structure with successive event order embedding, and train bi-directional LSTMs as sequence models considering multi-step sequence prediction. Using a series of approximately 200 episodes of kids videos named 'Pororo the Little Penguin', we give empirical results for story regeneration tasks and SEOE. In addition, each episode shows a trajectory-like shape on the latent space of the model, which gives the geometric information for the sequence models.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1413-1425
• /
• 2023

The increasing frequency of wildfires due to climate change is causing extreme loss of life and property. They cause loss of vegetation and affect ecosystem changes depending on their intensity and occurrence. Ecosystem changes, in turn, affect wildfire occurrence, causing secondary damage. Thus, accurate estimation of the areas affected by wildfires is fundamental. Satellite remote sensing is used for forest fire detection because it can rapidly acquire topographic and meteorological information about the affected area after forest fires. In addition, deep learning algorithms such as convolutional neural networks (CNN) and transformer models show high performance for more accurate monitoring of fire-burnt regions. To date, the application of deep learning models has been limited, and there is a scarcity of reports providing quantitative performance evaluations for practical field utilization. Hence, this study emphasizes a comparative analysis, exploring performance enhancements achieved through both model selection and data design. This study examined deep learning models for detecting wildfire-damaged areas using Landsat 8 satellite images in California. Also, we conducted a comprehensive comparison and analysis of the detection performance of multiple models, such as U-Net and High-Resolution Network-Object Contextual Representation (HRNet-OCR). Wildfire-related spectral indices such as normalized difference vegetation index (NDVI) and normalized burn ratio (NBR) were used as input channels for the deep learning models to reflect the degree of vegetation cover and surface moisture content. As a result, the mean intersection over union (mIoU) was 0.831 for U-Net and 0.848 for HRNet-OCR, showing high segmentation performance. The inclusion of spectral indices alongside the base wavelength bands resulted in increased metric values for all combinations, affirming that the augmentation of input data with spectral indices contributes to the refinement of pixels. This study can be applied to other satellite images to build a recovery strategy for fire-burnt areas.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_3
• /
• pp.913-924
• /
• 2022

In this paper, we presented a database construction of undersea images for the Habitats of Ecklonia cava and Sargassum and conducted an experiment for semantic segmentation using state-of-the-art (SOTA) models such as High Resolution Network-Object Contextual Representation (HRNet-OCR) and Shifted Windows-L (Swin-L). The result showed that our segmentation models were superior to the existing experiments in terms of the 29% increased mean intersection over union (mIOU). Swin-L model produced better performance for every class. In particular, the information of the Ecklonia cava class that had small data were also appropriately extracted by Swin-L model. Target objects and the backgrounds were well distinguished owing to the Transformer backbone better than the legacy models. A bigger database under construction will ensure more accuracy improvement and can be utilized as deep learning database for undersea images.

• Journal of the Korean Society for Library and Information Science
• /
• v.51 no.4
• /
• pp.99-120
• /
• 2017

In order to develop a semiautomatic support system that allows researchers concerned to efficiently analyze the technical trends for the ever-growing industry and market. This paper introduces a couple of Korean sentence generation models that can automatically generate definitional statements as well as descriptions of technical terms and concepts. The proposed models are based on a deep learning model called LSTM (Long Sort-Term Memory) capable of effectively labeling textual sequences by taking into account the contextual relations of each item in the sequences. Our models take technical terms as inputs and can generate a broad range of heterogeneous textual descriptions that explain the concept of the terms. In the experiments using large-scale training collections, we confirmed that more accurate and reasonable sentences can be generated by CHAR-CNN-LSTM model that is a word-based LSTM exploiting character embeddings based on convolutional neural networks (CNN). The results of this study can be a force for developing an extension model that can generate a set of sentences covering the same subjects, and furthermore, we can implement an artificial intelligence model that automatically creates technical literature.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.419-437
• /
• 2023

As the number of thin filers in Korea surpasses 12 million, there is a growing interest in enhancing the accuracy of assessing their credit default risk to generate additional revenue. Specifically, researchers are actively pursuing the development of default prediction models using machine learning and deep learning algorithms, in contrast to traditional statistical default prediction methods, which struggle to capture nonlinearity. Among these efforts, Graph Neural Network (GNN) architecture is noteworthy for predicting default in situations with limited data on thin filers. This is due to their ability to incorporate network information between borrowers alongside conventional credit-related data. However, prior research employing graph neural networks has faced limitations in effectively handling diverse categorical variables present in credit information. In this study, we introduce the Transformer embedded Graph Convolutional Network (TeGCN), which aims to address these limitations and enable effective default prediction for thin filers. TeGCN combines the TabTransformer, capable of extracting contextual information from categorical variables, with the Graph Convolutional Network, which captures network information between borrowers. Our TeGCN model surpasses the baseline model's performance across both the general borrower dataset and the thin filer dataset. Specially, our model performs outstanding results in thin filer default prediction. This study achieves high default prediction accuracy by a model structure tailored to characteristics of credit information containing numerous categorical variables, especially in the context of thin filers with limited data. Our study can contribute to resolving the financial exclusion issues faced by thin filers and facilitate additional revenue within the financial industry.

