• Journal of The Korean Association of Information Education
• /
• v.18 no.1
• /
• pp.121-132
• /
• 2014

The main goal of this research is to invest the difficulties in the process of developing serious game for science learning to acquire the suggestion. We analysed the journals written by developers of game scenario on the procedure of the development and on the review of the accomplished game. Then we also interviewed the scenario developers as well as game developers. When interviewing the game developers, we passed them the thoughts and questions of scenario developers. The possible difficulties in developing serious game are form the difference of cognitions on the skills which can be embodied in the development process and interaction and communication problem between scenario developers and game developers. As a consequence of the research, we acquired some suggestions for developing serious game as follows; 1) Scenario developers and game developers must understand the concepts of serious game as well as learning theory; 2) Both scenario developers and game developers should be aware of secured technological capacity and allowed time period. 3) Scenario developer should take a part of game developer' role as using a learning contents authoring tool; 4) Scenario developers have to consistently interact with game developers in developing the games; 5) Scenario should be concretely described in detail; and 6) A Supervisor is essential to control both scenario developers and game developers.

• Journal of The Korean Association For Science Education
• /
• v.39 no.3
• /
• pp.321-335
• /
• 2019

This study aims to explore the influencing factors and the process of enhancing science self-efficacy (SSE) and to lay the foundation in understanding science self-efficacy of students. The ten categories related to the science self-efficacy were derived through the coding of the interview data based on the grounded theory and paradigm analysis to develop a process model of science self-efficacy improvement. Through the process analysis, four cyclical phases were found in the process of enhancing SSE: 'Entering into learning science' phase, 'enhancing SSE' phase, 'adjustment' phase, and 'result' phase. More specifically, the phase of 'entering into learning science' is where students choose science track and stimulated to construct SSE. The phase of 'enhancing SSE' is where students taking a science track actively learn science and perform science activities. In the phase of 'adjustment', students come to have successful performance about learning science and performing science activities by using diverse strategies. Finally, 'result' phase indicates different appearances of students depending on SSE levels. The phases were non-linear and periodically repeat depending on situation. The core category in the selective coding was indicated to be 'enhancing science self-efficacy.' Students' SSE form by learning science and performing science activities. These finding may help better understand the behavior of students who are taking a science track by facilitating effective science learning through the increase of their SSE levels.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10a
• /
• pp.190-192
• /
• 2003

소비자의 신용 대출 규모가 점차 증가하면서 기업에서 고객의 신용 등급에 의한 정확한 고객 분류를 필요로 하고 있다 이를 위해 판별 분석과 신경망의 역전파(BP: Back Propagation)를 이용한 고객 분류 시스템이 연구되었다. 그러나, 판별 분석을 사용한 방법은 불규칙한 신용 거래의 성향을 보이는 비정규 분포의 고객 데이터의 영향으로 여러 개의 판별 함수와 판별점이 존재하여 분류 정확도가 떨어지는 단점이 있다. 신경망을 이용한 방법은 불규칙한 신용 거래의 성향을 보이는 고객 데이터에 의해서, 지역 최소점(Local Minima)에 빠져 최대의 분류 정확률을 보이는 분류자를 얻지 못하는 경우가 발생할 수 있다. 본 논문에서는 이러한 기존 연구의 분류 정확률을 저하시키는 단점을 해결하기 위해 SVM(Support Vector Machine)을 사용하여 고객의 신용 등급을 분류하는 방법을 제안한다. SVM은 SV(Support Vector)의 수에 의해서 학습 성능이 좌우되므로, 불규칙한 거래 성향을 보이는 고객에 대해서도 높은 차원으로의 매핑을 통하여, 효과적으로 학습시킬 수 있어 분류의 정확도를 높일 수 있다 하지만, SVM은 근사화 알고리즘(Approximation Algorithms)을 이용하므로 분류 정확도가 이론적인 성능에 미치지 못한다. 따라서, 본 논문은 점진적 앙상블 SVM을 사용하여, 기존의 고객 분류 시스템의 문제점을 해결하고 실제적으로 SVM의 분류 정확률을 높인다. 실험 결과는 점진적 앙상블 SVM을 이용한 방법의 정확성이 기존의 방법보다 높다는 것을 보여준다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.5
• /
• pp.3434-3439
• /
• 2015

