Search Results (66)

The purpose of this study is to explore and analyse user perceptions of peri-rural landscapes in the Seoul metropolitan region, using Icheon City as a case study. While the multifunctionality of peri-rural areas—providing ecological, cultural, and socioeconomic benefits—is increasingly recognised, the perceptual and experiential dimensions remain underexplored in South Korea. To address this gap, 10,578 Naver Blog posts were collected and refined, resulting in 8078 valid entries. Methodologically, this study introduces an innovative approach by integrating centrality analysis with latent Dirichlet allocation (LDA) topic modeling of user-generated content, supported by a bespoke dictionary of 170 local landscape resources. This combined framework allows simultaneous examination of structural associations and thematic narratives within user perceptions. The results indicate that resources such as Seolbong Urban Park, Seolbong Mountain, and the Cornus Fruit (sansuyu) Villages function as symbolic hubs in the perceptual network, while thematic clusters capture multi-dimensional concerns spanning leisure, ecology, culture, suburbanization, and real estate. Synthesised together, these findings demonstrate that user perceptions construct peri-rural landscapes not as isolated sites, but as spatially cohesive and thematically interconnected systems that mediate between urban and rural domains. Overall, this study contributes to metropolitan planning discourse by highlighting perceptual dimensions alongside functional and ecological dimensions. It shows that users cognitively construct peri-rural landscapes as systems that are both spatially cohesive and thematically interconnected, and that function as spaces that link urban and rural areas. Crucially, this study provides a replicable framework for using user-generated content to inform the planning and management of peri-rural landscapes in metropolitan areas. Full article

Most speech separation techniques require knowing the number of talkers mixed in an input, which is not always available in real situations. To address this problem, we present a novel speech separation method that automatically finds the number of talkers in input mixture recordings. The proposed method extracts the voices of individual talkers one by one in a deflationary manner and stops the extraction sequence when a predefined termination criterion is satisfied. The backbone separation model is built based on the transformer architecture with permutation-invariant training to avoid ambiguity in identifying talkers at the output. The experimental results on the Libri5Mix and Libri10Mix datasets show that the proposed method without the number of talkers as input significantly outperforms state-of-the-art models that are provided with the number of talkers. Full article

In this paper, we propose an active robotic 3D reconstruction methodology for achieving full object 3D reconstruction. Existing robotic 3D reconstruction approaches often struggle to cover the entire view space of the object or reconstruct occluded regions, such as the bottom or back side. To address these limitations, we introduce a two-stage robotic active 3D reconstruction pipeline, named See-Then-Grasp (STG), that employs a robot manipulator for direct interaction with the object. The manipulator moves toward the points with the highest uncertainty, ensuring efficient data acquisition and rapid reconstruction. Our method expands the view space of the object to include the entire perspective, including occluded areas, making the previous fixed view candidate approach time-consuming for identifying uncertain regions. To overcome this, we propose a gradient-based next best view pose optimization method that efficiently identifies uncertain regions, enabling faster and more effective reconstruction. Our method optimizes the camera pose based on an uncertainty function, allowing it to identify the most uncertain regions in a short time. Through experiments with synthetic objects, we demonstrate that our approach effectively addresses the next best view selection problem, achieving significant improvements in computational efficiency while maintaining high-quality 3D reconstruction. Furthermore, we validate our method on a real robot, showing that it enables full 3D reconstruction of real-world objects. Full article

The global VTuber market is experiencing rapid growth, with VTubers extending beyond mere content creators to be utilized in various fields such as social interaction, public relations, and health. VTubers have the potential to expand the existing content market and contribute to increasing economic and public value. This study aims to investigate the perception of VTubers in South Korea and to provide insights that can contribute to the global activation of the VTuber entertainment industry. For this purpose, unstructured data on VTubers from the past three years, during which interest in VTubers has significantly grown in South Korea, was collected. A total of 57,891 samples were gathered from Naver, Daum, and Google, of which 50 highly relevant data points between VTubers and users were selected for analysis. First, key terms such as ‘Broadcast’, ‘YouTube’, ‘Live’, ‘Game’, ‘Youtuber’, ‘Japan’, ‘Character’, ‘Video’, ‘Sing’, ‘Virtual’, ‘Woowakgood’, ‘Fan’, ‘Idol’, ‘Korea’, ‘Twitch’, ‘IsegyeIdol’, ‘Communication’, ‘Worldview’, ‘VTuberIndustry’, ‘Contents’, ‘AfricaTV’, ‘Nijisanji’, and ‘Streamer’ were extracted. Second, CONCOR analysis revealed four clusters: ‘Famous VTubers’, ‘Features of VTubers’, ‘VTuber Industry’, and ‘VTuber Platforms’. Based on these findings, the study offers various academic and practical implications regarding VTubers in South Korea and explores the potential for global growth in the VTuber industry. Full article

