• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.2037-2040
• /
• 2004

In this paper we introduce a new prototype based classifiers for overlapping data, where training pattern can be overlap on the feature space. The proposed classifier is based on the prototype from neural network classifier (NNC)[1] for overlap data. The method automatically chooses the initial center and two radiuses for each class. The center is used as a mean representative of training data for each class. The unclassified pattern is classified by measure distance from the class center. If the distance is in the lower (shorter radius) the unknown pattern has the high percentage of being in this class. If the distance is between the lower and upper (further radius), the pattern has the probability of being in this class or others. But if the distance is outside the upper, the pattern is not in this class. We borrow the words upper and lower from the rough set to represent the region of certainty [3]. The training algorithm to find number of cluster and their parameters (center, lower, upper) is presented. The clustering result is tested using patterns from Thai handwritten letter and the clustering result is very similar to human eyes clustering.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.11 no.3
• /
• pp.153-159
• /
• 2011

Pan-So-Ri singing a story as song is one of Korea traditional musics. it divide into two sect(east-sect, west-sect), and it is hard to classify two sect without knowledge about Pan-So-Ri. In this paper, we have propose a Pan-So-Ri classification program using PCD(Pitch Class Distribution) and Naive Bayesian Classifier. Attribute value of classifier is each appearance frequency of pitch. Experiment is conducted two time with different rounding off location of probability value. Better one show correct classification with east-sect 80%, west-sect 97%, and total accuracy of 88%. this result is used our program.

• The Journal of Korean Association of Computer Education
• /
• v.5 no.2
• /
• pp.11-19
• /
• 2002

In this paper, we present a simple statistical method for performing word sense disambiguation(WSD), specially for Korean transitive verbs, based on a supervised learning algorithm. This approach combines a set of indicators based on syntactic relations between surrounding words and an ambiguous verb. Experiments with 10 Korean verbs show that accuracy performance of our WSD method using indicators based on syntactic relations is 27% higher than the baseline performance. Moreover, our method using weighting mechanism based on each indicator type is 12% higher than a method which uses only an unordered set of surrounding words in the context.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.2
• /
• pp.775-792
• /
• 2015

Person re-identification is an important and challenging task in computer vision with numerous real world applications. Despite significant progress has been made in the past few years, person re-identification remains an unsolved problem. This paper presents a novel appearance-based approach to person re-identification. The approach exploits region covariance matrix and color histograms to capture the statistical properties and chromatic information of each object. Robustness against low resolution, viewpoint changes and pose variations is achieved by a novel signature, that is, the combination of Log Covariance Matrix feature and HSV histogram (LCMH). In order to further improve re-identification performance, third-party image sets are utilized as a common reference to sufficiently represent any image set with the same type. Distinctive and reliable features for a given image set are extracted through decision boundary between the specific set and a third-party image set supervised by max-margin criteria. This method enables the usage of an existing dataset to represent new image data without time-consuming data collection and annotation. Comparisons with state-of-the-art methods carried out on benchmark datasets demonstrate promising performance of our method.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09b
• /
• pp.71-75
• /
• 2003

오류 역전파 학습 알고리즘을 이용하여 영상 인식에 적용 할 경우에는 은닉층의 노드 수를 경험적으로 설정하므로, 학습시간과 지역최소화 및 정체현상이 발생한다. 그리고 ARTI 알고리즘은 입력 패턴과 저장 패턴간의 측정 방법인 유사성 검증 방법과 경계 변수의 설정에 따라 인식률이 좌우된다. 경계 변수의 값이 크면 입력 패턴과 저장 패턴사이에 약간의 차이만 있어도 새로운 카테고리(Category)로 분류하고, 반대로 경계 변수의 값이 적으면 입력 패턴과 저장 패턴 사이에 많은 차이가 있더라도 유사성이 인정되어 입력 패턴들을 대략적으로 분류한다. 따라서 ART1 알고리즘을 영상 인식에 적용하기 위해서는 경계 변수를 경험적으로 설정하므로 인식률에 부정적인 영향을 갖는 문제점이 있다. 따라서 본 논문에서는 개선된 ART1 알고리즘과 지도 학습 방법을 결합하여 신경망의 은닉층 노드를 동적으로 변화시키는 자가 생성지도 학습 알고리즘을 제안한다. 제안된 신경망에서 입력층과 은닉층의 학습 구조에는 ART1 알고리즘을 개선하여 적용하고, 은닉층과 출력층의 학습 구조에는 은닉층에서 승자로 선택된 노드와 출력층 노드와 연결된 가중치만을 조정하고 Delta-Bar-Delta 알고리즘을 적용한다. 제안된 방법의 학습 성능을 분석하기 위하여 학생증 영상에서 추출한 학번 패턴 분류에 적용한 결과, 기존의 신경망 학습 알고리즘보다 학습 성능이 개선됨을 확인하였다.

