• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권3호
• /
• pp.186-199
• /
• 2013

We pose pattern classification as a density estimation problem where we consider mixtures of generative models under partially labeled data setups. Unlike traditional approaches that estimate density everywhere in data space, we focus on the density along the decision boundary that can yield more discriminative models with superior classification performance. We extend our earlier work on the recursive estimation method for discriminative mixture models to semi-supervised learning setups where some of the data points lack class labels. Our model exploits the mixture structure in the functional gradient framework: it searches for the base mixture component model in a greedy fashion, maximizing the conditional class likelihoods for the labeled data and at the same time minimizing the uncertainty of class label prediction for unlabeled data points. The objective can be effectively imposed as individual mixture component learning on weighted data, hence our mixture learning typically becomes highly efficient for popular base generative models like Gaussians or hidden Markov models. Moreover, apart from the expectation-maximization algorithm, the proposed recursive estimation has several advantages including the lack of need for a pre-determined mixture order and robustness to the choice of initial parameters. We demonstrate the benefits of the proposed approach on a comprehensive set of evaluations consisting of diverse time-series classification problems in semi-supervised scenarios.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제14권4호
• /
• pp.295-304
• /
• 2014

Communication infrastructure supports wide variety of mobile services such as photo and file sharing, location tracking, social network services and instant messaging. However, instances like power-loss and natural disasters disrupt these communication infrastructures unable to render support to these mobile services. Delay-tolerant networks (DTNs) offer a solution to these problems at hand. By utilizing mobility and opportunistic contacts among mobile devices, a plausible communication network can be establish and enable support to mobile applications. This paper presents an energy-efficient, reliable message delivery routing scheme with message prioritization rules for DTN. It uses the context information of nodes (mobile devices) such as the contact history (location and time of contact), speed/velocity, moving direction to determine the best forwarders among nodes in the network. The remaining energy of the nodes is also used to determine the message types a node can deliver successfully. The simulation results show that proposed approach outperforms Epidemic and Prophet routing schemes in terms of delivery ratio, overhead ratio, delivered messages per types and remaining energy.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제9권3호
• /
• pp.206-212
• /
• 2009

This paper presents an optimized integrated task planning and control approach for manipulating a nonholonomic robot by mobile manipulators. Then, we derive a kinematics model and a mobility of the mobile manipulator(M2) platform considering it as the combined system of the manipulator and the mobile robot. to improve task execution efficiency utilizing the redundancy, optimal trajectory of the mobile manipulator(M2) platform are maintained while it is moving to a new task point. A cost function for optimality can be defined as a combination of the square errors of the desired and actual configurations of the mobile robot and of the task robot. In the combination of the two square errors, a newly defined mobility of a mobile robot is utilized as a weighting index. With the aid of the gradient method, the cost function is minimized, so the path trajectory that the M2 platform generates is optimized. The simulation results of the 2 ink planar nonholonomic M2 platform are given to show the effectiveness of the proposed algorithm.

• Journal of Information Processing Systems
• /
• 제8권4호
• /
• pp.621-652
• /
• 2012

Software Estimations provide an inclusive set of directives for software project developers, project managers, and the management in order to produce more realistic estimates based on deficient, uncertain, and noisy data. A range of estimation models are being explored in the industry, as well as in academia, for research purposes but choosing the best model is quite intricate. Estimation by Analogy (EbA) is a form of case based reasoning, which uses fuzzy logic, grey system theory or machine-learning techniques, etc. for optimization. This research compares the estimation accuracy of some conventional data mining models with a hybrid model. Different data mining models are under consideration, including linear regression models like the ordinary least square and ridge regression, and nonlinear models like neural networks, support vector machines, and multivariate adaptive regression splines, etc. A precise and comprehensible predictive model based on the integration of GRA and regression has been introduced and compared. Empirical results have shown that regression when used with GRA gives outstanding results; indicating that the methodology has great potential and can be used as a candidate approach for software effort estimation.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권3호
• /
• pp.230-245
• /
• 2005

In this paper, we develop a gesture-based input device equipped with accelerometers and gyroscopes. The sensors measure the inertial measurements, i.e., accelerations and angular velocities produced by the movement of the system when a user is inputting gestures on a plane surface or in a 3D space. The gyroscope measurements are integrated to give orientation of the device and consequently used to compensate the accelerations. The compensated accelerations are doubly integrated to yield the position of the device. With this approach, a user's gesture input trajectories can be recovered without any external sensors. Three versions of motion tracking algorithms are provided to cope with wide spectrum of applications. Then, a Bayesian network based recognition system processes the recovered trajectories to identify the gesture class. Experimental results convincingly show the feasibility and effectiveness of the proposed gesture input device. In order to show practical use of the proposed input method, we implemented a prototype system, which is a gesture-based remote controller (Magic Wand).

