• Annual Conference of KIPS
• /
• 2024.05a
• /
• pp.834-835
• /
• 2024

시장 경쟁력을 확보하고 기업을 성장시키기 위해서는 광고 행위가 필수적이므로 현재까지 효율적으로 광고하기 위한 여러 가지 방안들이 활용되었다. 이 중에는 타 업체와의 경쟁전략을 위해서 경쟁업체의 광고비를 파악하려는 과정도 포함 되어있다. 이에 디지털 광고 측면에서는 상대적으로 광고의 노출, 클릭, 시간 대 등의 관련 정보를 획득하기 용이하므로 본 연구에서는 대량의 데이터를 이용하고 XGBoost(Extreme Gradient Boosting) 알고리즘을 활용하여 크롤링된 데이터 그룹을 분석하고, 클릭 수를 예측하는 모델을 구현하였다. 실험 결과 모델의 RMSE(Root Mean Squared Error) Average 가 1.13 정도 나온 것을 확인하였고 이에 따른 과적합을 피하기 위한 방안을 검토하였다.

• Journal of Practical Engineering Education
• /
• v.15 no.3
• /
• pp.605-614
• /
• 2023

In In the era of digital transition, AI-based personalized services are emerging in the field of education. This research aims to examine the development strategies for implementing AI-based learning services in school. Focusing on AI-based math learning service "Math Cell" developed by i-Scream Edu, this study surveyed the functional requirements from the perspective of an educator. The results were analyzed for importance and suitability using IPA, and expert opinions were surveyed to explore specific development directions for the service. Consequently, importance in all areas such as diagnosis, learning, evaluation, and management averaged 4.82 and performance averaged 4.56, showing excellent results in most questions, and in particular, importance was higher than performance. Among certain detailed functions, concept learning, customized task presentation, evaluation result analysis function, dashboard-related functions, and learning materials in the dashboard were not intuitive for students to understand and had to be supplemented. This study provides meaningful insights by summarizing expert opinions on AI-based personalized mathematics learning services, thereby contributing to the exploration of the development strategies for "Math Cell".

• Korean Journal of Remote Sensing
• /
• v.38 no.5_3
• /
• pp.811-818
• /
• 2022

Recently, the number of natural and environmental disasters is rapidly increasing due to extreme weather caused by climate change, and the scale of economic losses and damage to human life is increasing accordingly. In addition, with urbanization and industrialization, the characteristics and scale of extreme weather appearance are becoming more complex and large in different ways from the past, and need for remote sensing and artificial intelligence technology for responding and managing global environmental disasters. This special issue investigates environmental disaster observation and management research using remote sensing and artificial intelligence technology, and introduces the results of disaster-related studies such as drought, flood, air pollution, and marine pollution, etc. in South Korea performed by the i-SEED (School of Integrated Science for Sustainable Earth and Environmental Disaster at Pukyong National University). In this special issue, we expect that the results can contribute to the development of monitoring and management technologies that may prevent environmental disasters and reduce damage in advance.

• Annual Conference of KIPS
• /
• 2021.05a
• /
• pp.202-204
• /
• 2021

다양한 IoT 기기로 구성된 모바일 IoT 환경에서는 IoT에서 수집된 데이터가 다른 IoT의 학습 데이터로 사용되는 순환 구조로 이루어져 있다. 따라서 해당 환경에서 데이터는 공유되는 자원이며 매우 중요한 요소이다. 특히 IoT 기기가 밀집된 지역에서는 많은 트래픽이 발생하기 때문에 전송지연 및 데이터 손실로 인한 시스템 성능이 저하되는 문제가 발생한다. 따라서 본 논문에서는 안전한 데이터 통신을 위한 블록체인 기반의 시스템 구조를 제안한다. 해당 시스템은 블록체인을 사용하여 IoT 기기의 이동성과 밀집도를 판별하고, 트래픽 밀집 구역이 발생하였을 경우 UAV를 활용하여 통신이 원활하게 이루어질 수 있도록 한다.

• Annual Conference of KIPS
• /
• 2022.05a
• /
• pp.260-263
• /
• 2022

현재의 교통정보시스템은 수집된 정보를 서버에서 가공하여 서비스하는 형태로 이루어져 있다. 이러한 형태는 네트워크 구성이 비교적 단순하고 유지관리 비용이 적게 든다는 장점이 있지만, 반면에 실시간성이 저하되고 보안이 제대로 보장되지 않을 수 있다는 문제가 있으며, 최근 많은 연구가 이루어지고 있는 VANET 환경에서의 교통정보시스템도 broadcast storm의 가능성을 안고 있다. 본 연구에서 제안하는 교통정보시스템은 자동차가 수집한 돌발 상황에 대한 데이터를 RSU(Road Side Unit)가 수신하고, 이후 메시지를 노드들에게 보낼 때 블록체인에 업로드함으로써 보안성과 broadcast storm 문제들을 해결할 수 있으며, raw data 를 IPFS 에 저장하여 시스템 고도화에 사용할 수 있어 참여자들이 교통 상황에 대해 신속하게 대응할 수 있도록 하는 장점을 갖는다.

