• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.455-460
• /
• 2023

Text augmentation is a methodology that creates new augmented texts by transforming or generating original texts for the purpose of improving the performance of NLP models. However existing text augmentation techniques have limitations such as lack of expressive diversity semantic distortion and limited number of augmented texts. Recently text augmentation using large language models and few-shot learning can overcome these limitations but there is also a risk of noise generation due to incorrect generation. In this paper, we propose a text augmentation method called TAGS that generates multiple candidate texts and selects the appropriate text as the augmented text. TAGS generates various expressions using few-shot learning while effectively selecting suitable data even with a small amount of original text by using contrastive learning and similarity comparison. We applied this method to task-oriented chatbot data and achieved more than sixty times quantitative improvement. We also analyzed the generated texts to confirm that they produced semantically and expressively diverse texts compared to the original texts. Moreover, we trained and evaluated a classification model using the augmented texts and showed that it improved the performance by more than 0.1915, confirming that it helps to improve the actual model performance.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.461-470
• /
• 2023

While speech animation generation employing deep learning has been actively researched for English, there has been no prior work for Korean. Given the fact, this paper for the very first time employs supervised deep learning to generate Korean speech animation. By doing so, we find out the significant effect of deep learning being able to make speech animation research come down to speech recognition research which is the predominating technique. Also, we study the way to make best use of the effect for Korean speech animation generation. The effect can contribute to efficiently and efficaciously revitalizing the recently inactive Korean speech animation research, by clarifying the top priority research target. This paper performs this process: (i) it chooses blendshape animation technique, (ii) implements the deep-learning model in the master-servant pipeline of the automatic speech recognition (ASR) module and the facial action coding (FAC) module, (iii) makes Korean speech facial motion capture dataset, (iv) prepares two comparison deep learning models (one model adopts the English ASR module, the other model adopts the Korean ASR module, however both models adopt the same basic structure for their FAC modules), and (v) train the FAC modules of both models dependently on their ASR modules. The user study demonstrates that the model which adopts the Korean ASR module and dependently trains its FAC module (getting 4.2/5.0 points) generates decisively much more natural Korean speech animations than the model which adopts the English ASR module and dependently trains its FAC module (getting 2.7/5.0 points). The result confirms the aforementioned effect showing that the quality of the Korean speech animation comes down to the accuracy of Korean ASR.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_2
• /
• pp.827-835
• /
• 2023

Drone light detection and ranging (LiDAR) is a state-of-the-art surveying technology that enables close investigation of the top of the mountain slope or the inaccessible slope, and is being used for field surveys in mountainous terrain. To build topographic information using Drone LiDAR, a preprocessing process is required to effectively separate ground and non-ground points from the acquired point cloud. Therefore, in this study, the point group data of the mountain topography was acquired using an aerial LiDAR mounted on a commercial drone, and the application and accuracy of the cloth simulation filtering algorithm, one of the ground separation techniques, was verified. As a result of applying the algorithm, the separation accuracy of the ground and the non-ground was 84.3%, and the kappa coefficient was 0.71, and drone LiDAR data could be effectively used for landslide field surveys in mountainous terrain.

• Industry Promotion Research
• /
• v.8 no.4
• /
• pp.147-162
• /
• 2023

This study explored the experience of social support in the education and rearing of children of immigrant women through international marriage and found its essential meaning. First of all, the husband's social support was very important, but the relationship with the husband had a different effect on childrearing and education. Parents-in-law had a positive and negative impact on child rearing and education of them due to cultural conflicts between the two countries. Their own mother was a strong support that gave them great strength just by being there, and as their children grew up, they regarded their mother as the source of bilingual education for their children. Other supporters around them were Korean friends who connected Korean society by sharing information on child care and education. Friends who spoke and communicated in their native language were emotional and psychological supporters that bonded the same experience of parenting and education for their children. In conclusion, the research participants expected a better life for themselves and their children by using a multi-layered social support system as well as a transnational family network in the process of child education and care. Accordingly, it was proposed to systematically improve the laws, systems, and policy support so that the social support system can be further strengthened at the family, community, and transnational levels for the education and care of children of immigrant women through international marriage.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.6
• /
• pp.255-266
• /
• 2022

The quality control of coarse aggregate among aggregates, which are the main ingredients of concrete, is currently carried out by SPC(Statistical Process Control) method through sampling. We construct a smart factory for manufacturing innovation by changing the quality control of coarse aggregates to inspect the coarse aggregates based on this image by acquired images through the camera instead of the current sieve analysis. First, obtained images were preprocessed, and HED(Hollistically-nested Edge Detection) which is the filter learned by deep learning segment each object. After analyzing each aggregate by image processing the segmentation result, fineness modulus and the aggregate shape rate are determined by analyzing result. The quality of aggregate obtained through the video was examined by calculate fineness modulus and aggregate shape rate and the accuracy of the algorithm was more than 90% accurate compared to that of aggregates through the sieve analysis. Furthermore, the aggregate shape rate could not be examined by conventional methods, but the content of this paper also allowed the measurement of the aggregate shape rate. For the aggregate shape rate, it was verified with the length of models, which showed a difference of ±4.5%. In the case of measuring the length of the aggregate, the algorithm result and actual length of the aggregate showed a ±6% difference. Analyzing the actual three-dimensional data in a two-dimensional video made a difference from the actual data, which requires further research.

