• Journal of The Korean Association of Information Education
• /
• v.25 no.6
• /
• pp.1015-1024
• /
• 2021

There are various educational programming environments in which students can train artificial intelligence (AI) using block-based programming languages, such as Entry, Machine Learning for Kids, and Teachable Machine. However, these programming environments are designed so that students can train AI through a separate menu, and then use the trained model in the code editor. These approaches have the advantage that students can check the training process more intuitively, but there is also the disadvantage that both the training menu and the code editor must be used. In this paper, we present a novel artificial intelligence block that can perform both AI training and programming in the code editor. While this AI block is presented as a Scratch block, the training process is performed through a Python server. We describe the blocks in detail through the process of training a model to classify a blue pen and a red pen, and a model to classify a dental mask and a KF94 mask. Also, we experimentally show that our approach is not significantly different from Teachable Machine in terms of performance.

• Annual Conference of KIPS
• /
• 2021.11a
• /
• pp.637-639
• /
• 2021

인공지능 기반의 생활폐기물의 인식 및 선별에서, 선별 정확도의 저하는 인식 대상의 형태적 다양성과 학습데이터 부족 및 불균등성에 기인한다. 본 연구에서는 비전 인공지능 기반의 효과적인 폐기물 선별을 위한 인식 시스템 및 감독학습 기반의 인공지능 학습 기법을 제안한다. 생활폐기물 중 순환자원적 가치가 높은 CAN, PET, 그리고 이와 형상적으로 유사한 폐기물에 대해 본 연구에서 제안된 시스템에서 물체원형 및 훼손된 형태의 총 18 종 이미지 데이터를 대상으로, 감독학습기반의 인공지능 모델 제작에서 최적의 데이터 레이블링을 위한 분류체계를 제시한다.

• The KIPS Transactions:PartB
• /
• v.10B no.7
• /
• pp.795-802
• /
• 2003

To live autonomously, intelligent agents such as robots or virtual characters need ability that recognizes given environment, and learns and chooses adaptive actions. So, we propose an action selection/learning mechanism in intelligent agents. The proposed mechanism employs a hybrid system which integrates a behavior-based method using the reinforcement learning and a cognitive-based method using the symbolic learning. The characteristics of our mechanism are as follows. First, because it learns adaptive actions about environment using reinforcement learning, our agents have flexibility about environmental changes. Second, because it learns environmental factors for the agent's goals using inductive machine learning and association rules, the agent learns and selects appropriate actions faster in given surrounding and more efficiently in extended surroundings. Third, in implementing the intelligent agents, we considers only the recognized states which are found by a state detector rather than by all states. Because this method consider only necessary states, we can reduce the space of memory. And because it represents and processes new states dynamically, we can cope with the change of environment spontaneously.

• Industry Promotion Research
• /
• v.9 no.1
• /
• pp.213-221
• /
• 2024

This study aims to determine whether learning motivation and grit mediate in the relationship between emotional intelligence and learning engagement among Chinese high school students. Data were collected through a survey targeting 304 high school students purposively sampled from a high school in China. The collected data was analyzed using SPSS PC+ Win ver. 25.0 and SPSS PROCESS macro ver. 4.2. The statistical methods applied were frequency analysis, reliability analysis, correlation analysis, and dual mediation effect analysis. The conclusion of the study is as follows. First, emotional intelligence, learning motivation, grit, and learning engagement all showed significant positive correlations. Second, high school students' learning motivation and grit double-mediated in the relationship between emotional intelligence and learning engagement. Based on these results, this study proposed a plan to improve high school students' academic engagement by utilizing not only emotional intelligence but also learning motivation and grit.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.05a
• /
• pp.310-313
• /
• 2006

지능형 학습 시스템(ITS, Intelligent Tutoring System)은 학습자의 학습 스타일을 인지하여 학습자에 맞는 학습전략을 세우고 적절한 학습 서비스를 제공하는 시스템이다. 기존의 학습시스템은 학습자의 학습 스타일 보다는 학습 컨텐츠에 중심을 두어 학습자에게 맞는 학습 전략을 적절히 세우는 과정이 부족했다. 이에 본 논문에서는 학습자의 학습과정에서 발생한 데이터를 기반으로 학습자의 학습 스타일을 파악하는 방법을 제안한다. 이를 위해 서양 건축양식 학습을 위한 교육 컨텐츠를 이용하였으며, 수집된 데이터를 분석하여 Folder & Silverman 이 제시한 학습 스타일에 근거한 학습자의 학습 스타일을 추출하였다. 실험에서는 70명의 데이터를 수집하였고, 학습자가 교육 컨텐츠를 학습한 순서에 대한 시계열 데이터를 기반으로 학습자 성향을 알아보기 위하여 은닉 마코프 모델(Hidden Markov Model)을 사용하였다. 은닉 마코프 모델을 적용하여 얻은 분석 결과를 가지고 각 학습자에게 맞는 학습 스타일을 진단하였다. 은닉 마코프 모델에서 얻은 학습 스타일 진단 모델은 향후에 학습자 학습 스타일을 파악하는데 사용할 수 있으며, ITS에 있어 학습자 성향 분석 모듈로 고려해볼 수 있다.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.704-706
• /
• 2023

이 논문에서는 과학기술분야 특화 한국어 사전학습 언어모델인 KorSciDeBERTa를 소개한다. DeBERTa Base 모델을 기반으로 약 146GB의 한국어 논문, 특허 및 보고서 등을 학습하였으며 모델의 총 파라미터의 수는 180M이다. 논문의 연구분야 분류 태스크로 성능을 평가하여 사전학습모델의 유용성을 평가하였다. 구축된 사전학습 언어모델은 한국어 과학기술 분야의 여러 자연어처리 태스크의 성능향상에 활용될 것으로 기대된다.