As a criterion of information theoretic learning, the Euclidean distance (ED) of two error probability distribution functions (minimum ED of error, MEDE) has been adopted in nonlinear (decision feedback, DF) supervised equalizer algorithms and has shown significantly improved performance in severe channel distortion and impulsive noise environments. However, the MEDE-DF algorithm has the problem of heavy computational complexity. In this paper, the recursive ED for MEDE-DF algorithm is derived first, and then the feed-forward and feedback section gradients for weight update are estimated recursively. To prove the effectiveness of the recursive gradient estimation for the MEDE-DF algorithm, the number of multiplications are compared and MSE performance in impulsive noise and underwater communication environments is compared through computer simulation. The ratio of the number of multiplications between the proposed DF and the conventional MEDE-DF algorithm is revealed to be $2(9N+4):2(3N^2+3N)$ for the sample size N with the same MSE learning performance in the impulsive noise and underwater channel environment.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.1
• /
• pp.159-170
• /
• 2021

As the main purpose of R&D changes from the center of knowledge creation to the center of economic value creation through technology transfer and commercialization, public technologies can also secure economic feasibility as well as make a social contribution. Korea has been focusing on fostering core human resources who can lead the commercialization of basic and original research results by launching the 'Support project for exploring startups linked with public technology-based markets' since 2015 in order to promote public technology startup. This study is based on the results of a survey for the purpose of analyzing the performance of this project. In addition, this study derived four factors related to the I-corps project performance from the results of this survey and verified the relationship between these factors through structural equation model analysis. In summary, it was confirmed that 'Application Level' and 'Business Model,' which are positively affected directly from 'Entrepreneurship Learning,' have positive effects on 'Financial Resources'. Furthermore, the indirect effect of 'Entrepreneurship Learning' on 'Financial Resources' was verified. In particular, the high level of impact of 'Entrepreneurship Learning' on 'Application Level,' and the impact of 'Application Level' on 'Business Model' and 'Financial Resources' were also positive.

• The Journal of Korean Association of Computer Education
• /
• v.17 no.2
• /
• pp.11-19
• /
• 2014

Presently the instructional model for internet ethics education is modeled after the integrated morality. The model is based on the assumption that ethical awareness will lead to ethical activities which is based on the theory that cognition is correlated to the behavioral domains. But the side effects of the information society in the cyber space increased even when the education for the awareness of ethics in the cyber space has been taught more aggressively than before. In this study, the relation of the cognition for information ethics and the ethical behavior in the cyber space was analyzed in order to find out the implications for the effective internet ethics education model. The tools used are the 'DIT (Defining Issues Test)' to measure the behavioral ability in the physical world, the Information Ethics Judgment to measure the behavioral ability in the cyber space, and the self diagnostic tool of 'Internet ethics awareness' to measure the level of cognitive knowledge for internet ethics. The correlation of three measures was analyzed. The results were college students' levels of ethics from three tools from are considerably low. Moral judgement and information ethics judgement were not correlated which means that the behavior in the physical world was not necessarily correlated to the behavior in the cyber space. The three measurements were not statistically significantly correlated. Therefore the cognitive awareness for the information ethics were not necessarily correlated to the ethical behavior in the cyber space. Ethical cognition and the moral behavior need to be taught with equal emphasis as they do not have strong correlation.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.128-132
• /
• 2012

