• Journal of Oral Medicine and Pain
• /
• 제47권2호
• /
• pp.107-108
• /
• 2022

Artificial intelligence (AI) refers to the use of machines to mimic intelligent human behavior. It involves interactions with humans in clinical settings, and augmented intelligence is considered as a cognitive extension of AI. The importance of AI in healthcare and medicine has been emphasized in recent studies. Machine learning models, such as genetic algorithms, artificial neural networks (ANNs), and fuzzy logic, can learn and examine data to execute various functions. Among them, ANN is the most popular model for diagnosis based on image data. AI is rapidly becoming an adjunct to healthcare professionals and is expected to be human-independent in the near future. The introduction of AI to the diagnosis and treatment of oral diseases worldwide remains in the preliminary stage. AI-based or assisted diagnosis and decision-making will increase the accuracy of the diagnosis and render treatment more precise and personalized. Therefore, dental professionals must actively initiate and lead the development of AI, even if they are unfamiliar with it.

• ETRI Journal
• /
• 제45권2호
• /
• pp.318-328
• /
• 2023

Recently, embedded systems, such as mobile platforms, have multiple processing units that can operate in parallel, such as centralized processing units (CPUs) and neural processing units (NPUs). We can use deep-learning compilers to generate machine code optimized for these embedded systems from a deep neural network (DNN). However, the deep-learning compilers proposed so far generate codes that sequentially execute DNN operators on a single processing unit or parallel codes for graphic processing units (GPUs). In this study, we propose PartitionTuner, an operator scheduler for deep-learning compilers that supports multiple heterogeneous PUs including CPUs and NPUs. PartitionTuner can generate an operator-scheduling plan that uses all available PUs simultaneously to minimize overall DNN inference time. Operator scheduling is based on the analysis of DNN architecture and the performance profiles of individual and group operators measured on heterogeneous processing units. By the experiments for seven DNNs, PartitionTuner generates scheduling plans that perform 5.03% better than a static type-based operator-scheduling technique for SqueezeNet. In addition, PartitionTuner outperforms recent profiling-based operator-scheduling techniques for ResNet50, ResNet18, and SqueezeNet by 7.18%, 5.36%, and 2.73%, respectively.

• 한국IT서비스학회지
• /
• 제20권3호
• /
• pp.41-56
• /
• 2021

New drug discovery and development research enable clinical treatment that saves human life and improves the quality of life, but the possibility of success with new drugs is significantly low despite a long time of 14 to 16 years and a large investment of 2 to 3 trillion won in traditional methods. As artificial intelligence is expected to radically change the new drug development paradigm, artificial intelligence new drug discovery and development projects are underway in various forms of collaboration, such as joint research between global pharmaceutical companies and IT companies, and government-private consortiums. This study uses the TOE framework and the Value-based Adoption Model, and the technical, organizational, and environmental factors that should be considered for the acceptance of AI technology at the level of the new drug research organization are the value of artificial intelligence technology. By analyzing the explanatory power of the relationship between perception and intention to use, it is intended to derive practical implications. Therefore, in this work, we present a research model in which technical, organizational, and environmental factors affecting the introduction of artificial intelligence technologies are mediated by strategic value recognition that takes into account all factors of benefit and sacrifice. Empirical analysis shows that usefulness, technicality, and innovativeness have significantly affected the perceived value of AI drug development systems, and that social influence and technology support infrastructure have significant impact on AI Drug Discovery and Development systems.

• Child Health Nursing Research
• /
• 제28권4호
• /
• pp.291-298
• /
• 2022

Purpose: This study aimed to identify students' awareness of the use of a chatbot (A-uC), a type of artificial intelligence technology, for violence prevention among elementary school students. Methods: The participants comprised 215 students in the fourth to sixth grades in Chuncheon, South Korea, and data were collected via a self-reported questionnaire. Results: The mean A-uC score was 3.43±0.83 out of 5 points. The mean scores for the 4 sub-dimensions of the A-uC tool were 3.48±0.80 for perceived value, 3.44±0.98 for perceived usefulness, 3.63±0.92 for perceived ease of use, and 3.15±1.07 for intention to use. Significant differences were observed in A-uC scores (F=59.26, p<.001) according to the need for the use of chatbots in violence prevention education. The relationships between intention to use and the other A-uC sub-dimensions showed significant correlations with perceived value (r=.85, p<.001), perceived usefulness (r=.76, p<.001), and perceived ease of use (r=.64, p<.001). Conclusion: The results of this study suggest that chatbots can be used in violence prevention education for elementary school students.

• International journal of advanced smart convergence
• /
• 제10권4호
• /
• pp.72-83
• /
• 2021

The use of the predictive analytics (PA) powered by the artificial intelligence (AI) is more important in the movie sector during the COVID-19 pandemic, because Hollywood witnessed the impact of the 'Netflix Effect' and began to invest in data and AI. Our purpose is to discover a few cases of the AI centered PA in the movie industry value chain based on five objectives of PA: Compete, grow, enforce, improve, and satisfy. Even if movie companies' interest is to predict future success for competing with over-the-tops (OTTs) at a first glance, it is observed, once they start to use the PA with the AI, they try to utilize the enhanced PA platforms for remaining four objectives. As a result, ScriptBook, Vault, Pilot, Cinelytic and Merlin Video (Merlin) are use cases for the objective 'compete.' Movio of Vista Group International and Datorama of Salesforce are use cases for the objective 'grow.' Industrial Light & Magic (ILM) and Geena Davis Institute on Gender in Media (GDI) with Disney are use cases for the objective 'enforce.' Watson, Benjamin, and Greenlight Essential are use cases for the objective 'improve.' Disney Research (DR) with Simon Fraser University and California Institute of Technology is the use case for the objective 'satisfy.'