• Human Ecology Research
• /
• v.51 no.5
• /
• pp.497-511
• /
• 2013

This study aimed to examine Korean men's experiences of work-family balance in dual income families with children under six years of age. We focused on identifying the difficulty of balancing work and family considering their individual, social, and cultural conditions. The method was a qualitative study involving two in-depth interviews with each of 12 men, and analyzing the data through the grounded theory approach. From the results, a model of men's work-family experience was constructed. It demonstrates the central phenomena (difficulties of balancing), the causal conditions (lacking time for family, seeking support from the employer, and learning husband's roles insufficiently), the contextual conditions (remaining paternalism and changing husband's roles), the intervening conditions (workplace, childcare support, and wife characteristics), and strategies (help from relatives, utilizing daycare centers, controlling birth, managing work conditions, and using family polices). We clarify the overall picture of working and family life experiences, and also show how men deal with their problems in their circumstances by balancing working and family life. In conclusion, males have difficulty participating in family life autonomously because of having less decision-making power than the wife. Moreover, the great responsibilities of the breadwinner disturb the work-family balance. Men devote themselves to working to hold a job instead of spending time with their family. However, they ultimately value work-family balance with respect to 'keeping a peaceful family life'.

• Journal of the Korean association of regional geographers
• /
• v.20 no.4
• /
• pp.484-503
• /
• 2014

This study is to point out importance of narrative as a way of learning values and interests for learners. Thus, geography value instruction models with narrative was developed and applied to teaching model. The results are as follows: First, narrative is useful to value education. Because narrative includes the contextual information, leaners are able to make a moral decision in respect of socio-cultural approach and to reach more truthful and practical decision with empathic understanding. Second, comparing with an expository text, the narrative text is more interesting, understandable, and preferred. Third, Web of meaning, one of scaffolding skills, is helpful to expand the scope of learner's thinking and group activity. Fourth, learner's awareness toward the topic changes. However, it is required to develop a practical method for elaborate assessment tool and for learners' active participation.

• Journal of the Korean Society for information Management
• /
• v.29 no.4
• /
• pp.273-294
• /
• 2012

Images have been substantially searched and used due to not only the advanced internet and digital technologies but the characteristics of a younger generation. The purpose of this study aims to discuss the ways on expanding the access points to images by analyzing the needs of users in context of everyday life. In order to achieve the purpose of this study, 105 questions of image seeking in NAVER, which is one of social Q&A services in Korea, were analyzed. For the analysis, a two-dimensional framework with image uses and image attributes were utilized. The findings of this study demonstrate that considerable use purposes on data oriented pole, such as information processing, information dissemination and learning are identified. On the other hand, image attributes from the needs of image show that non-visual aspects including contextual attributes are recognized substantially in addition to the traditional semantic attributes.

• Journal of Korean Elementary Science Education
• /
• v.41 no.1
• /
• pp.44-64
• /
• 2022

This study explored the elementary school teachers' experiences while teaching students with low achievement in science based on the grounded theory. In-depth interviews and analysis were conducted on 13 teachers with experiences in teaching students with low achievement in science within the last three years and more than five years of field experience until the theoretical saturation of data on the teaching experiences for students with low achievement in science. The analysis results were as follows. First, the teaching experiences of elementary school teachers for underachievers in science were classified into 119 concepts, 41 subcategories, and 17 categories. Based on the paradigm model, the categories were structured and presented as causal conditions, contextual conditions, intervening conditions, action/interaction strategies and consequences based on the central phenomenon of 'difficulty in teaching students with low achievement in science'. Second, the core category of elementary school teachers' teaching underachievers in science was assumed to be 'overcoming difficulties and teaching underachievers in science'. And according to the properties and dimensions of the core category, teachers who teaching students with low achievement in science were divided into four types: 'compromising-', 'overcoming-', 'accepting-', and 'conflicting-reality type'. Third, a conditional matrix was presented to summarize and integrate the results of this study by classifying the teaching experience of elementary school teachers for underachievers in science into educational providers and educational demanders. On the basis of these findings, educational implications for teaching students with low achievement in science were discussed.