• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.426-429
• /
• 1999

Many similar class components are stored in object-storage but the object-storage has needed the retrieval function of correct component for reuse. Accordingly this paper designed the class component retrieval viewer of the object-storage by using the improved spreading activation strategy. Object-storage has made up of information of inheritance relation, superclass, subclass, and we defined the queries about each class function. Also we specified connectionist relaxation of the each class and query, finally we gained retrieval result which showed highest activation value order of class component information including the query function.

• Proceedings of the Korean Society for Cognitive Science Conference
• /
• 2010.05a
• /
• pp.2-7
• /
• 2010

Formally, we may define cognitive science as the convergent study between symbolic and connectionist approaches at macro and micro levels. Since what we refer to as the human mind is regarded as a mathematical product of the human brain and the computing machine, we can obtain two mathematical dynamical projections: one from the set of human brains to the set of mind, the other from the set of computing machines to the set of mind. Then, we are having a new projection from the classical models to the quantum mind.

• Proceedings of the Korean Information Science Society Conference
• /
• 1998.10c
• /
• pp.3-5
• /
• 1998

본 연구팀에서는 기존의 기호주의 전문가 시스템의 경우 지식표현 체계가 의미구조를 반영하지 못함으로써 발생하는 경직성문제를 해결하기 위해 CSN(Connectionist Semantic Network) 모델을 제안하였다. 그러나 CSN모델은 상위개념간의 관계를 표현하기 위해 단순한 전향 신경망을 사용함으로써 상위개념간의 일반적이고 구조화된 지식표현 및 추론에 어려움이 있었다. CSN 모델의 이런 문제점을 위해 본 논문에서는 상위개념간의 일반적이고 구조화된 지식표현과 추론이 용이한 기호주의 표현 체계와 이 표현 체계 안에 효과적으로 의미구조를 반영할 수 있는 연결주의 학습 모델인 CSN을 결합한 하이브리드 구조를 제안하고, 실험을 통하여 제안된 하이브리드 구조의 타당성을 보인다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.1 no.1
• /
• pp.9-25
• /
• 1991

This tutorial paper has been written for biologists, physicians or beginners in fuzzy sets theory and applications. This field is introduced in the framework of medical diagnosis problems. The paper describes and illustrates with practical examples, a general methodology of special interest in the processing of borderline cases, that allows a graded assignment of diagnoses to patients. A pattern of medical knowledge consists of a tableau with linguistic entries or of fuzzy propositions. Relationships between symptoms and diagnoses are interpreted as labels of fuzzy sets. It is shown how possibility measures (soft matching) can be used and combined to derive diagnoses after measurements on collected data. The concepts and methods are illustrated in a biomedical application on inflammatory protein variations. In the case of poor diagnostic classifications, it is introduced appropriate ponderations, acting on the characterizations of proteins, in order to decrease their relative influence. As a consequence, when pattern matching is achieved, the final ranking of inflammatory syndromes assigned to a given patient might change to better fit the actual classification. Defuzzification of results (i.e. diagnostic groups assigned to patients) is performed as a non fuzzy sets partition issued from a "separating power", and not as the center of gravity method commonly employed in fuzzy control. It is then introduced a model of fuzzy connectionist expert system, in which an artificial neural network is designed to build the knowledge base of an expert system, from training examples (this model can also be used for specifications of rules in fuzzy logic control). Two types of weights are associated with the connections: primary linguistic weights, interpreted as labels of fuzzy sets, and secondary numerical weights. Cell activation is computed through MIN-MAX fuzzy equations of the weights. Learning consists in finding the (numerical) weights and the network topology. This feed forward network is described and illustrated in the same biomedical domain as in the first part.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.6
• /
• pp.1692-1705
• /
• 2024

