• Journal of the Korea Society of Computer and Information
• /
• v.25 no.3
• /
• pp.27-32
• /
• 2020

In this paper, we propose an algorithm to measure the accumulated status of scrap boxes where metal scraps are accumulated. The accumulated status measuring is defined as a multi-class classification problem, and the method with deep learning classify the accumulated status using only the scrap box image. The learning was conducted by the Transfer Learning method, and the deep learning model was NASNet-A. In order to improve the accuracy of the model, we combined the Random Forest classifier with the trained NASNet-A and improved the model through post-processing. Testing with 4,195 data collected in the field showed 55% accuracy when only NASNet-A was applied, and the proposed method, NASNet with Random Forest, improved the accuracy by 88%.

• The Mathematical Education
• /
• v.62 no.3
• /
• pp.401-416
• /
• 2023

This study systematically examined the influence of preservice teachers' self-efficacy and AI anxiety, on the intention to use AI programs 'knock-knock! mathematics expedition' in mathematics lessons based on a technology acceptance model. The research model was established with variables including self-efficacy, AI anxiety, perceived ease of use, perceived usefulness, and intention of use from 254 pre-service teachers. The structural relationships and direct and indirect effects between these variables were examined through structural equation modeling. The results indicated that self-efficacy significantly affected perceived ease of use, perceived usefulness, and intention to use. In contrast, AI anxiety did not significantly influence perceived ease of use and perceived usefulness. Perceived ease of use significantly affected perceived usefulness and intention to use and perceived usefulness significantly affected intention to use. The findings offer insights and strategies for encouraging the use of 'knock-knock! mathematics expedition' by preservice teachers in mathematics lessons.

• Convergence Security Journal
• /
• v.23 no.5
• /
• pp.3-8
• /
• 2023

The Fourth Industrial Revolution is bringing about great changes in many industrial fields, and among them, active research is being conducted on convergence technology using artificial intelligence. Among them, the demand is increasing day by day in the field of object recognition using artificial intelligence and digital transformation using recognition results. In this paper, we proposed an optimal learning model construction method to accurately recognize letters, symbols, and lines in images and save the recognition results as files in a standardized format so that they can be used in simulations. In order to recognize letters, symbols, and lines in images, the characteristics of each recognition target were analyzed and the optimal recognition technique was selected. Next, a method to build an optimal learning model was proposed to improve the recognition rate for each recognition target. The recognition results were confirmed by setting different order and weights for character, symbol, and line recognition, and a plan for recognition post-processing was also prepared. The final recognition results were saved in a standardized format that can be used for various processing such as simulation. The excellent performance of building the optimal learning model proposed in this paper was confirmed through experiments.

• Journal of the Korean Society of Radiology
• /
• v.18 no.3
• /
• pp.267-273
• /
• 2024

Using a cloud-based vertex AI platform that can develop an artificial intelligence learning model without coding, this study easily developed an artificial intelligence learning model by the non-professional general public and confirmed its clinical applicability. Nine dental diseases and 2,999 root disease X-ray images released on the Kaggle site were used for the learning data, and learning, verification, and test data images were randomly classified. Image classification and multi-label learning were performed through hyper-parameter tuning work using a learning pipeline in vertex AI's basic learning model workflow. As a result of performing AutoML(Automated Machine Learning), AUC(Area Under Curve) was found to be 0.967, precision was 95.6%, and reproduction rate was 95.2%. It was confirmed that the learned artificial intelligence model was sufficient for clinical diagnosis.

• ETRI Journal
• /
• v.44 no.2
• /
• pp.244-259
• /
• 2022

Bicycle- or bike-sharing systems (BSSs) have received increasing attention as a secondary transportation mode due to their advantages, for example, accessibility, prevention of air pollution, and health promotion. However, in BSSs, due to bias in bike demands, the bike rebalancing problem should be solved. Various methods have been proposed to solve this problem; however, it is difficult to apply such methods to small cities because bike demand is sparse, and there are many practical issues to solve. Thus, we propose a demand prediction model using multiple classifiers, time grouping, categorization, weather analysis, and station correlation information. In addition, we analyze real-world relocation data by relocation managers and propose a relocation algorithm based on the analytical results to solve the bike rebalancing problem. The proposed system is compared experimentally with the results obtained by the real relocation managers.

