Earticle

Wireless Mesh Networks is increasingly being applied widely with explosive traffic demand. This leads to a great challenge for traditional routing protocols in ensuring Quality of Service. We propose a QoS-guaranteed intelligent routing algorithm in this paper for WMN with heavy traffic load using reinforcement learning to improve its performance. We build a reward function for the Q-Learning algorithm to choose a route so that the packet delivery ratio is the highest. Concurrently, the learning rate coefficient is flexibly changed to determine constraints of the end-to-end delay. Our performance evaluations show that the proposed algorithm has significantly improved performance compared with other well-known routing algorithms.

Reconfigurable intelligent surfaces (RISs) represent a highly promising technology that enhances the capacity and coverage of wireless networks by intelligently reconfiguring the wireless propagation environment in highly advanced wireless communications. The objective of this study is to solve the problem of secrecy rate maximization for multiple RIS-aided millimeter-wave communications by jointly optimizing the active RISs and the RIS phase shifts of the considered system. For this nonconvex problem, we propose multitask learning in a deep neural network to predict the RIS phase shift and active RISs. Numerical results based on realistic, three-dimensional, ray-tracing simulations show that the proposed solution can predict the RIS phase and active RIS with an accuracy rate > 96%. These results confirm the viability of RIS-aided secure wireless communications.

Due to the complex functioning of Smart Healthcare Systems (SHS), many security concerns have been raised in the past. It provisions the attackers to hamper the working of SHS in a variety of ways, e.g., injection of false data to replace vital signs, tampering of medical devices to prevent informing critical situations, etc. In this work, a novel ML-based framework, i.e., SmartHealth is proposed to secure IoMT devices in SHS. SmartHealth watches the vital signs gathered through different IoMT to analyze the change in various body activities to differentiate between normal activities and dangerous security attacks. The performance of the SmartHealth is also analyzed for three different dangerous attacks. During performance analysis, it has been observed that SmartHealth can identify wicked activities in IoMT 92% times accurately with an F1-score of 90%.

Recently, in computer vision behavior recognition is an active research area that plays a significant role in smart cities for crime prevention and urban safety. However, without base knowledge of Artificial Intelligence (AI) designing an efficient model is very difficult because we need data and programing skills for implementing the system. To tackle this problem, we designed and implemented a system that allows a user having no professional knowledge to easily and conveniently create a deep learning model. The interface of this system consists of Data Selection, Model Training and Testing, and Model Parameter values according to domains and categories. In addition, we designed a function to check the test results for the model selected by the user. This system allows users to quickly and easily create and test models.

교육공학이 발달하고 교육현장에 이를 실현하는 목적은 곧 교수효과를 높이고 학습자들의 학습결과를 극대화하는데 있을 것이다. 이 연구는 멀티미디어 기술이 교수-학습현장에 일반화 되면서 World Wide Web과 Courseware 들이 가지고 있는 한계들을 보다 더 효율적인 측면과 인간에 가까운 지능화된 교수체계(Intelligent Tutoring System: ITS)를 실현하는데 필요한 연구들을 분석하였다. 연구방법으로는 ERIC(교육연구정보센터) Database를 이용하여 자료를 수집하였고, 자료수집에 이용한 Descriptors는 “Intelligent Teaching and Learning"과 "Intelligent Instruction"이었다. 여기서 얻은 자료를 인구학적(독일, 러시아, 미국, 벨기에, 한국 등) 연구자 관점(Ethnographic Conduct)에 따라 최종적으로 연구에 사용되었다. 연구결과 이들은 "Intelligent"의 접근을 Courseware, Coumputer Language, 그리고 Teaching Process & Tool 의 영역들에서 보다 더 인간에 가까워지는 수단으로 각기 개념들을 개발하거나 적용하고 있었다. 결과적으로는 교육자들과 학습자들의 수월성과 효율성을 위한 관점에서 지능적인 ITS, AI(Artificial Intelligence)등이 강조되고 있었다. 특히 심리-운동적 측면의 고려는 물론, 학습내용 이외에 예상되는 질문사항, 흥미, 다단계별 test, 상시 피드백, 연상과 이해 등의 고지능적 측면들 까지도 적용되고 있었다. 또한, 수업을 위한 교수방법 선택 과정에 까지 ITS 용어가 적용되고 있어 앞으로 각종 미디어와 컴퓨터의 교육적용이 Human-brainized 되고 있음을 알 수 있다. 따라서 실기가 많이 요구되는 교양실과 및 직업관련 교과들에서도 지능화된 ITS 개발과 교수-학습전략의 실천이 타 학문분야와 균형있는 경쟁력을 갖추기 위해서 교재개발이 촉진되어야 할 것이다.

