eISSN : 3022-7011
ISSUER : KIPS
 
After the Korea Information Processing Society (KIPS) Transactions journal was founded in 1994, it was reorganized into the KIPS Transactions: Computer and Communication Systems(2287-5891/2734-049X ) and the KIPS Transactions: Software and Data Engi neering(2287-5905/2734-0503) in 2012. Through the KIPS official meeting on January 8th, 2024, the new KIPS Transaction journal was founded by integrating two KIPS Journals, KIPS Transactions: Computer and Communication Systems and KIPS Transactions: Software and Data Engineering. The new journal aims to realize social value and contribute to the development of South Korea’s science and technology with support from the lottery fund of the Ministry of Strategy and Finance and the science/technology promotion fund of the Ministry of Science and ICT. It is indexed in the Korea Science Academic Database, Korea Citation Index (KCI), and EBSCO.

HighlightsMore

Performance Evaluation and Consideration of Shadow Stack on RISC-V Architecture

Kang Ha Young  Han Go Won  Park Sung Hwan  Kwon Dong Hyun
RISC-V is an open-source instruction set architecture, used in various hardware implementations, and can be flexibly expanded to meet system requirements through the RV64I base instruction set and 16 standard extensions. Currently, the RISC-V archit...

A Study on the Evaluation Methods for Assessing the Understanding of Korean Culture by Generative AI Models

Son Ki Jun  Kim Seung Hyun
Recently, services utilizing large-scale language models (LLMs) such as GPT-4 and LLaMA have been released, garnering significant attention. These models can respond fluently to various user queries, but their insufficient training on Korean data ra...

Protection on WebAssembly JIT-Compiled Code with Randomized Memory Protection Key

Shin Chae Won  Jeong Yun Seo  Bae Myeong Jin  Kwon Dong Hyun
WebAssembly(Wasm) is a powerful platform that enables compiled code in various programming languages to be executed in web browser and in varied runtime environments. Specifically, for performance optimization, Wasm runtime provides Just-in-Time (JI...

A Study on Automatic Metrics for Korean Text Abstractive Summarization

Sehwi Yoon  Youhyun Shin
This study aims to analyze and validate automatic evaluation metrics for Korean abstractive summarization. The unique linguistic characteristics of each language require evaluation metrics designed for them, underscoring the importance of research f...

Latest Publication   (Vol. 14, No. 3, Mar.  2025)