• Asia pacific journal of information systems
• /
• v.7 no.1
• /
• pp.99-111
• /
• 1997

This paper presents an application of an artificial neural net to the implementation of decision class analysis (DCA), together with the generation of a decision model influence diagram. The diagram is well-known as a good tool for knowledge representation of complex decision problems. Generating influence diagram model is known to in practice require much time and effort, and the resulting model can be generally applicable to only a specific decision problem. In order to reduce the burden of modeling decision problems, the concept of DCA is introduced. DCA treats a set of decision problems having some degree of similarityz as a single unit. We propose a method utilizing a feedforward neural net with supervised learning rule to develop DCA based on influence diagram, which method consists of two phases: Phase l is to search for relevant chance and value nodes of an individual influence diagram from given decision and specific situations and Phase II elicits arcs among the nodes in the diagram. We also examine the results of neural net simulation with an example of a class of decision problems.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.368-372
• /
• 2009

The method to get the feature have been proposed to recognize the user activity by setting specific action for making the user independent result in previous research. However, it was only applied in specific environment and it was difficult to implement because it regarded only some specific feature as the recognized object. To improve this problem we detected the normality/abnormality of the activity based on the repetition and the continuity of the past activity pattern. We applied the unsupervised learning method, not supervised, and clustered the data which was collected within a certain period of time and we regarded it as the basis of the evaluation of the repetition. We demonstrated to be able to detect the abnormal activity based on wether the data was generated repeatedly.

• Journal of Korean Institute of Industrial Engineers
• /
• v.42 no.6
• /
• pp.421-431
• /
• 2016

Over the last four decades, industrial engineering (IE) research in Korea has continued to evolve and expand to respond to social needs. This paper aims to identify research topics in IE research and explore their dynamic changes over time. The topic modeling approach, which automatically discovers topics that pervade a large and unstructured collection of documents, is adopted to identify research topics in domestic IE research. 1,242 articles published from 2001 to 2015 in two IE journals issued by the Korean Institute of Industrial Engineers were collected and their English abstracts were analyzed. Applying the Latent Dirichlet Allocation model led us to uncover 50 topics of domestic IE research. The top 10 most popular topics are revealed, and topic trends are explored by examining the dynamic changes over time. The four topics, technology management, financial engineering, data mining (supervised learning), efficiency analysis, are selected as hot topics while several traditional topics related with manufacturing are revealed as cold topics. The findings are expected to provide fruitful implications for IE researchers.

• 한국방재학회:학술대회논문집
• /
• 2011.02a
• /
• pp.94-94
• /
• 2011

배관 구조물에서는 내부 미세 균열에서부터 국부 좌굴, 볼트 풀림, 피로 균열 등과 같이 다양한 형태의 손상이 복합적으로 발생 가능하다. 이러한 복합 손상은 배관 구조물의 누수, 누유 등의 사고를 야기할 수 있다. 하지만 기존의 단일 스케일 계측 시스템으로부터 복합 손상에 의한 실시간 누수를 진단하기는 매우 어렵다. 본 연구 단계에서는 누수를 야기하는 복합 손상을 효율적으로 진단하기 위하여 선행 연구에서 제안된 압전센서를 이용한 자가 계측 회로 기반의 다중 스케일 계측 시스템을 구조물의 복합 손상 진단에 적용하였다. 자가 계측 회로 기반 다중 스케일 계측 시스템은 크게 두 가지 형태의 신호를 계측한다. 첫 번째 스케일은 임피던스 계측으로부터 특정 주파수 대역폭에 대한 구조 응답을 계측하며, 두 번째 스케일은 유도 초음파 계측으로부터 단일 중심 주파수에 해당하는 구조물의 응답을 계측한다. 복합 손상을 손상 유형별로 분류하기 위하여 E/M 임피던스(Electro-mechanical impedance)및 유도 초음파(Guided wave) 계측으로부터 추출한 특성을 이용하여 2차원 손상지수를 계산하고 이를 지도학습 기반 패턴인식 기법(Supervised learning based pattern recognition) 중 확률론적 신경망 기법(Probabilistic Neural Network, PNN)에 적용한다. 제안된 기법의 적용성 검토를 위하여 파이프 구조물에 인위적으로 다중 손상을 생성시켜 시험을 수행하였다. 본 연구에서 제안된 기법이 실제 배관 구조물에 성공적으로 적용된다면 손상 부재의 거동 및 구조물 성능의 손상에 대한 영향을 효율적으로 진단하고 평가함으로써 배관 구조물의 효과적인 유지관리가 가능할 것으로 예상된다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.11
• /
• pp.2813-2826
• /
• 2012

Various stakeholders are involved in the creation of software projects. In general, the higher the number of stakeholders involved during the requirements elicitation phase, the better are the chances of success for the project. However, it is rather difficult to consider the opinion of all the stakeholders owing to constraints on time and resources. Furthermore, conflicts between stakeholders can become inevitable when the number of stakeholders increases. Thus, the identification of key stakeholders is an important factor in ensuring the success of a project. In this paper, a methodical stakeholder conflict resolution model (s-CRM) is proposed by considering an actual industrial case study. The proposed model uses information gain based on entropy when measuring the impurity of information. We believe that the proposed s-CRM is effective in identifying the key stakeholders and in intuitively indicating those stakeholders whose elicited requirements need to be weighted. In addition, the model provides a solution for conflicts among stakeholders during requirements engineering.