• Korean Journal of Culture and Social Issue
• /
• v.9 no.spc
• /
• pp.35-55
• /
• 2003

The purpose of this study is to review the previous studies on the safety problems in Korea and to propose a psychological total safety system model. The model consisted of four agents; the government as the safety management agent, the suppliers of safety goods and services, consumer of safety goods and services, and civil movement institutions for safety. It was emphasized that the culture specific social representations of safety and accident have emerged in the course of rapid industrialization process in Korea during last 30 years. We delineated the social representations of the Korean people on safety and accident according to the model. A psychological analysis of drinking and driving behavior was performed as an application of the model. It was emphasized that safety psychologists have to develope and to apply the knowledge and the information from human engineering psychology and applied social psychology on safety and accidents.

• Knowledge Management Research
• /
• v.24 no.3
• /
• pp.149-172
• /
• 2023

With the development of information technology, new technologies to be introduced in each industry are continuously increasing. This study aims to verify the influence of ambivalent emotions experienced when encountering new technologies, the coping strategies they induce, and their impact on the decision-making process of technology adoption Specifically, this research investigates the emotions and responses to new technologies in the situational context where service providers must deliver services based on new technology in environments where no such services have been developed previously. Furthermore, it seeks to verify the influence of coping responses on the intention to use services based on new technologies. To this end, this study investigated the ambivalent emotions and coping responses of financial sector workers to new financial services based on metaverse technology. As a result of the analysis ambivalance had a significant effect on all four coping responses (disengagement-oriented coping, denial, indecision and compromise). Among them, denial, which is an inflexible response, and compromise, which is a flexible response, had a significant positive effect on the intention to use, and disengagement-oriented coping and indecision had a significant negative effect on the intention to use. The results of this study confirm the user's metaverse acceptance factor and user-centered influence, and are expected to provide guidelines for the introduction of services to practical workers with academic significance.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.6
• /
• pp.1321-1330
• /
• 2023

This study attempts to address the problem of 3D pose estimation for multiple human objects through a single image generated during the character development process that can be used in augmented reality. In the existing top-down method, all objects in the image are first detected, and then each is reconstructed independently. The problem is that inconsistent results may occur due to overlap or depth order mismatch between the reconstructed objects. The goal of this study is to solve these problems and develop a single network that provides consistent 3D reconstruction of all humans in a scene. Integrating a human body model based on the SMPL parametric system into a top-down framework became an important choice. Through this, two types of collision loss based on distance field and loss that considers depth order were introduced. The first loss prevents overlap between reconstructed people, and the second loss adjusts the depth ordering of people to render occlusion inference and annotated instance segmentation consistently. This method allows depth information to be provided to the network without explicit 3D annotation of the image. Experimental results show that this study's methodology performs better than existing methods on standard 3D pose benchmarks, and the proposed losses enable more consistent reconstruction from natural images.

• Journal of Cadastre & Land InformatiX
• /
• v.53 no.2
• /
• pp.79-92
• /
• 2023

With a growing interest in the Han River Skyline, which greatly influences Seoul's image, careful consideration of the skyline form has become crucial in the redevelopment plans for apartment complexes along the Han River. The Seoul Metropolitan City government has lifted the height limitations for apartments along the Hang River to cultivate a vibrant skyline. However, traditional skyline analysis often overlooks specific attributes, limiting the provision of precise guidelines for Seoul's unique skyline plans. Despite advancements in Digital Twin technology, only some tools effectively manage urban skylines with preferred shapes. Hence, this study aims to make a substantial contribution to the advancement of a Digital Twin 3D modeling program capable of effectively managing urban skylines. This is achieved through the utilisation of Conjoint Analysis, which assesses the importance of each attribute in determining the preferred skyline shape. Focusing on Apgujeong apartment complexes along the Han River currently undergoing redevelopment or planned for redevelopment, the study analyses the preferred skyline shape to propose standards for the Digital Twin 3D modeling program development. It also suggests that Conjoint Analysis can be beneficial in this process.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.18 no.2
• /
• pp.81-93
• /
• 2023

Under the digital economy, companies are facing a new business environment. Previous studies with a traditional perspectives on start-ups explained they are at a disadvantaged compared to large companies in mobilizing resources and building new relationships. However, recent researches on the digital economy and platform ecosystems have suggested that digital platforms can be an efficient means of overcoming the liabilities of smallness and newness for start-up companies. Through platforms start-ups can secure routes to mobilize resources, collect and analyze market information. Especially as various platorms are established within categorized industry or market, unique characteristics and market awareness regarding an individual startup company have been formed. Accordingly there is also an advantage that startup companies have umbrella effect by participating in the platform. From this point of view, this study describes the process by which a start-up company effectively utilizes a platform to overcome the liabilities of newness and smallness through the case the study of 'Genie the Bottle'. The results suggest that platforms play a crucial role for start-ups to secure awareness and legitimacy and grow the market in the beauty industry in which high consumer involvement is dominant.