An aspect-based sentiment analysis (ABSA) aims to perform a fine-grained analysis of text to identify sentiments and opinions associated with specific aspects. Recently, transformers and large language models have demonstrated exceptional performance in detecting aspects and determining their associated sentiments within text. However, understanding the decision-making processes of transformers remains a significant challenge, as they often operate as black-box models, making it difficult to interpret how they arrive at specific predictions. In this article, we examine the performance of various transformers on ABSA and we employ explainability techniques to illustrate their inner decision-making processes. Firstly, we fine-tune several pre-trained transformers, including BERT, RoBERTa, DistilBERT, and XLNet, on an extensive set of data composed of MAMS, SemEval, and Naver datasets. These datasets consist of over 16,100 complex sentences, each containing a couple of aspects and corresponding polarities. The models were fine-tuned using optimal hyperparameters and RoBERTa achieved the highest performance, reporting 89.16% accuracy on MAMS and SemEval and 97.62% on Naver. We implemented five explainability techniques, LIME, SHAP, attention weight visualization, integrated gradients, and Grad-CAM, to illustrate how transformers make predictions and highlight influential words. These techniques can reveal how models use specific words and contextual information to make sentiment predictions, which can improve performance, address biases, and enhance model efficiency and robustness. These also point out directions for further focus on the analysis of models’ bias in combination with explainability methods, ensuring that explainability highlights potential biases in predictions. Full article

Many older adults are increasingly embracing digital technology in the Republic of Korea. This study investigated the relationship between the digital skills of Korean older adults and their perceived health status and digital technology application for health promotion. This mixed-method study comprised a community survey of 434 older adults aged ≥65 in two cities in South Korea, followed by focus group interviews. Five types of digital skills, ‘operational internet skills’, ‘information navigation skills’, ‘social skills’, ‘creative skills’, and ‘mobile skills’, were measured using the LSE digital skill measurement instrument. Multivariable analysis identified the influence of digital skills on health-related outcomes. Among them, ‘social skills’ associated positively with self-rated health (β 0.37, 95%CI 0.08, 0.65). ‘Information navigation skills’ contributed positively to the use of digital technology and the internet for a healthy lifestyle in terms of improving eating habits (β 0.43, 95%CI 0.09, 0.77), accessing healthcare (β 0.53, 95%CI 0.21, 0.85), and accessing long-term care services (β 0.45, 95%CI 0.11, 0.79). Thematic analysis revealed that the study participants use Korean language-based resources such as Naver and Kakao Talk for social connection to promote a healthy lifestyle. This study concludes that encouraging initial and sustained use of the internet and enhancing digital skills among Korean older adults can promote active and healthy aging. Full article

With the introduction of deep learning, a significant amount of research has been conducted in the field of computer vision in the past decade. In particular, research on object detection (OD) continues to progress rapidly. However, despite these advances, some limitations need to be overcome to enable real-world applications of deep learning-based OD models. One such limitation is inaccurate OD when image quality is poor or a target object is small. The performance degradation phenomenon for small objects is similar to the fundamental limitations of an OD model, such as the constraint of the receptive field, which is a difficult problem to solve using only an OD model. Therefore, OD performance can be hindered by low image quality or small target objects. To address this issue, this study investigates the compatibility of super-resolution (SR) and OD techniques to improve detection, particularly for small objects. We analyze the combination of SR and OD models, classifying them based on architectural characteristics. The experimental results show a substantial improvement when integrating OD detectors with SR models. Overall, it was demonstrated that, when the evaluation metrics (PSNR, SSIM) of the SR models are high, the performance in OD is correspondingly high as well. Especially, evaluations on the MS COCO dataset reveal that the enhancement rate for small objects is 9.4% higher compared to all objects. This work provides an analysis of SR and OD model compatibility, demonstrating the potential benefits of their synergistic combination. The experimental code can be found on our GitHub repository. Full article

