• Journal of the Korea Convergence Society
• /
• v.12 no.11
• /
• pp.35-43
• /
• 2021

Recently, there has been a growing interest in zero-shot cross-lingual transfer, which leverages cross-lingual language models (CLLMs) to perform downstream tasks that are not trained in a specific language. In this paper, we point out the limitations of the data-centric aspect of quality estimation (QE), and perform zero-shot cross-lingual transfer even in environments where it is difficult to construct QE data. Few studies have dealt with zero-shots in QE, and after fine-tuning the English-German QE dataset, we perform zero-shot transfer leveraging CLLMs. We conduct comparative analysis between various CLLMs. We also perform zero-shot transfer on language pairs with different sized resources and analyze results based on the linguistic characteristics of each language. Experimental results showed the highest performance in multilingual BART and multillingual BERT, and we induced QE to be performed even when QE learning for a specific language pair was not performed at all.

• Journal of Internet Computing and Services
• /
• v.24 no.4
• /
• pp.25-36
• /
• 2023

Today, as AI (Artificial Intelligence) technology is introduced in various fields, including security, the development of technology is accelerating. However, with the development of AI technology, attack techniques that cleverly bypass malicious behavior detection are also developing. In the classification process of AI models, an Adversarial attack has emerged that induces misclassification and a decrease in reliability through fine adjustment of input values. The attacks that will appear in the future are not new attacks created by an attacker but rather a method of avoiding the detection system by slightly modifying existing attacks, such as Adversarial attacks. Developing a robust model that can respond to these malware variants is necessary. In this paper, we propose two methods of generating Adversarial attacks as efficient Adversarial attack generation techniques for improving Robustness in AI models. The proposed technique is the XAI-based attack technique using the XAI technique and the Reference based attack through the model's decision boundary search. After that, a classification model was constructed through a malicious code dataset to compare performance with the PGD attack, one of the existing Adversarial attacks. In terms of generation speed, XAI-based attack, and reference-based attack take 0.35 seconds and 0.47 seconds, respectively, compared to the existing PGD attack, which takes 20 minutes, showing a very high speed, especially in the case of reference-based attack, 97.7%, which is higher than the existing PGD attack's generation rate of 75.5%. Therefore, the proposed technique enables more efficient Adversarial attacks and is expected to contribute to research to build a robust AI model in the future.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.38 no.4
• /
• pp.196-203
• /
• 2022

Purpose: This study aimed to evaluate the accuracy and clinical usability of an identification model using deep learning for 79 dental implant types. Materials and Methods: A total of 45396 implant fixture images were collected through panoramic radiographs of patients who received implant treatment from 2001 to 2020 at 30 dental clinics. The collected implant images were 79 types from 18 manufacturers. EfficientNet and Meta Pseudo Labels algorithms were used. For EfficientNet, EfficientNet-B0 and EfficientNet-B4 were used as submodels. For Meta Pseudo Labels, two models were applied according to the widen factor. Top 1 accuracy was measured for EfficientNet and top 1 and top 5 accuracy for Meta Pseudo Labels were measured. Results: EfficientNet-B0 and EfficientNet-B4 showed top 1 accuracy of 89.4. Meta Pseudo Labels 1 showed top 1 accuracy of 87.96, and Meta pseudo labels 2 with increased widen factor showed 88.35. In Top5 Accuracy, the score of Meta Pseudo Labels 1 was 97.90, which was 0.11% higher than 97.79 of Meta Pseudo Labels 2. Conclusion: All four deep learning algorithms used for implant identification in this study showed close to 90% accuracy. In order to increase the clinical applicability of deep learning for implant identification, it will be necessary to collect a wider amount of data and develop a fine-tuned algorithm for implant identification.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.4
• /
• pp.159-172
• /
• 2023

In general, social problem-solving research aims to create important social value by offering meaningful answers to various social pending issues using scientific technologies. Not surprisingly, however, although numerous and extensive research attempts have been made to alleviate the social problems and issues in nation-wide, we still have many important social challenges and works to be done. In order to facilitate the entire process of the social problem-solving research and maximize its efficacy, it is vital to clearly identify and grasp the important and pressing problems to be focused upon. It is understandable for the problem discovery step to be drastically improved if current social issues can be automatically identified from existing R&D resources such as technical reports and articles. This paper introduces a comprehensive dataset which is essential to build a machine learning model for automatically detecting the social problems and solutions in various national research reports. Initially, we collected a total of 700 research reports regarding social problems and issues. Through intensive annotation process, we built totally 24,022 sentences each of which possesses its own category or label closely related to social problem-solving such as problems, purposes, solutions, effects and so on. Furthermore, we implemented four sentence classification models based on various neural language models and conducted a series of performance experiments using our dataset. As a result of the experiment, the model fine-tuned to the KLUE-BERT pre-trained language model showed the best performance with an accuracy of 75.853% and an F1 score of 63.503%.

