• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.2
• /
• pp.61-68
• /
• 2020

E-sports has grown steadily in recent years and has become a popular sport in the world. In this paper, we propose a win-loss prediction model of League of Legends at the start of the game. In League of Legends, the combination of a champion statistics of the team that is made through each player's selection affects the win-loss of the game. The proposed model is a deep learning model based on Bidirectional LSTM embedding which considers a combination of champion statistics for each team without any domain knowledge. Compared with other prediction models, the highest prediction accuracy of 58.07% was evaluated in the proposed model considering a combination of champion statistics for each team.

• Journal of The Korean Association of Information Education
• /
• v.13 no.4
• /
• pp.471-478
• /
• 2009

In this paper, we intend to relive game addiction by educating game programming for students that suffer from game addiction. We had analyzed popular games at the present day and extracted factors related to information education from game programming tool. In addition, we had designed the practical lesson plan based on the instructional design model. We had applied the game programming project to students for a year and the results were as follows. First, we verified the drop of addiction propensity by performing the examination of game addiction scale with before and after T-test. Second, according to the results of analysis based on grounded theory, it saved game time and relieved game addiction. In conclusion, this study shows that the game programming project relieved game addiction.

• Journal of KIISE:Software and Applications
• /
• v.35 no.3
• /
• pp.179-188
• /
• 2008

An important issue in multiagent reinforcement learning is how an agent should team its optimal policy through trial-and-error interactions in a dynamic environment where there exist other agents able to influence its own performance. Most previous works for multiagent reinforcement teaming tend to apply single-agent reinforcement learning techniques without any extensions or are based upon some unrealistic assumptions even though they build and use explicit models of other agents. In this paper, basic concepts that constitute the common foundation of multiagent reinforcement learning techniques are first formulated, and then, based on these concepts, previous works are compared in terms of characteristics and limitations. After that, a policy model of the opponent agent and a new multiagent reinforcement learning method using this model are introduced. Unlike previous works, the proposed multiagent reinforcement learning method utilize a policy model instead of the Q function model of the opponent agent. Moreover, this learning method can improve learning efficiency by using a simpler one than other richer but time-consuming policy models such as Finite State Machines(FSM) and Markov chains. In this paper. the Cat and Mouse game is introduced as an adversarial multiagent environment. And effectiveness of the proposed multiagent reinforcement learning method is analyzed through experiments using this game as testbed.

• Journal of Korea Game Society
• /
• v.15 no.3
• /
• pp.189-200
• /
• 2015

These days computer graphics technology has been applied in diverse IT fields. Needs for computer graphics such as 3D Printer, Head Mount Display, VR & AR are growing rapidly. Computer graphics will be more specialized and demanding for graphics specialists will be also increased. However, serious mathematical background obstructs people to learning computer graphics. An efficient computer graphics learning system would be helpful for graphics experts training. By analyzing the graphics theory, we propose an educational software system with that students can effectively learn computer graphics. Our system focuses on theoretical objects of computer graphics and enhances accessibility and intuition using web and blocks.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.1
• /
• pp.34-49
• /
• 2015

Market means of spectrum trading have been utilized as a vital method of spectrum sharing and access in future cognitive radio system. In this paper, we consider the spectrum trading with multiple primary carrier providers (PCP) leasing the spectrum to multiple secondary carrier providers (SCP) for a short period of time. Several factors including the price of the resource, duration of leasing, and the spectrum quality guides the proposed model. We formulate three trading policies based on the game theory for dynamic spectrum access in a LTE based cognitive radio system (CRS). In the first, we consider utility function based resource sharing (UFRS) without any knowledge of past transaction. In the second policy, each SCP deals with PCP using a non-cooperative resource sharing (NCRS) method which employs optimal strategy based on reinforcement learning. In variation of second policy, third policy adopts a Nash bargaining while incorporating a recommendation entity in resource sharing (RERS). The simulation results suggest overall increase in throughput while maintaining higher spectrum efficiency and fairness.