This study aimed to assess the potential contributions of the Myers–Briggs Type Indicator (MBTI) to the sustainable growth of communities by conducting a comprehensive analysis of social perceptions of the MBTI in South Korea through big data analysis. The investigation encompasses three primary stages: data collection, preprocessing, and analysis, involving text mining, network analysis, CONCOR analysis, and sentiment analysis. A total of 31,308 text data pieces (13.73 MB) from various sources, including news, blogs, and other sections of Naver and Google, over the past three years, were collected and analyzed using the keyword “MBTI”. Tools, such as Textom SV, UCINET, and NetDraw, were employed for data collection and analysis. The study’s key findings include the identification, through term frequency (TF) and TF-inverse document frequency analyses, of top-ranking terms, such as 16Types, 4Indicators, Test, Myself, OthersMBTI, Situation, and Contents. The CONCOR analysis further revealed six clusters, encompassing themes like interest in MBTI personality tests, application of 16 types in daily life, MZ’s MBTI consumption patterns, trending of MBTI characters, extension to K-Test, and professional use of MBTI. Moreover, sentiment analysis indicated that 68.5% of individuals in South Korea expressed a positive sentiment towards MBTI, while 31.5% conveyed a negative sentiment. The specific emotions identified included liking (Good Feeling), disgust, and interest, in order of prominence. In light of these findings, this study delineates a spectrum of perceptions regarding MBTI in South Korea, encompassing both positive interests and negative concerns. To ensure the responsible use of MBTI, it is imperative to implement reliable scientific testing and education, mitigate the potential harm of stereotyping, and reshape social perceptions surrounding MBTI usage. Only through these measures can MBTI genuinely contribute to the sustainable growth of communities without being confined to limiting stereotypes. Full article

Vietnamese is an under-resourced language. The requirement for a large-scale and high-quality Vietnamese speech corpus increases on demand. We introduce a new large-scale Vietnamese speech corpus with 100.5 h collected from various audio sources in the Internet. The raw collected audio was processed to obtain clean speech. Transcription of the clean speech was made manually. The new corpus was analyzed in terms of gender, topic and regional dialect. Results shows that the new corpus has good diversity of genders, topics and regional dialects. We also evaluated the new corpus using state-of-the-art automatic speech recognition models like LAS and Speech-Transformer for multiple scenarios. This is the first time that these models have been applied to Vietnamese speech recognition and obtained reasonable results. Simulation results showed that the new corpus would be a good dataset for the Vietnamese ASR tasks because it reflected correctly difficulties in recognizing speech from different dialects and topic domains. Full article

The rapid evolution of 5G technology, while offering substantial benefits, concurrently presents complex cybersecurity challenges. Current cybersecurity systems often fall short in addressing challenges such as the lack of realism of the 5G network, the limited scope of attack scenarios, the absence of countermeasures, the lack of reproducible, and open-sourced cybersecurity training environments. Addressing these challenges necessitates innovative cybersecurity training systems, referred to as “cyber ranges”. In response to filling these gaps, we propose the Cyber5Gym, an integrated cyber range that enhances the automation of virtualized cybersecurity training in 5G networks with cloud-based deployment. Our framework leverages open-source tools (i) Open5GS and UERANSIM for realistic emulation of 5G networks, (ii) Docker for efficient virtualization of the training infrastructure, (iii) 5Greply for emulating attack scenarios, and (iv) Shell scripts for automating complex training operations. This integration facilitates a dynamic learning environment where cybersecurity professionals can engage in real-time attack and countermeasure exercises, thus significantly improving their readiness against 5G-specific cyber threats. We evaluated it by deploying our framework on Naver Cloud with 20 trainees, each accessing an emulated 5G network and managing 100 user equipments (UEs), emulating three distinct attack scenarios (SMC-Reply, DoS, and DDoS attacks), and exercising countermeasures, to demonstrate the cybersecurity training. We assessed the effectiveness of our framework through specific metrics such as successfully establishing the 5G network for all trainees, accurate execution of attack scenarios, and their countermeasure implementation via centralized control of the master using automated shell scripts. The open-source foundation of our framework ensures replicability and adaptability, addressing a critical gap in current cybersecurity training methodologies and contributing significantly to the resilience and security of 5G infrastructures. Full article

Pedestrian detection is a critical task for safety-critical systems, but detecting pedestrians is challenging in low-light and adverse weather conditions. Thermal images can be used to improve robustness by providing complementary information to RGB images. Previous studies have shown that multi-modal feature fusion using convolution operation can be effective, but such methods rely solely on local feature correlations, which can degrade the performance capabilities. To address this issue, we propose an attention-based novel fusion network, referred to as INSANet (INtra-INter Spectral Attention Network), that captures global intra- and inter-information. It consists of intra- and inter-spectral attention blocks that allow the model to learn mutual spectral relationships. Additionally, we identified an imbalance in the multispectral dataset caused by several factors and designed an augmentation strategy that mitigates concentrated distributions and enables the model to learn the diverse locations of pedestrians. Extensive experiments demonstrate the effectiveness of the proposed methods, which achieve state-of-the-art performance on the KAIST dataset and LLVIP dataset. Finally, we conduct a regional performance evaluation to demonstrate the effectiveness of our proposed network in various regions. Full article