본 논문은 도시 교통 혼잡 문제를 해결하기 위해 IoT 기반 스마트 폴 인프라와 DQN 강화학습 알고리즘을 결합한 엣지지능형 교통 신호 최적화 프레임워크를 제안한다. 제안 방법은 교차로에 설치된 스마트 폴을 통해 차량 대기열, 교통 밀도, 대기시간 등의 데이터를 실시간으로 수집하고, 엣지 계층에서 DQN 기반 신호 제어 정책을 수행하도록 설계하였다. 또한 VEINS 기반 공동 시뮬레이션 환경에서 SUMO와 OMNeT++를 연동하여 고정시간 제어 및 감응식 제어 방식과 성능을 비교하였다. 또한 자율주행차량 환경에서 V2X 통신의 성능을 최적화하고, 높은 이동성과 네트워크 불안정성이 특징인 VANET환경에서 효과적인 지능형 교통 시스템(ITS)을 구현하기 위한 SDVN 기반 설계 방법을 제안한다. 기존의 VANET은 분산형 구조로 인해 패킷 손실, 지연 증가, 네트워크 단절 등 여러 한계가 존재하며, 이를 해결하기 위해 본 연구는 SDN 아키텍처를 차량 통신망에 통합하였다. SDN의 중앙집중 제어 특성을 활용하여 네트워크 토폴로지를 실시간으로 관리하고, 트래픽 상황에 따라 최적의 경로를 동적으로 설정할 수 있도록 하였다.

This paper proposes an edge-intelligent traffic signal optimization framework that integrates IoT-based smart pole infrastructure with a DQN reinforcement learning algorithm to address urban traffic congestion problems. The proposed approach is designed to collect real-time traffic information, including vehicle queue lengths, traffic density, and waiting times, through smart poles installed at intersections, while executing DQN-based signal control policies at the edge layer. Furthermore, a VEINS-based co-simulation environment integrating SUMO and OMNeT++ was employed to evaluate the performance of the proposed framework in comparison with conventional fixed-time and actuated traffic signal control methods. In addition, this study proposes an SDVN-based design methodology to optimize V2X communication performance in autonomous vehicle environments and to implement an effective ITS in VANET environments characterized by high mobility and network instability. Conventional VANETs suffer from several limitations, including packet loss, increased latency, and frequent network disconnections due to their distributed architecture. To overcome these challenges, this research integrates SDN architecture into vehicular communication networks. By leveraging the centralized control capabilities of SDN, the proposed approach enables real-time management of network topology and dynamic configuration of optimal communication routes according to changing traffic conditions.

Today, the network becomes the core element in all that is done efficiently and effectively. They include block transfer, linear transfer, and asynchronous transfer. Optical Burst Switching (OBS) is also classified with them. By picking on data sent with OBS, some security failures occur, and these comprise Replay Attacks, Spoofing, and Burst Header Packet (BHP) flooding attacks which are among these threats. The addressed methodology incorporates the application of the Support Vector Machine (SVM) algorithm to fight down BHP attacks. The simulation outcomes reveal that the performance which is obtained from the actual learning algorithm is the best at 97.7% in all four classes of flooding attacks which include NB-No Block, NB-Wait, No Block, or Block. This proposed Intelligent Identification of BHP Flooding Attack on OBS utilizing Machine Learning Technique (I2BHPOBSML) shows that it is giving better results than the past Works.

In order to solve the safety hazards caused by the operation failure of elevators, pump rooms, fire control facilities, access control and other equipment and facilities in colleges, this paper uses the Internet of Things technology to realize the ubiquitous access of equipment and facilities in colleges, and collect the operating status data of the equipment and facilities in real time. Establish an equipment and facility operation data center, use big data and deep learning technology to realize online monitoring and real-time early warning of equipment and facilities, and send the monitoring and early warning information to managers and equipment maintenance personnel for processing, and realize the monitoring, early warning, and processing of equipment and facilities Automated process management to build an intelligent monitoring system for college equipment and facilities.

