• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.603-608
• /
• 2023

본 연구는 한국어 Vision-Language Pre-training 모델 학습을 위한 대규모 시각-언어 멀티모달 데이터셋 구축에 대한 필요성을 연구한다. 현재, 한국어 시각-언어 멀티모달 데이터셋은 부족하며, 양질의 데이터 획득이 어려운 상황이다. 따라서, 본 연구에서는 기계 번역을 활용하여 외국어(영문) 시각-언어 데이터를 한국어로 번역하고 이를 기반으로 생성형 AI를 활용한 데이터셋 구축 방법론을 제안한다. 우리는 다양한 캡션 생성 방법 중, ChatGPT를 활용하여 자연스럽고 고품질의 한국어 캡션을 자동으로 생성하기 위한 새로운 방법을 제안한다. 이를 통해 기존의 기계 번역 방법보다 더 나은 캡션 품질을 보장할 수 있으며, 여러가지 번역 결과를 앙상블하여 멀티모달 데이터셋을 효과적으로 구축하는데 활용한다. 뿐만 아니라, 본 연구에서는 의미론적 유사도 기반 평가 방식인 캡션 투영 일치도(Caption Projection Consistency) 소개하고, 다양한 번역 시스템 간의 영-한 캡션 투영 성능을 비교하며 이를 평가하는 기준을 제시한다. 최종적으로, 본 연구는 ChatGPT를 이용한 한국어 멀티모달 이미지-텍스트 멀티모달 데이터셋 구축을 위한 새로운 방법론을 제시하며, 대표적인 기계 번역기들보다 우수한 영한 캡션 투영 성능을 증명한다. 이를 통해, 우리의 연구는 부족한 High-Quality 한국어 데이터 셋을 자동으로 대량 구축할 수 있는 방향을 보여주며, 이 방법을 통해 딥러닝 기반 한국어 Vision-Language Pre-training 모델의 성능 향상에 기여할 것으로 기대한다.

• Journal of Internet Computing and Services
• /
• v.24 no.5
• /
• pp.51-66
• /
• 2023

In this paper, we propose a new type of process discovery framework, which is named as control-path-driven process group discovery framework, to be used for process mining and process reengineering in supporting life-cycle management of business process models. In addition, we develop a process mining system based on the proposed framework and perform experimental verification through it. The process execution event logs applied to the experimental effectiveness and verification are specially defined as Process BIG-Logs, and we use it as the input datasets for the proposed discovery framework. As an eventual goal of this paper, we design and implement a control path-driven process group discovery algorithm and framework that is improved from the ρ-algorithm, and we try to verify the functional correctness of the proposed algorithm and framework by using the implemented system with a BIG-Log dataset. Note that all the process mining algorithm, framework, and system developed in this paper are based on the structural information control net process modeling methodology.

• KIPS Transactions on Computer and Communication Systems
• /
• v.12 no.12
• /
• pp.357-362
• /
• 2023

Bridges are important transportation infrastructure, but they are subject to damage and cracking due to various environmental factors and constant traffic loads, which accelerate their aging. With many bridges now older than their original construction, there is a need for systems to ensure safety and diagnose deterioration. Bridges are already utilizing structural health monitoring (SHM) technology to monitor the condition of bridges in real time or periodically. Along with this technology, the development of intelligent bridge monitoring technology utilizing artificial intelligence and Internet of Things technology is underway. In this paper, we study an edge system technique for predicting bridge safety using fast Fourier transform and dimensionality reduction algorithm for maintenance of aging bridges. In particular, unlike previous studies, we investigate whether it is possible to form a dataset using sensor data collected from actual bridges and check the safety of bridges.

• Advanced Industrial SCIence
• /
• v.2 no.4
• /
• pp.11-16
• /
• 2023

The subway is a means of public transportation that plays an important role in the transportation system of modern cities. However, congestion often occurs due to sudden breakdowns and system outages, causing inconvenience. Therefore, in this paper, we conducted a study on failure prediction and prevention using machine learning to efficiently operate the subway system. Using UC Irvine's MetroPT-3 dataset, we built a subway breakdown prediction model using logistic regression. The model predicted the non-failure state with a high accuracy of 0.991. However, precision and recall are relatively low, suggesting the possibility of error in failure prediction. The ROC_AUC value is 0.901, indicating that the model can classify better than random guessing. The constructed model is useful for stable operation of the subway system, but additional research is needed to improve performance. Therefore, in the future, if there is a lot of learning data and the data is well purified, failure can be prevented by pre-inspection through prediction.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.12
• /
• pp.525-534
• /
• 2023

Gas leak detection system is a key to minimize the loss of life due to the explosiveness and toxicity of gas. Most of the leak detection systems detect by gas sensors or thermal imaging cameras. To improve the performance of gas leak detection system using single-modal methods, the paper propose multimodal approach to gas sensor data and thermal camera data in developing a gas type identification model. MultimodalGasData, a multimodal open-dataset, is used to compare the performance of the four models developed through multimodal approach to gas sensors and thermal cameras with existing models. As a result, 1D CNN and GasNet models show the highest performance of 96.3% and 96.4%. The performance of the combined early fusion model of 1D CNN and GasNet reached 99.3%, 3.3% higher than the existing model. We hoped that further damage caused by gas leaks can be minimized through the gas leak detection system proposed in the study.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.369-377
• /
• 2023