• The Journal of the Korea Contents Association
• /
• v.6 no.3
• /
• pp.85-92
• /
• 2006

Problem-Based Learning(PBL) is one of constructionism's model which is learner-centered learning. In this paper, a role-play base PBL using character is proposed to study importance of information security. It is different from the other PBL models in that it reflects the characteristics of learners, learning task. And it is learning support system which the learners preside their own learning activities using Internet and mobile devices. By experimental result showed that proposed method was more effective than traditional teacher-oriented teaching method about information security in self-directed learning, cooperative learning, contents-making and attraction.

• Journal of KIISE:Software and Applications
• /
• v.26 no.12
• /
• pp.1507-1518
• /
• 1999

본 논문은 제어대상체의 지식을 이용하여 적절한 의사결정을 내리거나 또는 지속적으로 변화하는 주변환경에 적응해 나갈 수 있는 지능시스템 설계를 위한 내배엽성 모델링 방법론을 제시한다. 이러한 지능적 내배엽성 시스템은 의사결정 모델, 지식기반의 내부모델, 그리고 내부모델의 구축모델 등을 기반으로 달성될 수 있다. 학습기능의 모델링을 위하여 수정된 귀납추론 방법과 적응형 전문가 시스템 방법이 제안되었다. 제시된 방법론은 지능적 학습 및 의사결정 기능을 갖춘 지능적 카드경기자 모델링의 예를 통하여 그 가능성을 검증하였다. Abstract This paper presents an endomorphic modeling methodology for designing intelligent systems that can determine by itself using its knowledge of the world and adapt itself to continuously changing circumstances. We have developed such an intelligent endomorphic system by integrating the decision making component and knowledge based internal model with internal model construction model. Learning capabilities are established using the modified inductive reasoning and adaptive expert system techniques we developed. Proposed methodology has been successfully applied to a design of intelligent card game players capable of supporting the intelligent learning and decision making.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.8
• /
• pp.17-25
• /
• 2013

TIn this paper, we propose a model automatically scoring a student's answer for a descriptive problem by using a voting method. Considering the model construction cost, the proposed model does not separately construct the automatic scoring model per problem type. In order to utilize features useful for automatically scoring the descriptive answers, the proposed model extracts feature values from the results, generated by comparing the student's answer with the answer sheet. For the purpose of improving the precision of the scoring result, the proposed model collects the scoring results classified by a few machine learning based classifiers, and unanimously selects the scoring result as the final result. Experimental results show that the single machine learning based classifier C4.5 takes 83.00% on precision while the proposed model improve the precision up to 90.57% by using three machine learning based classifiers C4.5, ME, and SVM.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.01a
• /
• pp.59-64
• /
• 2001

The robot soccer simulation game is a dynamic multi-agent environment. In this paper we suggest a new reinforcement learning approach to each agent's dynamic positioning in such dynamic environment. Reinforcement learning is the machine learning in which an agent learns from indirect, delayed reward an optimal policy to chose sequences of actions that produce the greatest cumulative reward. Therefore the reinforcement learning is different from supervised learning in the sense that there is no presentation of input pairs as training examples. Furthermore, model-free reinforcement learning algorithms like Q-learning do not require defining or learning any models of the surrounding environment. Nevertheless it can learn the optimal policy if the agent can visit every state- action pair infinitely. However, the biggest problem of monolithic reinforcement learning is that its straightforward applications do not successfully scale up to more complex environments due to the intractable large space of states. In order to address this problem. we suggest Adaptive Mediation-based Modular Q-Learning (AMMQL)as an improvement of the existing Modular Q-Learning (MQL). While simple modular Q-learning combines the results from each learning module in a fixed way, AMMQL combines them in a more flexible way by assigning different weight to each module according to its contribution to rewards. Therefore in addition to resolving the problem of large state effectively, AMMQL can show higher adaptability to environmental changes than pure MQL. This paper introduces the concept of AMMQL and presents details of its application into dynamic positioning of robot soccer agents.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.10
• /
• pp.1309-1316
• /
• 2001

In the majority of todays animation and computer games, the behaviours of characters are controlled by pre-defined game logic or pre-generated motion. As game developers strive for richer and more interactive games, they often encounter limitations with this approach. This paper attempts to construct a game model using Genetic Algorithms (GAs) in order to produce more intelligent and compelling computer games. Based on learning ability, the use of GAs will enable the characters to continually evolve, providing a changing and dynamic game environment. A real-time game was implemented to investigate the performance and limitations of the system.