Age is an essential piece of demographic information for social profiling, as different social and behavioral characteristics are age-related. To acquire age information, most of the previously conducted social profiling studies have predicted age information. However, age predictions in social profiling have been very limited, because it is difficult or impossible to obtain age information from social media. Moreover, age-prediction results have rarely been used to study human dynamics. In these circumstances, this study focused on naver.com, a nationwide social media website in Korea. Although the social profiles of news commenters on naver.com can be analyzed and used, the age information is incomplete (i.e., partially open to the public) owing to anonymity and privacy protection policies. Therefore, no prior research has used naver.com for age predictions or subsequent analyses based on the predicted age information. To address this research gap, this study proposes a method that uses a machine learning approach to predict the age information of anonymous commenters on unlabeled (i.e., with age information hidden) news articles on naver.com. Furthermore, the predicted age information was fused with the section information of the collected news articles, and fuzzy differences between age groups were analyzed for topics of interest, using the proposed correlation–similarity matrix and fuzzy sets of age differences. Thus, differentiated from the previous social profiling studies, this study expands the literature on social profiling and human dynamics studies. Consequently, it revealed differences between age groups from anonymous and incomplete Korean social media that can help in understanding age differences and ease related intergenerational conflicts to help reach a sustainable South Korea. Full article

This paper presents a model for generating expressive robot motions based on human expressive movements. The proposed data-driven approach combines variational autoencoders and a generative adversarial network framework to extract the essential features of human expressive motion and generate expressive robot motion accordingly. The primary objective was to transfer the underlying expressive features from human to robot motion. The input to the model consists of the robot task defined by the robot’s linear velocities and angular velocities and the expressive data defined by the movement of a human body part, represented by the acceleration and angular velocity. The experimental results show that the model can effectively recognize and transfer expressive cues to the robot, producing new movements that incorporate the expressive qualities derived from the human input. Furthermore, the generated motions exhibited variability with different human inputs, highlighting the ability of the model to produce diverse outputs. Full article

Online streaming contents are creating greater service uncertainty, as consumers need to experience such contents before making a decision to continue to purchase them. Few studies have investigated the interaction between eWOM (online word-of-mouth) and online streaming content service characteristics with regard to the performance of online streaming contents and explained how this interaction can promote the role of service characteristics in service performance outcomes or remedy service uncertainty attributable to these characteristics. Thus, in order to test the interaction effects, this paper examines the moderating effects of service (webtoon) characteristics (i.e., author experience, genre (drama or fantasy), completion, transfer to paid service, and publication time (Wednesday)) on the relationship between eWOM and certain online streaming contents’ service performance measures; in this case, the publication period and content gamification. Based on scrawled data from 154 webtoons published on Naver Webtoon, a multivariate regression analysis with interaction terms showed that author experience and genre interact with the number of reviews to affect gamification. The transfer to a paid service interacts crucially with review ratings and the number of reviews to influence both the publication period and gamification. Online streaming content completion and publication times are factors that interact with review ratings and thus affect the publication period. Service providers need to cope with service uncertainties when attempting to further their online streaming content service by considering the service characteristics as well as customers’ responses through eWOM. Full article

It is presently being questioned whether the metaverse is mere hype or the next transformative vision. It should be examined how the issues associated with the metaverse are being dealt with socially, and accordingly, how the public’s interest has changed. This paper aims to explore the metaverse’s issues and its rapidly changing trends in South Korea during the pandemic period of 2020–2021, in which the term was very widely used. This study conducted a semantic network analysis using online news big data with a text mining approach to analyze online news content from search engine portals such as Naver, Daum, and Google. TF-IDF, degree centrality, word cloud visualization, and CONCOR analysis were used within the Textom and UCINET6 programs. This research provides valuable insights into how the metaverse is being embraced and discussed within the South Korean context, shedding light on its potential impact and the changing dynamics of public engagement. The results showed that the topics of the public’s interests in the metaverse varied in the year 2021 as compared to 2020, and the opportunities and concerns revolving around it are referred to at the same time. The study found that there were significant changes in the subjects that gained public interest in the metaverse between 2020 and 2021. In 2020, the term “Metaverse” became popular in the news due to its increasing popularity in the world of virtual online gaming, particularly among younger populations. This was further accelerated by the COVID-19 pandemic restrictions, resulting in a rise in virtual experiences. In contrast, the year 2021 was marked as the time when the concept of the metaverse gained widespread recognition and established itself as a platform for business and financial opportunities, suggesting the growing interest of older generations in the metaverse. Full article