전통적인 웹기반의 학습시스템은 다양한 학습콘텐츠를 운영한다. 그러나 학습자가 자신에게 적절한 학습콘텐츠를 선택하는 것은 쉬운 일이 아니다. 본 논문에서는 학습자가 학습을 계획할 때 학습이력 및 다양한 학습정보를 담고 있는 학습자 프로파일을 기초로 하여, 학습자에게 가장 적합한 선호도와 피드백을 제공할 수 있는 학습콘텐츠 제공 방법을 제안하였다. 즉 학습자의 프로파일을 분석하여 학습자에게 제공할 긍정적 피드백과 평가를 결정한다. 또한 학습자가 틀린 부분에 대한 지식을 스스로 학습할 수 있도록 학습콘텐츠를 적응적으로 제공한다. 그 결과, 학습자 프로파일 분석을 통해 긍정적인 피드백을 바탕으로 자신이 오류를 배울 수 있도록 하여, 학습의 적용 결과가 가장 잘 나오는 형태로 학습 콘텐츠를 적응적으로 제공해 주는 기법을 적용하였다. 본 논문에서 구현한 지능형 학습시스템은 실제 학습자에게 적용하였으며 사용한 학습자들을 대상으로 학습 만족도를 설문하여 그 결과를 분석하였다.

Typical Web-based learning system is operating the variety of learning contents. However, it is not easy to efficiently select appropriate learning contents. In this paper, we propose a learning content delivery method that can provide the most suitable preferences and feedback to the learner. By analyzing the profile of the learner, it determines the positive feedback and evaluation to be provided to the learner. The result of applying learning techniques were applied to provide the best learning content to adaptively out the form. The proposed method appears as a learning experience and learning outcomes are higher after a study was conducted to suggest that could help in the learning progress of the students themselves. This paper are applied to real learners. And the learners using the system were surveyed by the questionnaire on learning experience and learning outcomes were analyzed.

방대해지고 있는 온라인 패션 시장에서는 소비자도 자신이 원하는 스타일에 대해 키워드 검색으로 원하는 패션 스타일을 일일이 찾기란 쉽지 않은 일이다. 이를 해소해줄 수 있는 것은 소비자의 니즈를 반영한 패션 추천이다. 기존 온라인 쇼핑 사이트는 소비자의 니즈를 파악하고 추천하기 위하여 설문조사 형식으로 소비자의 선호 스타일을 파악하는 것이 대부 분이었다. 본 논문에서는 기존 방법의 한계점을 해소하고자 암묵적 프로파일링 방법을 통하여 소비자들의 니즈와 선호하는 스타일에 대해 간편하고 효과적으로 파악할 수 있는 모델을 제안하였다. 또한 이렇게 수집된 데이터로 학습한 딥러닝기반의 지능형 선호 패션 모델을 통하여 이미지 자체에 대한 특성을 반영하도록 학습하는 방법을 제안하였다. 제안한 모델의 정성적 평가를 통하여 의미있는 결과를 얻을 수 있었다.

In the massive online fashion market, it is not easy for consumers to find the fashion style they want by keyword search for their preferred style. It can be resolved into consumer needs based fashion recommendation. Most of the existing online shopping sites have collected cumtomer’s preference style using the online quastionnair. In this paper, we propose a simple but effective novel model that resolve the traditional method in fashion profiling for consumer’s preference style and needs using implicit profiling method. In addition, we proposed a learning model that reflects the characteristics of the images itself through the deep learning-based intelligent preferred fashion model learned from the collected data. We show that the proposed model gave meaningful results through the qualitative evaluation.

정보통신기술의 발달 및 이용 증가로 인해 우편 사업의 매출 비중이 감소하고 경영수지 적자를 기록하고 있다. 이에 대한 개선 방법 중 미래의 우편 물량을 예측하고 해당 결과를 통해 미리 대비하는 효율적인 운영 방법을 제시한다. 본 논문은 특정 지역에서 발생할 우편 물량을 예측하는 방법으로 GAN(Generative Adversarial Network)을 이용한 데이터 증폭 기술과 딥러닝 기반 우편 물량 예측 기술을 통합한 지능형 우편 물량 예측 기술을 제시한다.