Diffusion-based Audio-to-Visual Generation for High-Quality Bird Images
Adel Toleubekova  Joo Yong Shim  XinYu Piao  Jong-Kook Kim
Accurately identifying bird species from their vocalizations and generating corresponding bird images is still a challenging task due to limited training data and environmental noise in audio data. To address this limitation, this paper introduces a diffusion-based audio-to-image generation approach that satisfies both the need to accurately identify bird sounds and generate bird images. The main idea is to use a conditional diffusion model to handle the complexities of bird audio data, such as pitch variations and environmental noise while establishing a robust connection between the auditory and visual domains. This enables the model to generate high-quality bird images based on the given bird audio input. Plus, the proposed approach is integrated with deep audio processing to enhance its capabilities by meticulously aligning audio features with visual information and learning to map intricate acoustic patterns to corresponding visual representations. Experimental results demonstrate the effectiveness of the proposed approach in generating better images for bird classes compared to previous methods
The Transactions of the Korea Information Processing Society, Vol. 14, No. 3, pp. 135-142, Mar. 2025
https://doi.org/10.3745/TKIPS.2025.14.3.135
Audio-to-visual generation Diffusion models Image Generation Audio features Multi-modal generation
An Edge Detection Accelerator Using Otsu Method Based on Iterative Operation
Joowan Kang  Insu Jeong  Seungjun Jo  Byungin Moon
Edge detection is a key technology that significantly enhances the performance of vision-based systems by extracting object boundaries, and its importance is increasingly recognized in various real-time applications. To generate an edge image, selecting a threshold to distinguish between edge and non-edge regions is essential. The Otsu method is an effective technique for calculating the optimal threshold that divides an image into two classes. In particular, the Otsu method automatically selects a suitable threshold for each image, enabling class classification in environments where images change in real-time. However, implementing the Otsu method in hardware typically involves the use of logarithmic approximation techniques, w hich improve computational efficiency but increase hardw are resource consumption. This paper proposes an edge detection accelerator that enhances hardware resource efficiency by introducing a histogram-based iterative computation approach to the process of finding the maximum between-class variance in the Otsu method. Performance evaluation using 100 images from the ImageNet dataset revealed that the proposed method achieved an absolute error of 5.04 compared to the theoretical value in Otsu threshold calculation, and the binary classification using the computed threshold achieved an accuracy of 97.88%. Also, synthesis results show that the proposed hardware architecture utilizes 9,442 slice LUTs and 5,060 slice registers in the Otsu module. The proposed edge detection accelerator reduces hardware resource consumption while maintaining high computational accuracy compared to conventional logarithmic approximation techniques, making it suitable for various real-time image processing applications.
The Transactions of the Korea Information Processing Society, Vol. 14, No. 3, pp. 143-148, Mar. 2025
https://doi.org/10.3745/TKIPS.2025.14.3.143
Otsu method optimal threshold histogram Edge detection Hardware Accelerator
A Survey of Phishing Campaign Trends and the Classification of Detection Techniques
Ji-Hoon Park  Sang-Hoon Choi  Ki-Woong Park
Recent phishing attacks have demonstrated notable cost-effectiveness and efficiency through the use of phishing kits and AI technologies, resulting in a substantial rise in phishing efforts. Furthermore, phishing is evolving in complexity as perpetrators leverage psychological weaknesses, enhanced generation methods, and evasion tactics. Credentials obtained via phishing are frequently utilised for subsequent assaults, causing more harm. To safeguard users against phishing assaults, it is essential to examine how contemporary phishing strategies circumvent existing detection systems and to derive insights from the newest research developments. This study categorises phishing detection solutions into four types: URL and domain-based, web page component-based, visual similarity-based, and message content-based detection, while highlighting the challenges faced by each approach.
The Transactions of the Korea Information Processing Society, Vol. 14, No. 3, pp. 149-160, Mar. 2025
https://doi.org/10.3745/TKIPS.2025.14.3.149
Phishing Campaign Phishing Detection Machine Learning LLM feature extraction
A Study on Enhancing Zero-Shot Dense Retrieval Using Query and Hypothetical Document Embedding Combination
Lee Subin  Ho Bae
Dense retrieval transforms user queries and documents into high-dimensional embedding vectors and calculates their similarity in vector space, enabling effective contextual understanding. It is widely applied in search engines, recommendation systems, and legal domains, outperforming traditional keyword-based methods in handling complex queries. However, it struggles with short, ambiguous queries and adapting to new domains in zero-shot settings. Recent advancements, such as HyDE, leverage large language models (LLMs) to generate hypothetical documents to augment queries. Yet, relying solely on hypothetical embeddings may fail to fully capture user intent. This study proposes a novel framework that combines query and hypothetical document embeddings, dynamically adjusting their contributions based on query complexity. This approach enhances semantic richness and customization for more accurate search results. Experiments on MS MARCO and BEIR datasets show up to 8% performance improvement over HyDE and demonstrate superior nDCG@10 results with dynamic weights compared to fixed-weight methods. This framework offers a scalable, efficient solution applicable to various domains and complex query environments.
The Transactions of the Korea Information Processing Society, Vol. 14, No. 3, pp. 161-171, Mar. 2025
https://doi.org/10.3745/TKIPS.2025.14.3.161
Dense Retrieval Generative Model Embeddings Hypothetical Document Query
Detecting Money Market Macro Liquidity Event Using Column Sampling and Bagging
Jun Gyu Ahn  Hyoun-goo kang
Understanding liquidity in the financial market is important for raising funds for financial companies or corporations. Financing means raising funds necessary for corporate operations. In this study, we propose a liquidity event detection model in the financial market that considers the overall state of the financial market and the economy by utilizing macroeconomic variables. In order to understand the overall state of the financial market and the economy, macroeconomic variables from the United States and Korea are considered as model input variables. Machine learning models utilizing a large number of macroeconomic variables may exhibit model overfitting due to the curse of dimensionality. To alleviate this phenomenon, this study conducted a sampling-based column search and utilized bootstrap aggregation(bagging) to alleviate model variance and overfitting.
The Transactions of the Korea Information Processing Society, Vol. 14, No. 3, pp. 172-178, Mar. 2025
