• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2001.05a
• /
• pp.257-260
• /
• 2001

일반적으로 기계학습은 교사신호의 유무에 따라 교사학습과 비교사학습, 그리고 간접교사에 의한 강화학습으로 분류할 수 있다. 강화학습이란 용어는 원래 실험 심리학에서 동물의 학습방법 연구에서 비롯되었으나, 최근에는 공학 특히 인공생명분야에서 뉴럴 네트워크의 학습 알고리즘으로 많은 관심을 끌고 있다. 강화학습은 제어기 또는 에이전트의 행동에 대한 보상을 최대화하는 상태-행동 규칙이나 행동발생 전략을 찾아내는 것이다. 본 논문에서는 최근 많이 연구되고 있는 강화학습의 방법과 연구동향을 소개하고, 특히 인공생명 연구에 있어서 강하학습의 중요성을 역설한다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 1999.10a
• /
• pp.99-107
• /
• 1999

인공지능(Artificial Intelligence : AI)이란 인간이 가지고 있는 지각, 인식, 이해, 기억, 판단, 학습, 사고, 발견, 창조 등과 같은 지적인 능력을 기계에 실현하고자 하는 분야이다. 즉 기계에 인간이 가진 지능을 복제하여 우리 인간이 소유하고 있는 추론과 학습 메커니즘 등을 이용하여 신속한 계산을 수행할 수 있도록 컴퓨터의 실제 능력을 향상시키는 것이다. 인공지능 분야는 다양한 분야에 응용되어 왔으며 수많은 기술들이 개발되었다. 석유 분야도 예외는 아니며 석유 지질, 탐사, 매장량 평가, 시추, 생산, 수송, 정제 등 다양한 분야에 걸쳐서 인공지능 분야가 활용되어 많은 문제 해결에 적용되고 있다. 이와 같은 취지에서 본 논문에서는 우선 석유 분야별로 활용되고 있는 인공 지능 분야들을 정리하고 그 후 인공지능 분야별로 실제 해결하고 있는 석유 분야의 문제들을 다시 한번 정리하였다. 그 후 특히 석유 분야가 있어서 실제 개발되어 사용중인 전문가 시스템들을 정리하였다. 마지막으로 향후 석유 분야의 전문가 시스템 발전 방향을 간략히 분석하였다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.12
• /
• pp.6455-6460
• /
• 2013

Learner modeling forms the foundations for intelligent tutoring systems that provide adaptive and active learning guidance for learning and education quality enhancement. The aim of this study was to develop learner modeling technologies to form the foundation of intelligent tutoring systems. Specific research tasks include learner modeling building techniques, diverse learner state diagnosis methods and educational data mining.

• Annual Conference of KIPS
• /
• 2024.10a
• /
• pp.356-359
• /
• 2024

현대 사회에서 큰 문제중 하나인 보이스 피싱은 다양한 기술을 사용하여 많은 사람에게 피해를 주고 있다. 이러한 문제를 해결하고자 많은 시도가 있었고 그 중 한 방법이 인공지능을 활용하여 보이스 피싱을 탐지하는 방법이다. 하지만 인공지능을 활용하는 과정에서 데이터 프라이버시 보장이 이루어지지 않는 문제가 발생한다. 본 논문에서는 이러한 문제를 해결하고자 연합학습을 사용한 개인정보 보호에 특화된 보이스 피싱 탐지 기술을 제안한다. 연합학습은 사용자의 데이터를 중앙 서버로 전송하지 않고 로컬 디바이스에서 학습한 모델만을 서버로 전송하여 개인정보 유출을 방지하는 인공지능 학습 방법이다. 또한 스노클을 활용한 오토 데이터 라벨링 기법을 적용하여 피싱 여부를 자동으로 분류하고 탐지 성능을 향상시킨다. 본 기술은 일반적인 인공지능 학습 기술과 비교하여 좋은 성능을 보이며 특히 피싱 탐지에서 중요한 부분인 긍정 클래스 예측 부분에서 높은 성능을 보이고 있다.

• Journal of the Korean Applied Science and Technology
• /
• v.37 no.5
• /
• pp.1268-1284
• /
• 2020

The purpose of this study was to examine mediating effect of emotional intelligence on relationship between nursing students' learning engagement and academic resilience, find whether self-regulation would moderating the relationship between learning engagement and emotional intelligence. The data were collected from 277 nursing students from three colleges in J province and were analyzed with a regression analysis and bootstrapping. As a result of the study, first, the fit of the causal model between learning engagement, emotional intelligence, academic resilience and self-regulation of nursing students was found to be good, and the causal relationship between variables was predicted appropriately. Second, partially mediating effect of emotional intelligence on the path of nursing students' learning engagement affecting academic resilience. Third, self-regulation had moderating effect on learning engagement affecting emotional intelligence. Finally, the significance of this study is that the influence of various variables that can affect the academic resilience of nursing college students was verified, and in order to improve academic resilience, a strategy that considers the subjects' learning engagement, emotional intelligence, and self-regulation.