• Journal of Gifted/Talented Education
• /
• v.23 no.5
• /
• pp.697-713
• /
• 2013

The research aimed to investigate characteristics of middle school students enrolled in a science gifted education center affiliated with university in terms of multiple intelligence, self-regulated learning and personality traits. The 89 subjects in the study responded to questionnaires of multiple intelligence, self-regulated learning ability and a personality trait in October, 2011. It was found that both science and math gifted students presented intrapersonal intelligence as strength and logical-mathematical intelligence as weakness. While physics and earth science gifted ones showed spatial intelligence as strength, chemistry and biology gifted ones did intrapersonal intelligence. For self-regulated learning ability, both science and mathematics gifted students tend to show higher levels than general students, in particular, cognitive and motivation strategies comparatively higher than meta-cognition and environment condition strategies. Characteristics of personal traits widely distributed across science and mathematics gifted students, showing that each gifted student presented distinct characteristics individually. Those gifted students showing certain intelligence such as spatial, intrapersonal, or natural intelligences as strength also showed different characteristics of self-regulated learning ability and personal traits among students showing same intelligence as strength. It was concluded that science and mathematics gifted students showed various characteristics of multiple intelligences, self-regulated learning ability, and personal traits across science and mathematics areas.

• Communications of Mathematical Education
• /
• v.35 no.4
• /
• pp.445-473
• /
• 2021

The purpose of this study is to derive implications for the design of the next curriculum by analyzing the textbooks designed as a new subject in the 2015 revised curriculum. In the mathematics curriculum documents of , 'related learning elements' are presented instead of 'learning elements'. 'Related learning elements' are defined as mathematical concepts or principles that can be used in the context of artificial intelligence, but there are no specific restrictions on the amount and scope of dealing with 'related learning elements'. Accordingly, the aspects of 'related learning elements' reflected in the textbooks were analyzed focusing on the textbook format, the amount and scope of contents, and the ways of using technological tools. There were differences in the format of describing 'related learning elements' in the textbook by textbook and the amount and scope of handling mathematics concepts. Although similar technological tools were dealt with in each textbook so that 'related learning elements' could be used in the context of artificial intelligence, the focus was on computations and interpretation of results. In order to fully reflect the intention of the curriculum in textbooks, a systematic discussion on 'related learning elements' will be necessary. Additionally, in order for students to experience the use of mathematics in artificial intelligence, substantialized activities that can set and solve problems using technological tools should be included in textbooks.

• Journal of Internet Computing and Services
• /
• v.8 no.5
• /
• pp.91-97
• /
• 2007

Game program had been classed by 3D or on-line game etc, and engine and game programming simply, But, game programmer's kind more classified new, Artifical Intelligence game programmer's role is important. This paper makes game character study and moved by intelligence using reinforcement learning algorithm. Fought with character enemy using developed game, Confirmed whether embodied game character is facile by intelligence, As result of an experiment, we know, studied character defends excellently than randomly moved character.

• KSCI Review
• /
• v.15 no.1
• /
• pp.125-129
• /
• 2007

지능형 홈 네트워크 서비스의 일반적 보급화로 사용자의 필요와 욕구에 밀착한 개인화 서비스를 위한 사용자의 프로파일 및 다양한 상태 정보, 센서 및 기타 환경정보를 통한 동적 상황인지가 가능토록 하는 상황인지(context-aware) 서비스에 대한 필요성이 증대되고 있다. 사용자 행위 학습에 따른 지능적 자동 제어 시스템 구축에서 먼저 고려해야 할 사항은 사용자 행위 학습에 따른 지능적 자동 제어에 대한 기준을 마련하는 것이다. 홈 네트워크 내의 정보가전기기들 환경에 대한 정보를 지속적으로 수집하고 학습 알고리즘을 통하여 분석하며, 분석되어진 정보를 바탕으로 사용자의 성향을 파악하는 것을 주요인으로 간주해야 할 것이다. 이에 따라 본 논문에서는 사용자 능동적 행위에 따른 지능형 홈 제어 시스템을 제안하였다. 또한 지속적인 모니터링을 통하여 사용자의 성향이 파악되면 상황에 따른 최적의 환경을 제공할 수 있도록 홈 네트워크 제어 시스템을 구축하는 것으로 목적으로 하였다. 사용자의 행동 패턴을 분석하고 이를 기반으로 지능적인 서비스를 제공함으로써 사용자 중심의 능동적 서비스 효과들을 얻을 수 있을 것으로 기대한다.