In the industrial manufacturing sector, quality control is pivotal for minimizing defect rates; inadequate management can result in additional costs and production delays. This study underscores the significance of detecting texture defects in manufactured goods and proposes a more precise defect detection technique. While the DFR(Deep Feature Reconstruction) model adopted an approach based on feature map amalgamation and reconstruction, it had inherent limitations. Consequently, we incorporated a new loss function using statistical methodologies, integrated a skip connection structure, and conducted parameter tuning to overcome constraints. When this enhanced model was applied to the texture category of the MVTec-AD dataset, it recorded a 2.3% higher Defect Segmentation AUC compared to previous methods, and the overall defect detection performance was improved. These findings attest to the significant contribution of the proposed method in defect detection through the reconstruction of feature map combinations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.4
• /
• pp.826-842
• /
• 2024

As 5G and AI continue to develop, there has been a significant surge in the healthcare industry. The COVID-19 pandemic has posed immense challenges to the global health system. This study proposes an FL-supported edge computing model based on federated learning (FL) for predicting clinical outcomes of COVID-19 patients during hospitalization. The model aims to address the challenges posed by the pandemic, such as the need for sophisticated predictive models, privacy concerns, and the non-IID nature of COVID-19 data. The model utilizes the FATE framework, known for its privacy-preserving technologies, to enhance predictive precision while ensuring data privacy and effectively managing data heterogeneity. The model's ability to generalize across diverse datasets and its adaptability in real-world clinical settings are highlighted by the use of SHAP values, which streamline the training process by identifying influential features, thus reducing computational overhead without compromising predictive precision. The study demonstrates that the proposed model achieves comparable precision to specific machine learning models when dataset sizes are identical and surpasses traditional models when larger training data volumes are employed. The model's performance is further improved when trained on datasets from diverse nodes, leading to superior generalization and overall performance, especially in scenarios with insufficient node features. The integration of FL with edge computing contributes significantly to the reliable prediction of COVID-19 patient outcomes with greater privacy. The research contributes to healthcare technology by providing a practical solution for early intervention and personalized treatment plans, leading to improved patient outcomes and efficient resource allocation during public health crises.

• Smart Media Journal
• /
• v.12 no.11
• /
• pp.1-9
• /
• 2023

Stress is a significant issue in modern society, often triggered by external or internal factors that are difficult to manage. When high stress persists over a long term, it can develop into a chronic condition, negatively impacting health and overall well-being. However, it is challenging for individuals experiencing chronic stress to recognize their condition, making early detection and management crucial. Using biosignals measured from wearable devices to detect stress could lead to more effective management. However, there are two main problems with using biosignals: first, manually extracting features from these signals can introduce bias, and second, the performance of classification models can vary greatly depending on the subject of the experiment. This paper proposes a model that reduces bias using convo utional autoencoders, which can represent the key features of data, and enhances generalizability by employing soft voting, a method of ensemble learning, to minimize performance variability. To verify the generalization performance of the model, we evaluate it using LOSO cross-validation method. The model proposed in this paper has demonstrated superior accuracy compared to previous studies using the WESAD dataset.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.34 no.3
• /
• pp.403-416
• /
• 2024

Malware is being commercialized and sold on the black market, primarily driven by financial incentives. With the increasing demand driven by these sales, the scope of attacks via malware has expanded. In response, there has been a surge in research efforts leveraging artificial intelligence for detection and classification. However, adversaries are integrating various anti-analysis techniques into their malware to thwart analytical efforts. In this study, we introduce the "Malware Analysis with Dynamic Extraction (MADE)" framework, a hybrid binary analysis tool devised to procure datasets from advanced malware incorporating Anti-Analysis techniques. The MADE framework has the proficiency to autonomously execute dynamic analysis on binaries, encompassing those laden with Anti-VM and Anti-Debugging defenses. Experimental results substantiate that the MADE framework can effectively circumvent over 90% of diverse malware implementations using Anti-Analysis techniques and can adeptly extract relevant datasets.

• Korean Chemical Engineering Research
• /
• v.61 no.4
• /
• pp.523-541
• /
• 2023

As accessibility to 3D printers increases, there is a growing frequency of exposure to chemicals associated with 3D printing. However, research on the toxicity and harmfulness of chemicals generated by 3D printing is insufficient, and the performance of toxicity prediction using in silico techniques is limited due to missing molecular structure data. In this study, quantitative structure-activity relationship (QSAR) model based on data-centric AI approach was developed to predict the toxicity of new 3D printing materials by imputing missing values in molecular descriptors. First, MissForest algorithm was utilized to impute missing values in molecular descriptors of hazardous 3D printing materials. Then, based on four different machine learning models (decision tree, random forest, XGBoost, SVM), a machine learning (ML)-based QSAR model was developed to predict the bioconcentration factor (Log BCF), octanol-air partition coefficient (Log Koa), and partition coefficient (Log P). Furthermore, the reliability of the data-centric QSAR model was validated through the Tree-SHAP (SHapley Additive exPlanations) method, which is one of explainable artificial intelligence (XAI) techniques. The proposed imputation method based on the MissForest enlarged approximately 2.5 times more molecular structure data compared to the existing data. Based on the imputed dataset of molecular descriptor, the developed data-centric QSAR model achieved approximately 73%, 76% and 92% of prediction performance for Log BCF, Log Koa, and Log P, respectively. Lastly, Tree-SHAP analysis demonstrated that the data-centric-based QSAR model achieved high prediction performance for toxicity information by identifying key molecular descriptors highly correlated with toxicity indices. Therefore, the proposed QSAR model based on the data-centric XAI approach can be extended to predict the toxicity of potential pollutants in emerging printing chemicals, chemical process, semiconductor or display process.