• Annual Conference of KIPS
• /
• 2000.10a
• /
• pp.387-390
• /
• 2000

웹 기반의 학습평가 시스템 중에는 미리 입력하여 놓은 평가문제를 정해진 순서대로 학습자에게 제시하므로 인하여 학습자의 수준이 전혀 고려되지 않은 문제점이 있었다. 학습자의 수준에 맞는 평가문제를 제공하는 많은 시스템에서도 단순히 문제의 난이도를 지정해 놓고 이를 선택적으로 추출하여 제공해주는 정도의 수준에 머무르고 있다. 결국 학습자의 학업성취도가 고려되지 않은 학습평가가 이루어지게 되는 것이다. 따라서 본 연구에서는 지능형 에이전트를 이용함으로써 좀 더 타당성 있는 평가문제를 학습자에게 제시하여, 평가문제마다 보다 효율적인 학습평가가 이루어 질 수 있도록 지능형 에이전트를 이용한 학습평가시스템을 설계 제안하였다.

• The Journal of the Korea Contents Association
• /
• v.12 no.5
• /
• pp.346-356
• /
• 2012

The purpose of this study is to examine the effect of opinion leader behavior on E-learning satisfaction on the mediating role of social intelligence. For the study, opinion leader behavior were defined two groups such as information mediated behavior, influence behavior and e-learning satisfaction were defined two groups such as contents satisfaction, operation satisfaction. On the basis of theoretical linkages between the constructs, a conceptual model and hypotheses were established. Data were collected from 153 graduated students by structured questionnaires. Collected data were analyzed by PLS(Partial Least Square) statistics program and findings are as follows. Empirical results indicate that opinion leader behavior has a positive impact on opinion leader's social intelligence, social intelligence has a positive impact on E-learning satisfaction. Opinion leader's social intelligence has partially mediated effect on the relationship between opinion leader behavior and E-learning satisfaction.

• Journal of Digital Convergence
• /
• v.17 no.10
• /
• pp.251-258
• /
• 2019

In the future, various products are created in various fields using artificial intelligence. In this age, it is a very important problem to know the operation principle of artificial intelligence learning method and to use it correctly. This paper introduces artificial intelligence learning methods that have been known so far. Learning of artificial intelligence is based on the fixed point iteration method of mathematics. The GD(Gradient Descent) method, which adjusts the convergence speed based on the fixed point iteration method, the Momentum method to summate the amount of gradient, and finally, the Adam method that mixed these methods. This paper describes the advantages and disadvantages of each method. In particularly, the Adam method having adaptivity controls learning ability of machine learning. And we analyze how these methods affect digital signals. The changes in the learning process of digital signals are the basis of accurate application and accurate judgment in the future work and research using artificial intelligence.

• Korean Journal of Cognitive Science
• /
• v.31 no.4
• /
• pp.137-154
• /
• 2020

Psychological properties of learners have influence on learning behaviors in various ways. The purpose of this study was to examine how the goal orientation of learners affected the learning time distribution method. Regulatory focus and theories of intelligence were measured and manipulated in order to differentiate participants' goal-oriented state. Two variables are known to be key variables influencing learner's goal orientation, inducing the approach-avoidance strategy and mastery-performance oriented attitude. In the experiment, the control focus was divided into two groups based on the inclination test score (regulatory Focus Questionnaire, RFQ), and TOI(theory of intelligence) was temporally induced through manipulation to confirm the interaction between the two variables. Participants were able to determine the order of learning freely by learning a set of Spanish-Korean word pairs and then selecting the items they would like to re-learn. Word pairs consisted of difficult or easy items, and learners could learn the same word many times if they wanted to. In the results, promotion-incremental group showed allocating difficult word-pairs in early time.

• Journal of Korean Home Economics Education Association
• /
• v.33 no.4
• /
• pp.85-101
• /
• 2021

The purpose of this study was to make suggestions for improvement by analyzing the activity tasks in the clothing life area in middle school 「Technology & Home Economics」 textbooks of the 2015 revised curriculum. For this purpose, the multiple intelligence teaching-learning strategy analysis criteria were reconstructed and used for analysis. The activity tasks of the clothing life area of 「Technology & Home Economics I」 textbooks from 12 different publishers were analyzed based on the reconstructed analysis criteria, and the content validity was verified by 11 experts. The content validity, assessed by CVI was 0.94. According to the results, the logical·mathematical intelligence accounted for the highest proportion with 31.02%, followed by linguistic intelligence(23.81%), visual/spatial intelligence(17.08%), intrapersonal intelligence(14.71%), interpersonal intelligence(5.79%), bodily/kinesthetic intelligence(5.22%), naturalistic intelligence(2.37%), and musical intelligence(0.00%). The results showed that the teaching-learning strategies most frequently implemented in clothing life area were logical/mathematical intelligence, linguistic intelligence, visual/spatial intelligence, and intrapersonal intelligence. On the other hand, teaching-learning strategies related to interpersonal intelligence, bodily/kinesthetic intelligence, and naturalistic intelligence were used at a relatively low proportion. Therefore, it is recommended to expand the teaching-learning strategies of interpersonal, bodily/kinesthetic, naturalistic and musical intelligence, for a more balanced intelligence development of students.