• Composites Research
• /
• v.36 no.3
• /
• pp.205-210
• /
• 2023

In this study, we developed a transfer learning framework based on homogenization data for efficient prediction of the effective mechanical properties and thermal conductivity of cellular foam structures. Mean-field homogenization (MFH) based on the Eshelby's tensor allows for efficient prediction of properties in porous structures including ellipsoidal inclusions, but accurately predicting the properties of cellular foam structures is challenging. On the other hand, finite element homogenization (FEH) is more accurate but comes with relatively high computational cost. In this paper, we propose a data-driven transfer learning framework that combines the advantages of mean-field homogenization and finite element homogenization. Specifically, we generate a large amount of mean-field homogenization data to build a pre-trained model, and then fine-tune it using a relatively small amount of finite element homogenization data. Numerical examples were conducted to validate the proposed framework and verify the accuracy of the analysis. The results of this study are expected to be applicable to the analysis of materials with various foam structures.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.6B
• /
• pp.503-512
• /
• 2009

Watershed models, which are a tool for water cycle mechanism, are classified as the distributed model and the lumped model. Currently, the distributed models have been more widely used than lumped model for many researches and applications. The lumped model estimates the parameters in the conceptual and empirical sense, on the other hand, in the case of distributed model the first-guess value is estimated from the grid-based watershed characteristics and rainfall data. Therefore, the distributed model needs more detailed parameter adjustment in its calibration and also one should precisely understand the model parameters' characteristics and sensitivity. This study uses Jungnang basin as a study area and $Vflo^{TM}$ model, which is a physics-based distributed hydrologic model, is used to analyze its parameters' sensitivity. To begin with, 100 years frequency-design rainfall is derived from Huff's method for rainfall duration of 6 hours, then the discharge is simulated using the calibrated parameters of $Vflo^{TM}$ model. As a result, hydraulic conductivity and overland's roughness have an effect on runoff depth and peak discharge, respectively, while channel's roughness have influence on travel time and peak discharge.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.399-401
• /
• 2022

A definitive diagnosis to classify the breast malignancy status may be achieved by microscopic analysis using surgical open biopsy. However, this procedure requires experts in the specializing of histopathological image analysis directing to time-consuming and high cost. To overcome these issues, deep learning is considered practically efficient to categorize breast cancer into benign and malignant from histopathological images in order to assist pathologists. This study presents a pre-trained convolutional neural network model architecture with a 100% fine-tuning scheme and Adagrad optimizer to classify the breast cancer histopathological images into benign and malignant using a 40× magnification BreaKHis dataset. The pre-trained architecture was constructed using the InceptionResNetV2 model to generate a modified InceptionResNetV2 by substituting the last layer with dense and dropout layers. The results by demonstrating training loss of 0.25%, training accuracy of 99.96%, validation loss of 3.10%, validation accuracy of 99.41%, test loss of 8.46%, and test accuracy of 98.75% indicated that the modified InceptionResNetV2 model is reliable to predict the breast malignancy type from histopathological images. Future works are necessary to focus on k-fold cross-validation, optimizer, model, hyperparameter optimization, and classification on 100×, 200×, and 400× magnification.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.11 no.3
• /
• pp.226-233
• /
• 2023

In this study, two phases were conducted to investigate the direct injection of gaseous CO₂ into cement mortar. The aim was to advance carbon capture, utilization, and storage (CCUS) technology by harnessing industrial waste CO₂ from the domestic ready-mixed concrete industry. In the first phase, the factors influencing the physical properties of cement mortar when using gaseous CO₂ were identified. This included a review of materials to achieve physical properties comparable to a reference formulation. As a result of this phase, it was confirmed that traditional approaches, such as adjusting the water-to-cement ratio, had limitations in achieving the desired physical properties. Consequently, the second phase focused on the optimization of CO₂-injected mortar. This involved studying the CO₂ application and mixing method for cement mortar. Changes in properties were observed when gaseous CO₂ was injected into the mortar. The optimal injection quantity and time to enhance the compressive strength of mortar were determinded. As a result, this study indicated that an extra mixing time exceeding 120 seconds was necessary, compared to conventional mortar. The optimal CO₂ injection rate was identified as 0.1 to 0.2 % by weight of cement, taking both flowability and compressive strength performance into account. Increasing the CO₂ injection time did not further enhance strength. For this approach to be employed as a CCUS technology, additional studies are required, including a microstructural analysis evaluating the amount of immobilized CO₂.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.69-90
• /
• 2023

The purpose of this study is to evaluate the service attributes and consumer reactions of telemedicine apps in South Korea and visualize their differentiation by constructing positioning maps. We crawled 23,219 user reviews of 6 major telemedicine apps in Korea from the Google Play store. Topics were derived by BERTopic modeling, and sentiment scores for each topic were calculated through KoBERT sentiment analysis. As a result, five service characteristics in the application attribute category and three in the medical service category were derived. Based on this, a two-dimensional positioning map was constructed through principal component analysis. This study proposes an objective service evaluation method based on text mining, which has implications. In sum, this study combines empirical statistical methods and text mining techniques based on user review texts of telemedicine apps. It presents a system of service attribute elicitation, sentiment analysis, and product positioning. This can serve as an effective way to objectively diagnose the service quality and consumer responses of telemedicine applications.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.2
• /
• pp.139-145
• /
• 2024

This paper presents the design of a voltage controlled oscillator (VCO) which is one of the key building blocks in modern wireless communication systems with small VCO gain (K_vco) variation. To compensate conventional large K_vco variation, a series varactor bank has been added to the conventional LC-tank with parallel capacitor bank array. And also, in order to achieve excellent phase noise performance while maintaining wide tuning range, a mixed coarse/fine tuning scheme(series varactor array and parallel capacitor array) is chosen. The switched varactor array bank is controlled by the same digital code for switched capacitor array without additional digital circuits. For use at a low voltage of 1.2V, the proposed current reference circuit in this paper used a current reference circuit for safety with the common gate removed more safely. Implemented in a TSMC 0.13㎛ CMOS RF technology, the proposed VCO can be tuned from 4.4GH to 5.3GHz with the K_vco (VCO gain ) variation of less than 9.6%. While consuming 3.1mA from a 1.2V supply, the VCO has -120dBc/Hz phase noise at 1MHz offset from the carrier of the 5.3 GHz.