• 한국인터넷방송통신학회논문지
• /
• 제21권4호
• /
• pp.57-62
• /
• 2021

인공지능을 위한 병렬연산 능력이 향상됨에 따라 인공지능 적용 분야가 다양한 방향으로 확대되고 있다. 특히 방대한 데이터를 처리해야 하는 IoT센서의 데이터를 처리하기 위해 인공지능이 도입되고 있다. 하지만 시간에 따른 데이터의 중요도가 달라지는 IoT 시계열 데이터 특성상 기존의 인공지능 학습 기법을 그대로 적용하기에는 한계점이 있다. 본 과제에서는 IoT 센서 데이터를 효과적으로 처리하기 위해 시간가중치기반 및 사용자 상태값 기반 인공지능 처리기법을 연구한다. 상기 기법을 통해 기존 인공지능 학습을 적용시키는 것 보다 높은 센서 정확도를 확보 할 수 있게 된다. 이에 더해, 해당 연구를 기반으로 다양한 분야에서 인공지능 학습을 적용하는 방안을 제시하고, 지속적인 연구를 통해 다양한 분야로의 확장을 기대할 수 있다.

• International Journal of Computer Science & Network Security
• /
• 제23권12호
• /
• pp.13-26
• /
• 2023

Recently, the development of wireless network technologies has been advancing in several directions: increasing data transmission speed, enhancing user mobility, expanding the range of services offered, improving the utilization of the radio frequency spectrum, and enhancing the intelligence of network and subscriber equipment. In this research, a series of contradictions has emerged in the field of wireless network technologies, with the most acute being the contradiction between the growing demand for wireless communication services (on operational frequencies) and natural limitations of frequency resources, in addition to the contradiction between the expansions of the spectrum of services offered by wireless networks, increased quality requirements, and the use of traditional (outdated) management technologies. One effective method for resolving these contradictions is the application of artificial intelligence elements in wireless telecommunication systems. Thus, the development of technologies for building intelligent (cognitive) radio and cognitive wireless networks is a technological imperative of our time. The functions of artificial intelligence in prospective wireless systems and networks can be implemented in various ways. One of the modern approaches to implementing artificial intelligence functions in cognitive wireless network systems is the application of fuzzy logic and fuzzy processors. In this regard, the work focused on exploring the application of fuzzy logic in prospective cognitive wireless systems is considered relevant.

• 마이크로전자및패키징학회지
• /
• 제29권3호
• /
• pp.1-12
• /
• 2022

인공지능 기술의 급속한 발전으로 기존 센서에 인간의 지능과 유사한 기능을 부여하기 위한 연구가 큰 주목을 받고 있다. 기존에는 주로 센서로써의 기초 성능지표, 예를 들어 감도 및 속도 등을 향상시키기 위한 연구가 주로 진행되었지만, 최근에는 분류나 예측 등의 인공지능을 센서에 결합하기 위한 시도가 확대되고 있다. 이를 바탕으로 최근 질병 감지 센서, 모션 감지 센서 및 가스 센서 등 거의 센서 전 분야에서 지능형 센서에 대한 연구 결과가 활발히 보고되고 있다. 본 논문에서는 인공지능의 기본적인 개념, 종류 및 메커니즘과 더불어, 최근 보고된 지능형 센서에의 적용 사례에 대해 알아보고자 한다.

• 디지털융복합연구
• /
• 제18권7호
• /
• pp.47-56
• /
• 2020

본 연구는 인공지능 활용 교육에 대한 초등교사의 인식을 종합적으로 분석하는 목적을 지닌다. 최근 학교 교육 현장에서 인공지능 기술의 활용에 대한 관심이 증대되고 있다. 하지만 초등학교 교사들이 이를 어떻게 인식하는지를 확인하는 연구는 미흡하다. 본 연구는 초등교사 69명을 대상으로 기술통계, 중다회귀분석, 의미변별척도를 활용하여 초등교사들이 교육에서 인공지능 활용에 대해 어떻게 인식하는지를 총체적으로 분석하였다. 연구 결과, 초등교사들은 인공지능 기술이 수업 시간 내 활동을 보조하는데 가장 적합하다고 응답하였으며 교수학습 방법 측면에서는 문제중심학습이 가장 적절하다고 인식하고 있었다. 인공지능의 교육적 활용에 대해 영향을 미치는 요소는 학습 내용, 학습 자료, 인공지능 기기로 나타났다. 인공지능 활용 교육은 개별학습, 참여 촉진, 흥미 유발 등의 특성을 지닌다고 인식하였다. 향후 최적화된 교육 운영을 가능하게 하는 수업 전략이나 모형 개발 등이 이루어질 필요가 있다.

• 한국정보통신학회논문지
• /
• 제25권7호
• /
• pp.877-883
• /
• 2021

인공지능 기술은 제조 로봇, 인공지능 스피커, 로봇 청소기 등 산업 및 스마트홈 분야에서 다양하게 사용되고 있다. 본 논문에서는 RCMS(Real-time Cash Management System)에서 활용하기 위한 인공지능 기반 1:1 챗봇(chatbot) 시스템을 설계 및 구현하였다. 구현한 RCMS 챗봇은 기존 온라인 게시판의 1만 3천 5백여건의 질의응답을 기반으로 연구비 사용, 시스템 사용법 등 9개 영역에 총 210개의 질의시나리오로 구축하였다. 챗봇은 부족한 상담인원 문제를 해소하고, 근무시간 이후에 연구자의 문의에 대응하여 사용자의 만족도를 제고 할 것으로 예상되며, 연구자의 상담문의가 가장 많았던 사용비목에 대한 추천 서비스는 상담건수를 감소시켜 다른 상담문의에 대한 답변의 질적 수준 향상이 기대된다.