최근 개발자와 사용자 사이에서 게임의 인공지능에 대한 관심이 늘어나고 있다. 그래픽과 사운드 요소의 한계로 인 해 화려하고 웅장한 게임보다 다양한 재미를 줄 수 있는 게임을 원하기 때문이다. 기존의 게임 인공지능 기법은 단순 하여 쉽게 질리고, 사용자에게 다양한 재미를 지속적으로 제공하기 어렵다. 그러나 학습 능력을 갖춘 게임은 다양하고 예측하기 어려운 특성으로 인해 사용자에게 끝없는 재미를 제공할 수 있다. 본 논문에서는 이러한 학습 능력을 게임 내의 에이전트에게 부여하기 위해 심리학 이론 중 하나인 관찰학습 이론을 적용한 알고리즘을 제시하고자 한다.

Cyber education system service is in the field of software service which is highlighted after the latter half of 1990'. But the progress of this service is impeded by the lack of back office which contributes to the evaluation of learning achievement and the management of learning progress. This article points out the problem of current back office which is the most important in the cyber education system, and focuses on the new intelligent learning achievement evaluation module. First, we define the cause and effect between the learning stages using by fuzzy implication which is the important part of fuzzy theory. Next, we suggest the model which generates the results of the learning achievement evaluation. This model, suggested by this article, may contribute to the development of the cyber education system by improving the current on-line education service.

하천 인근에 위치한 빗물펌프장은 유수지를 대상으로 적절한 규칙에 따라 펌프를 가동함으로써 도심지 및 농경지 침수 피 해를 예방하는 기능을 수행한다. 현재 대부분의 빗물펌프장은 유수지의 수위를 기준으로 단순한 규칙 기반의 펌프운영 정책을 사용하고 있다. 최근 지구온난화로 인한 기후 변화가 예측하기 어려운 강우량의 변화를 발생시키고 있다. 따라서 단순한 펌프 정책으로는 지구온난화로 인한 갑작스러운 유수지 변화에 적절하게 대처하기 어렵다. 본 논문은 강우량과 저수량, 유수지 수 위 등의 정보를 이용해 시스템이 적정 유수지 수위을 유지할 수 있도록 펌프 가동을 선택할 수 있는 심층 강화학습 기반의 자 동 빗물펌프 운용 방법을 제시한다. 제안한 방법의 타당성을 검증하기 위해 강우-유출 모의 모델인 Storm Water Management Model(SWMM)을 이용해 모의실험을 수행하고 현장에서 사용되고 있는 기존 펌프 정책과 성능을 비교하였다.

The rainwater pumping station located near a river prevents river overflow and flood damages by operating several pumps according to the appropriate rules against the reservoir. At the present time, almost all of rainwater pumping stations employ pumping policies based on the simple rules depending only on the water level of reservoir. The ongoing climate change caused by global warming makes it increasingly difficult to predict the amount of rainfall. Therefore, it is difficult to cope with changes in the water level of reservoirs through the simple pumping policy. In this paper, we propose a pump operating method based on deep reinforcement learning which has the ability to select the appropriate number of operating pumps to keep the reservoir to the proper water level using the information of the amount of rainfall, the water volume and current water level of the reservoir. In order to evaluate the performance of the proposed method, the simulations are performed using Storm Water Management Model(SWMM), a dynamic rainfall-runoff-routing simulation model, and the performance of the method is compared with that of a pumping policy being in use in the field.

본 연구는 학습부진학생들을 대상으로 탐구형 소프트웨어인 Excel과 GSP의 활용한 탐구활동을 통해 해석기하의 개념을 형성해가는 과정을 이해하고자 하였다. 본 연구를 위해 중학교의 논증기하에 대한 개념과 고등학교에서 배우는 해석기하의 개념을 관계적으로 이해할 수 있도록 Skemp의 목표 지향적 학습을 위한 지능 모델이 7차시로 구성되었고 2011년 7월～9월에 5명의 학습부진학생을 대상으로 연구가 수행되었다. 연구결과로는 탐구형 소프트웨어를 통해 관계적 이해(유형의 목표를 성취하기 위한)에 비길 수 있는 두 가지는 유형, 즉 직관적 사고 활동을 통한 관계적 이해가 된 경우로 이 과정에 탐구형 소프트웨어를 통하여 반영적 사고가 일어날 수 있다는 것과 유형, 즉 반영적 사고 활동을 통한 도구적 이해가 일어난 경우로 스키마 학습과 같은 장점을 얻게 함으로써 관계적 이해와 같은 효과를 얻을 수 있었다. 더욱이 이러한 관계적 이해가 성취되었을 때 아주 초보적인 단계에서 논리적 이해를 할 수 있는 여지와 양식 2의 기호를 통한 의사소통의 능력에 해당하는 수준까지 가능함을 보여주었다.