In this paper, we present a method that integrates a Grammar Transducer as an external language model to enhance the accuracy of the pre-trained Korean End-to-end (E2E) Automatic Speech Recognition (ASR) model. The E2E ASR model utilizes the Connectionist Temporal Classification (CTC) loss function to derive hypothesis sentences from input audio. However, this method reveals a limitation inherent in the CTC approach, as it fails to capture language information from transcript data directly. To overcome this limitation, we propose a fusion approach that combines a clause-level n-gram language model, transformed into a Weighted Finite-State Transducer (WFST), with the E2E ASR model. This approach enhances the model's accuracy and allows for domain adaptation using just additional text data, avoiding the need for further intensive training of the extensive pre-trained ASR model. This is particularly advantageous for Korean, characterized as a low-resource language, which confronts a significant challenge due to limited resources of speech data and available ASR models. Initially, we validate the efficacy of training the n-gram model at the clause-level by contrasting its inference accuracy with that of the E2E ASR model when merged with language models trained on smaller lexical units. We then demonstrate that our approach achieves enhanced domain adaptation accuracy compared to Shallow Fusion, a previously devised method for merging an external language model with an E2E ASR model without necessitating additional training.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.6 no.4
• /
• pp.238-242
• /
• 2013

We consider the problem of constructing observers for nonlinear systems with unknown inputs. Connectionist networks, also called neural networks, have been broadly applied to solve many different problems since McCulloch and Pitts had shown mathematically their information processing ability in 1943. In this thesis, we present a genetic neuro-control scheme for nonlinear systems. Our method is different from those using supervised learning algorithms, such as the backpropagation (BP) algorithm, that needs training information in each step. The contributions of this thesis are the new approach to constructing neural network architecture and its training.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1997.10a
• /
• pp.45-49
• /
• 1997

In this paper, we controlled a Rotary Inverted Pendulum System using Neuro-Fuzzy Controller(NFC). The inverted pendulum system is widely used as a typical example of an unstable nonlinear control system which is difficult to control. Fuzzy theory have been because membership functions and rules of a fuzzy controller are often given by experts or a fuzzy logic control system. This controller is a feedforward multilayered network which integrates the basic elements and functions of a tradtional fuzzy logic controller into a connectionist structure which has distributed learning abilities. Such NFC can be constructed from training examples by learning rule, and the structure can be trained to develop fuzzy logic rules and find optimal input/output membership functions. Using this controller, we presented the results that controlled a Rotary Inverted Pendulum System and the associated algorithms.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.5
• /
• pp.812-820
• /
• 1995

This paper proposes an approach of identifying nonlinear mappings from input/output data. The approach is based on the universal approximation by the fuzzy integration of local affine mappings. A connectionist model realizing the universal approximator is suggested by using a processing unit based on both the radial basis function and the weighted sum scheme. In addition, a learning method with self-organizing capability is proposed for the identifying of nonlinear mapping relationships with the given input/output data. To show the effectiveness of our approach, the proposed model is applied to the function approximation and the prediction of Mackey-Glass chaotic time series, and the performances are compared with other approaches.

• Proceedings of the KAIS Fall Conference
• /
• 2006.05a
• /
• pp.316-319
• /
• 2006

Connectionist networks, also called neural networks, have been broadly applied to solve many different problems since McCulloch and Pitts had shown mathematically their information processing ability in 1943. In this thesis, we present a genetic neuro-control scheme for nonlinear systems. Our method is different from those using supervised learning algorithms, such as the backpropagation (BP) algorithm, that needs training information in each step. The contributions of this thesis are the new approach to constructing neural network architecture and its training.

• Proceedings of the KSME Conference
• /
• 2004.04a
• /
• pp.1183-1188
• /
• 2004

The inverse heat conduction problem (IHCP) is a problem of estimating boundary condition from temperature measurement at one or more interior points. Neural networks are general information processing systems inspired by the connectionist theory of human brain. By properly training the network by the learning rule, the neural network method can handle many non-linear or other complex problems. In this work, neural network is applied to complicated inverse heat conduction problems. Efficiency of the procedure is enhanced by incorporating the radial basis functions (RBF). The RBF is trained faster than other neural network and can find smooth solution. In order to demonstrate the effectiveness of the current scheme, a typical one-dimensional IHCP is considered. At one surface, the temperature as well as the heat flux is known. The unknown temperature of interest is estimated on the other side of the slab. The results from the proposed method based on RBF neural network are compared with the conventional method.