• The Journal of Information Systems
• /
• v.30 no.4
• /
• pp.95-118
• /
• 2021

Purpose This study verified the moderating effect of consumer knowledge in relation to the factors affecting the acceptance intention of autonomous vehicles by adding trust to the United Theory of Acceptance and Use of Technology model for the commercialization of autonomous vehicles. Design/methodology/approach For this purpose, this study conducted a survey on general consumers who are interested in automobiles. A total of 250 questionnaires were distributed and collected, and 242 questionnaires were used for analysis. To test the hypotheses, multiple regression analysis and multiple group analysis were performed. Findings Performance expectations, effort expectations, social influence, and trust were found to have a positive effect on the acceptance intention of autonomous vehicles. In addition, consumer knowledge between performance expectation and acceptance intention and between effort expectation and acceptance intention was confirmed as a variable that can moderate the relationship.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2021.11a
• /
• pp.223-224
• /
• 2021

The process before the model learning stage in AI R&D can be subdivided into data collection/cleansing-data purification-data labeling. After that, according to the purpose of development, it goes through a stage of verifying the model by performing learning by using the algorithm of the artificial intelligence model. Several studies describe an important part of AI research as the learning stage, and try to increase the accuracy by changing the structure and layer of the AI model. However, if the refinement and labeling process of the learning data is tailored only to the model format and is not made for the purpose of development, the desired AI model cannot be obtained. The latest research reveals that most AI research failures are the failure of the learning data rather than the structure of the AI model. analyzed.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.4
• /
• pp.659-666
• /
• 2023

Among the tools based on generative artificial intelligence, the possibility of using ChatGPT is being explored. However, studies that have confirmed what factors should be considered when using it educationally based on learners' actual perceptions are insufficient. Through qualitative research method, this study was to derive consideration factors when using ChatGPT in the education. The results showed that there were five key factors as follows: critical thinking on generated information, recognizing it as a tool to support learning and avoiding dependent use, conducting prior training on ethical usage, generating clear and appropriate questions, and reviewing and synthesizing answers. It is necessary to develop an instructional design model that comprehensively composes the above elements.

• Journal of the Microelectronics and Packaging Society
• /
• v.30 no.3
• /
• pp.11-19
• /
• 2023

Recently with the rapid advancement of artificial intelligence (AI) technologies such as Chat GPT, AI semiconductors have become important. AI technologies require the ability to process large volumes of data quickly, as they perform tasks such as big data processing, deep learning, and algorithms. However, AI semiconductors encounter challenges with excessive power consumption and data bottlenecks during the processing of large-scale data. Thus, the latest packaging technologies are required for AI semiconductor computations. In this study, the authors have described packaging technologies applicable to AI semiconductors, including interposers, Through-Silicon-Via (TSV), bumping, Chiplet, and hybrid bonding. These technologies are expected to contribute to enhance the power efficiency and processing speed of AI semiconductors.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.137-138
• /
• 2022

소스 코드 식별자의 이름을 잘 정하는 것은 소프트웨어 엔지니어링에서 중요한 문제로 다루어지고 있다. 프로그램 엔티티의 의미있고 간결한 이름은 코드 이해도에 중요한 역할을 하며, 소프트웨어 유지보수 관리 비용을 줄이는 데에 큰 효과가 있다. 이러한 코드 식별자 중 평균적으로 가장 복잡한 식별자는 '메소드 이름'으로 알려져 있다. 본 논문에서는 메소드 내용과 일관성 있는 적절한 메소드 이름 생성을 자연어 처리 태스크 중 하나인 '추상적 텍스트 요약'으로 치환하여 수행하는 트랜스포머 기반의 인코더-디코더 모델을 제안한다. 제안하는 모델은 Github 오픈소스를 크롤링한 Java 데이터셋에서 기존 최신 메소드 이름 생성 모델보다 약 50% 이상의 성능향상을 보였다. 이를 통해 적절한 메소드 작명에 필요한 비용 절감 달성 및 다양한 소스 코드 관련 태스크를 언어 모델의 성능을 활용하여 해결하는 데 도움이 될 것으로 기대된다.