The purpose of this study was to examine How the exploratory activities using Excel and GSP which are exploratory software, in learning analytic geometry affected on the underachievers’ analytic geometry concept development process. The subjects of 5 students who received the 8th~9th grades from their examination of the last semester, participated in a total of 7 units based on Skemp's intelligent learning model. The results of the study showed that there were two important cases found to nearly achieve the category . One was reflective thinking could happen through exploratory software in category . The other was the exploratory activities which could have the same effectiveness as the relational understanding in category , as Skemp mentioned that there is a room to be achieved in the elementary level when such relational understanding is achieved.

Recently, Since people always carry mobile devices and can learn whenever and wherever by using not only phone, text, SNS, and TV, but also using learning device, recorder, e-book, e-dictionary, and internet surfing due to rapid distribution of smartphone, the interest in intelligent learning agent related to situated learning is increasing more. The method of practicing the situated learning theories developed so far provides learning contents to learners by analyzing their environment information along with their basic information. However, if learning contents are recommended by considering only basic and partial information of a learner, in many cases, they do not match with the demand from the learner, so it could become factors that decrease learner’s learning desire and obstruct inducing learning motivation. Therefore, this study propose an intelligent learning agent that automatically recommends learning contents that learners want in timely manner in their real life by analyzing not only learners’ location information, but also their available time for learning and Internet activity pattern.

Class imbalance is a problem that is very much critical in many real-world application domains of machine learning. When examples of one class in a training data set vastly outnumber examples of the other class(es), traditional data mining algorithms tend to create suboptimal classification models. Researchers have rigorously studied several techniques to alleviate the problem of class imbalance, including resampling algorithms, and feature selection approaches to this problem. In this paper, we present a new hybrid feature selection algorithm dubbed as Class Imbalance Learning using Intelligent Under Sampling (CILIUS), for learning from skewed training data. This algorithm provides a simpler and faster alternative by using C4.5 as base algorithm. We conduct experiments using four UCI data sets from various application domains using five learning algorithms for comparison and five evaluation metrics. Experimental results show that our method has higher Area under the ROC Curve, F-measure, Precision, TP rate and low TN rate values than many existing class imbalance learning methods.

This paper presents a framework of intelligent tutorial system to incorporate adaptive learning and assess the relative performance of adaptive learning system over general classroom learning. In conventional classroom teaching and in distance learning the students may be from different backgrounds, their need of study and goal may be different; above all their ability to learn may not be similar. So teaching style and fitting teaching material may differ from student to student. A layered architecture of an Intelligent Tutorial System [ITS] for adaptive learning is proposed that assess the requirement, goal and capability of a student and dynamically sets a path for study, the instruction materials are dynamically selected as per the student’s level of understanding from a given set of instruction materials. The system monitors the student and changes the path of study automatically as per the performance. Formative and summative test are taken by the system for decision making – as to which material is best for student and for assessing student’s understanding level. The system also intelligently helps the user to overcome the misconceptions, as human teacher. Experimental results show the impressive improvement of the performances of students in distance mode learning using ITS following the proposed framework.

Recently, the prevalence of high-performance mobile devices and development of IT technologies, including augmented reality technology and location information service, etc. has led to the formation of an educational environment allowing learners to practice context learning theory ideally, subsequently contributing to active progress of relevant studies. Therefore, it is difficult to expect leaners to have an interest and a commitment since it cannot connect a lot of information coming through the perspective of the learner in a real-world environment with situated learning in real time. Accordingly, this paper proposes a intelligent English situated learning system that can practice situation context learning more reasonably using a location-based service and a recognition technology that automatically recognizes text information on street signboards that are easily accessible in everyday life, yet provide learners with a lot of information. The proposed system provides learners with English conversation learning contents that can be used in the business sector related to trade name recognized through text information on street signboards from images captured by cameras.