물은 생물의 생존을 위해 필수적인 요소로 인류가 시작된 이래로 물을 효율적으로 이용하고 안전하게 관리하기 위한 노력은 계속되어 왔다. 최근 지구 온난화가 주요 원인으로 알려진 국지성 집중호우의 피해는 매우 심각하며, 이로 인해 치수에 대한 중요성은 날로 커지고 있다. 지금까지 사용해 왔던 홍수 예 경보 과정은 특정 지점의 유출량을 예측하기 위해서 강우-유출 모형을 운영하였다. 그러나 물리적 모형의 경우 운영에 필요한 매개변수의 결정과정이 복잡하고, 매개변수 결정을 위해 많은 자료를 필요로 한다. 또한 그 매개변수의 결정과정은 많은 불확실성을 포함하고 있어서 모형의 운영을 위한 전처리과정과 계산과정을 거치는 동안 발생한 오차가 누적되어 결과물 속에는 많은 오차가 포함되어 있다. 본 연구에서는 기존의 홍수 예 경보 시스템의 문제점과 불확실성을 최대한 감소시키고 더 우수한 유출량 예측을 위해 neuro-fuzzy 추론 기법을 이용한 모형인 ANFIS(Adaptive Neuro-Fuzzy Inference System)를 사용하여 하천수위를 예측하였다. ANFIS는 신경회로망과 퍼지이론을 결합한 기법으로 신경회로망의 구조와 학습 능력을 이용하여 제어환경에서 획득한 입 출력 정보로부터 언어변수의 membership 함수와 제어규칙을 제어 대상에 적합하도록 자동으로 조종하는 기법이다. 본 연구에서는 ANFIS를 사용하여 탄천 하류에 위치한 대곡교의 수위를 예측하였다. 분석을 위해 2007년부터 2011년까지의 탄천 유역의 관측 강우자료와 수위 자료 중 강우강도와 지속시간, 강우 형태에 따라 7개의 강우사상을 선정하였다. 학습자료 및 보정자료의 변화에 따른 예측 오차를 비교하여 모형의 적용성과 적정성을 평가하였다. 적용결과 입력자료 구성의 경우 해당 시간의 강우량 및 수위자료와 10분 전 강우자료를 이용한 모델이 가장 우수한 예측을 보였고, 학습자료의 경우 자료의 길이가 길고, 최대홍수량이 큰 경우 가장 우수한 예측 결과를 보였다. 본 연구의 적용결과 가장 우수한 모형의 경우 30분 예측 첨두수위 오차는 0.32%, RMSE는 0.05m 이고 예측시간이 길어짐에 따라 오차가 비선형적으로 증가하는 경향을 보였다.

• KIPS Transactions on Computer and Communication Systems
• /
• v.8 no.1
• /
• pp.17-28
• /
• 2019

In the financial investment management strategy, the distributed investment selecting and combining various financial assets is called portfolio management theory. In recent years, the blockchain based financial assets, such as cryptocurrencies, have been traded on several well-known exchanges, and an efficient portfolio management approach is required in order for investors to steadily raise their return on investment in cryptocurrencies. On the other hand, deep learning has shown remarkable results in various fields, and research on application of deep reinforcement learning algorithm to portfolio management has begun. In this paper, we propose an efficient financial portfolio investment management method based on Asynchronous Advantage Actor-Critic (A3C), which is a representative asynchronous reinforcement learning algorithm. In addition, since the conventional cross-entropy function can not be applied to portfolio management, we propose a proper method where the existing cross-entropy is modified to fit the portfolio investment method. Finally, we compare the proposed A3C model with the existing reinforcement learning based cryptography portfolio investment algorithm, and prove that the performance of the proposed A3C model is better than the existing one.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.2 no.2
• /
• pp.69-78
• /
• 1999

In this paper, a neural network approach to forecast Korean regional precipitation is presented. We first analyze the characteristics of the conventional models for time series prediction, and then propose a new model and its learning method for the precipitation forecast. The proposed model is a layered network in which the outputs of a layer are buffered within a given period time and then fed fully connected to the upper layer. This study adopted the dual connections between two layers for the model. The network behavior and learning algorithm for the model are also described. The dual connection structure plays the role of the bias of the ordinary Multi-Layer Perceptron(MLP), and reflects the relationships among the features effectively. From these advantageous features, the model provides the learning efficiency in comparison with the FIR network, which is the most popular model for time series prediction. We have applied the model to the monthly and seasonal forecast of precipitation. The precipitation data and SST(Sea Surface Temperature) data for several decades are used as the learning pattern for the neural network predictor. The experimental results have shown the validity of the proposed model.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.7
• /
• pp.1881-1892
• /
• 1997

This paper proposes the method for recognition of printed chinese characters by probabilistic VQ networks and multi-stage recognizer has hierarchical structure. We use modular neural networks, because it is difficult to construct a large-scale neural network. Problems in this procedure are replaced by probabilistic neural network model. And, Confused Characters which have significant ratio of miss-classification are reclassified using the entropy theory. The experimental object consists of 4,619 chinese characters within the KSC5601 code except the same shape but different code. We have 99.33% recognition rate to the training data, and 92.83% to the test data. And, the recognition speed of system is 4-5 characters per second. Then, these results demonstrate the usefulness of our work.