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제8권1호
• /
• pp.42-51
• /
• 2008

We believe that smart spaces, offering pervasive services, will proliferate. However, at present, those islands of smart spaces should be joined seamlessly with each other. As users move about, they will have to roam from one autonomous smart space to another. When they move into the new island of smart space, they should setup their devices and service manually or not have access to the services available in their home spaces. Sometimes, there will conflicts between users when they try to occupy the same space or use a specific device at the same time. It will also be critical to elder people who suffer from Alzheimer or other cognitive impairments when they travel from their smart space to other visited spaces (e.g., grocery stores, museums). Furthermore our experience in building the Gator Tech Smart House reveals to us that home residents generally do not want to lose or be denied all the features or services they have come to expect simply because they move to a new smart space. The seamless inter-space interaction requirements and issues are raised automatically when the ubiquitous pervasive computing system tries to establish the user's service environment by allocating relevant resources after the user moves to a new location where there are no prior settings for the new environment. In this paper, we raise and present several critical inter-space interactions issues impacting middleware architecture design of ubiquitous pervasive computing. We propose requirements for resolving these issues on seamless inter-space operation. We also illustrate our approach and ideas via a service scenario moving around two smart spaces.

• 한국인터넷방송통신학회논문지
• /
• 제13권5호
• /
• pp.183-188
• /
• 2013

객체를 분류하기 위하여 유사한 특징을 기반으로 하는 다양한 클러스터해석은 데이터 마이닝에서 필수적이다. 그러나 많은 데이터베이스에 포함되어 있는 범주형 데이터의 경우에 기존의 분할접근방법은 객체간의 불확실성을 처리하는데 한계가 있다. 범주형 데이터의 분할과정에서 식별불가능에 의한 동치류의 불확실성에 대한 접근논리가 러프집합의 대수학적인 논리에만 국한되어서 알고리즘의 안정성과 효율성이 떨어지는 요인으로 작용하고 있다. 본 논문에서는 범주형 데이터에 존재하는 속성의 의존도를 고려하기 위하여 정보이론적인 척도를 기반으로 러프엔트로피를 정의하고 MMMR이라는 알고리즘을 제안하여 분할속성을 추출한다. 제안된 방법의 성능을 분석하고 비교하기 위하여 K-means, 퍼지에 의한 방법과 표준편차를 이용한 기존의 방법과 비교우위를 ZOO데이터에 국한하여 알아본다. ZOO데이터를 이용하여 기존의 범주형 알고리즘과의 비교우위를 살펴보고 제안된 알고리즘의 효율성을 검증한다.

• 한국지구과학회지
• /
• 제38권7호
• /
• pp.496-510
• /
• 2017

본 연구는 기상청의 기상레이더 관측망을 이용한 하이브리드 고도면 강우추정 기법 기반의 새로운 정량적 합성강수량 추정 방법을 제시한다. HSR기법은 지형클러터, 빔차폐, 비 기상 에코 및 밝은 띠의 영향을 받지 않는 하이브리드 고도면의 반사도를 합성하는 것이 특징이다. HSR 합성반사도는 정적 HSR (STATIC)과 단일편파레이더에 대한 퍼지로직 기법과 이중편파레이더에 대한 시선방향 질감 기반의 품질관리 절차를 사용하는 동적 HSR (DYNAMIC) 합성으로 구분된다. STATIC과 DYNAMIC은 2014년 5월부터 10월까지 10개의 강우 사례에 대해 기상청 현업용 합성강우(MOSAIC)와 비교검증 하였다. 차폐 영역에서 STATIC, DYNAMIC, MOSAIC의 상관계수는 각각 0.52, 0.78, 0.69이며 평균 상대 오차는 각각 34.08, 30.08, 40.71%로 분석되었다.

• 한국지구과학회지
• /
• 제36권1호
• /
• pp.109-124
• /
• 2015

레이더 반사도를 이용한 강수추정의 개선을 위해 새로운 접근 방식인 경북대학교에서 개발한 하이브리드 고도면을 이용한 강수량 추정기법(Hybrid Surface Rainfall, KNU-HSR)을 사용하였다. KNU-HSR기법은 지형에코와 레이더 빔차폐의 영향을 받지 않는 2차원 하이브리드 고도면에서의 반사도를 이용하여 강수량을 추정한다. 본 연구에서는 정적 HSR 및 동적 HSR기법이 사용되었으며 비교 검증되었다. 정적 HSR은 빔차폐지도와 지형에코지도를 사용하며, 동적 HSR은 정적 HSR에 추가적으로 실시간 퍼지로직 품질관리를 통한 품질지수지도를 사용한다. 검증을 위해 상관계수(correlation coefficient), 총비율(total ratio), 평균편의(mean bias), 정규화된 표준편차(normalized standard deviation), 평균 상대오차(mean relative error)를 사용하였으며, 10개 강우사례의 지상우량계 강우자료를 이용하여 두 HSR의 강우추정 성능을 평가하였다. 모든 검증지수에서 동적 HSR은 반사도 보정을 하지 않은 정적 HSR에 비해 더 우수한 성능을 보였다. 동적 HSR은 레이더로부터 근거리에서는 과대추정하였으며 원거리에서는 빔 폭 확장 및 빔 고도증가로 인해 과소추정하였다. 동적 HSR의 정규화된 표준편차와 평균상대오차는 레이더로부터의 거리에 관계없이 가장 좋은 결과를 보였다. 정적 HSR은 약한 강우강도에서 상당히 과대추정하였으나 동적 HSR은 모든 강우강도에서 1.0에 총비율을 보였다. 반사도의 시스템오차 보정 후, 동적 HSR의 정규화된 표준편차와 평균상대오차는 각각 약 20%와 15%로 개선되었다.