https://doi.org/10.3745/TKIPS.2025.14.3.172
Money Market Macro Liquidity Event Detection Variable selection Ensemble
An XR-Based Interface for Human-Robot Remote Control
Cho Su Been  Park Sung Jin  You Bum-Jae  Park Jung-Min
In the recent digital environment, humans are increasingly becoming active participants who directly interact with various technological elements. Recent research has explored the application of extended reality (XR) in solving robotics problems, aiming to reduce human risks through collaboration with robots. XR provides workers with a sense of realism and immersion, making it a valuable tool in remote robotic operations. This study presents the development of XR environment that replicates a remote workspace equipped with a robot, enabling workers to control the robot through the XR interface. The proposed XR space allows workers to monitor the status of both the XR space and the remote workspace, generate remote control commands to the robot, and observe the results of these commands in real time. Furthermore, an interface was developed to facilitate the seamless execution of remote control tasks and provide feedback within the XR environment.
The Transactions of the Korea Information Processing Society, Vol. 14, No. 3, pp. 179-185, Mar. 2025
https://doi.org/10.3745/TKIPS.2025.14.3.179
Extended Reality Human-Robot Interaction Remote Control
An Optimized Method with Cost and Performance to Supporting Multi-Cloud Service Deployment
Jinhyeok Jeon  Sumin Jeong  Joonseok Park  Keunhyuk Yeom
Multi-cloud is a technology that connects multiple cloud platforms to increase the stability of cloud service provision and ensure that cloud services are not dependent on the platform. To increase the stability of multi-cloud services, a multi-region strategy can be applied to deploy multiple virtual machines in various regions to utilize loose coupling between services. However, when individual functions of cloud services are distributed and operated, such as in a multi-region strategy, the cost of cloud services may increase and service performance may deteriorate due to different regional operational costs and virtual machine performance by region. Therefore, in this paper, we propose a method to determine the optimal placement location that satisfies the virtual machine operational cost and RTT (Round-Trip Time) requirements by applying NSGA-II (Non-dominated Sorting Genetic Algorithm-II), a genetic algorithm for multi-objective optimization. In addition, we conducted multi-objective optimization experiments on fourteen cloud regions operated by Google Cloud Platform and Amazon Web Service, which are actual cloud operating environments, by applying the proposed method. As a result of the experiment, it was confirmed that the execution time of the proposed multi-objective optimization algorithm was minimized to 1.53 seconds. The multi-objective optimization method presented in this paper is expected to be used as a base technology for virtual machine placement location decision-making to provide multi-cloud services.
The Transactions of the Korea Information Processing Society, Vol. 14, No. 3, pp. 186-193, Mar. 2025
https://doi.org/10.3745/TKIPS.2025.14.3.186
Multi-Cloud Multi-cloud Service Multi Region NSGA-II Multi objective optimization
Supporting Product Sales Strategy Through LLM-Based Consumer Review Analysis
He-sse Park  Dong-Gun Lee  Yeong-Seok Seo
The proliferation of online shopping has made product reviews a crucial factor in consumer purchasing decisions. Reviews provide essential information to consumers, such as product quality, price, and delivery, while also offering companies opportunities to refine their sales strategies and improve their products. However, the sheer volume of reviews poses challenges in efficiently analyzing and utilizing this data. To address this issue, this study proposes a model that automatically collects and systematically analyzes review data to enhance sales strategies. In this study, review data was effectively collected by combining web crawling with OCR technology. Particularly in web environments where crawling is restricted, OCR was used to convert review images into text, thereby expanding the scope of data collection. The collected review data was then analyzed using Large Language Model (LLM) like Chat GPT-4o, allowing for detailed classification of consumer opinions into various aspects such as quality, price, and delivery. The analysis results were utilized to identify key feedback elements and to design a user-friendly app UI that facilitates comparisons between a company's products and those of competitors. This UI includes features for data visualization, real-time search, and personalized analytical insights, enabling both consumers and companies to easily understand and utilize review data. This study introduces a novel approach to understanding consumer behavior by integrating IT technology with psychological analysis. It empirically demonstrates not only the tendency to justify purchasing decisions for high-priced products but also the possibility that a similar pattern can be observed in low-priced products, highlighting the importance of marketing strategies that reflect consumer psychological characteristics. The findings highlight the importance of developing marketing strategies that account for consumers’ psychological traits. Ultimately, this research showcases the potential for more effective, data-driven marketing strategies.
The Transactions of the Korea Information Processing Society, Vol. 14, No. 3, pp. 194-202, Mar. 2025
https://doi.org/10.3745/TKIPS.2025.14.3.194
LLM OCR Data Engineering Consumer Psychology Psychology Web Scraping Web crawling Artificial intelligence Consumer Behavior
SViT: A Novel Multimodal Learning Approach for Ship Distance Estimation via Time-Series Data Visualization
Sun Choi  Jeongmin Choi  Hyunbae Chang  Jhonghyun An
Accurately detecting and estimating ship distances is crucial for maritime safety, resource management, and preventing illegal activities. However, traditional Automatic Identification Systems (AIS) are highly vulnerable to signal interference and tampering, making them unreliable. To address this issue, we propose SViT, a novel multimodal learning approach. SViT converts time-series sensor data into 2D image representations, allowing Vision Transformer (ViT) models to analyze them effectively. It further enhances computational efficiency and noise reduction through hierarchical sensor feature selection. To ensure stable ship movement predictions, we introduce a novel loss function combining MSE, Smoothness Loss, and Gradient Loss. Experimental results show that SViT outperforms LSTM-based models in training speed and predictive stability. This study presents a robust framework for utilizing complex multisensor data, with potential applications in security, surveillance, and maintenance.
The Transactions of the Korea Information Processing Society, Vol. 14, No. 3, pp. 203-213, Mar. 2025
https://doi.org/10.3745/TKIPS.2025.14.3.203
Underwater Ship Estimation multivariate time series forecasting Multimodal Vision Transformer (ViT)

Forms

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)

Old Journals

Indexing

All publications of TKIPS is indexed in DOI, EBSCO, Google Scholar, Crossref, and CrossCheck.

TKIPS is also selected as the Journal for Accreditation by NRF (National Research Foundation of Korea).

Related Journals