• Electronics and Telecommunications Trends
• /
• v.38 no.5
• /
• pp.1-11
• /
• 2023

The emergence of generative hyperscale artificial intelligence (AI) has enabled new services, such as image-generating AI and conversational AI based on large language models. Such services likely lead to the influx of numerous users, who cannot be handled using conventional AI models. Furthermore, the exponential increase in training data, computations, and high user demand of AI models has led to intensive hardware resource consumption, highlighting the need to develop domain-specific semiconductors for hyperscale AI. In this technical report, we describe development trends in technologies for hyperscale AI processors pursued by domestic and foreign semiconductor companies, such as NVIDIA, Graphcore, Tesla, Google, Meta, SAPEON, FuriosaAI, and Rebellions.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.354-359
• /
• 2022

에세이 자동 평가는 주어진 에세이를 읽고 자동으로 평가하는 작업이다. 본 논문에서는 효과적인 에세이 자동 평가 모델을 위해 Argument Mining 작업을 사용하여 에세이의 논증 구조가 반영된 에세이 표현을 만들고, 에세이의 평가 항목별 표현을 학습하는 방법을 제안한다. 실험을 통해 제안하는 에세이 표현이 사전 학습 언어 모델로 얻은 표현보다 우수함을 입증했으며, 에세이 평가를 위해 평가 항목별로 다른 표현을 학습하는 것이 보다 효과적임을 보였다. 최종 제안 모델의 성능은 QWK 기준으로 0.543에서 0.627까지 향상되어 사람의 평가와 상당히 일치한다.

• Annual Conference of KIPS
• /
• 2022.11a
• /
• pp.550-552
• /
• 2022

최근 사물인터넷(IoT) 기기가 활성화됨에 따라 웨어러블 장치 환경에서 장기간 모니터링 및 수집이 가능해짐에 따라 생체 신호 처리 및 ECG 분석 연구가 활성화되고 있다. 그러나, ECG 데이터는 부정맥 비트의 불규칙적인 발생으로 인한 클래스 불균형 문제와 근육의 떨림 및 신호의 미약등과 같은 잡음으로 인해 낮은 신호 품질이 발생할 수 있으며 훈련용 공개데이터 세트가 작다는 특징을 갖는다. 이 논문에서는 ECG 1D 신호를 2D 스펙트로그램 이미지로 변환하여 잡음의 영향을 최소화하고 전이학습과 메타학습의 장점을 결합하여 클래스 불균형 문제와 소수의 데이터에서도 빠른 학습이 가능하다는 특징을 갖는다. 따라서, 이 논문에서는 ECG 스펙트럼 이미지를 사용하여 2D-CNN 메타-전이 학습 기반 부정맥 분류 기법을 제안한다.

• Steel and Composite Structures
• /
• v.52 no.4
• /
• pp.475-485
• /
• 2024

The mechanical behavior of the steel tube encased high-strength concrete (STHC) composite walls under constant axial load and cyclically increasing lateral load was studied. Conclusions are drawn based on experimental observations, grey evolutionary algorithm and finite element (FE) simulations. The use of steel tube wall panels improved the load capacity and ductility of the specimens. STHC composite walls withstand more load cycles and show more stable hysteresis performance than conventional high strength concrete (HSC) walls. After the maximum load, the bearing capacity of the STHC composite wall was gradually reduced, and the wall did not collapse under the influence of the steel pipe. For analysis of the bending capacity of STHC composite walls based on artificial intelligence tools, an analysis model is proposed that takes into account the limiting effect of steel pipes. The results of this model agree well with the test results, indicating that the model can be used to predict the bearing capacity of STHC composite walls. Based on a reasonable material constitutive model and the limiting effect of steel pipes, a finite element model of the STHC composite wall was created. The finite elements agree well with the experimental results in terms of hysteresis curve, load-deformation curve and peak load.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.5
• /
• pp.29-36
• /
• 2022

In this paper, we propose an efficient agricultural products price prediction model based on dataset which provided in DACON. This model is XGBoost and CatBoost, and as an algorithm of the Gradient Boosting series, the average accuracy and execution time are superior to the existing Logistic Regression and Random Forest. Based on these advantages, we design a machine learning model that predicts prices 1 week, 2 weeks, and 4 weeks from the previous prices of agricultural products. The XGBoost model can derive the best performance by adjusting hyperparameters using the XGBoost Regressor library, which is a regression model. The implemented model is verified using the API provided by DACON, and performance evaluation is performed for each model. Because XGBoost conducts its own overfitting regulation, it derives excellent performance despite a small dataset, but it was found that the performance was lower than LGBM in terms of temporal performance such as learning time and prediction time.