ACIS International Conference on Computer and Information Science, 2005
Rational Unified Process (RUP) provides a component-based development process which is use-case d... more Rational Unified Process (RUP) provides a component-based development process which is use-case driven, architecture-centric, and iterative and incremental. This paper describes our experience of RUP application to the development of a web-based project management system, which had been peformed by a five-member team for one year. The paper introduces how we applied RUP in the development of the system and
Pacific Rim International Conference on Artificial Intelligence, 2002
The MLP-based speaker verification system can have to provide fast speaker enrollment process as ... more The MLP-based speaker verification system can have to provide fast speaker enrollment process as well as high speaker recognition rate and quick speaker verification process. The multilayer perceptron (MLP) presents higher pattern recognition rate without assuming underlying density distribution against the existing parametric pattern recognition methods, and enables rapid operation by sharing internal parameters between various models. Among a variety
We present a novel approach for computing link-based similarities among objects accurately by uti... more We present a novel approach for computing link-based similarities among objects accurately by utilizing the link information pertaining to the objects involved. We discuss the problems with previous link-based similarity measures and propose a novel approach for computing link based similarities that does not suffer from these problems. In the proposed approach each target object is represented by a vector. Each element of the vector corresponds to all the objects in the given data, and the value of each element denotes the weight for the corresponding object. As for this weight value, we propose to utilize the probability of reaching from the target object to the specific object, computed using the "Random Walk with Restart" strategy. Then, we define the similarity between two objects as the cosine similarity of the two vectors. In this paper, we provide examples to show that our approach does not suffer from the aforementioned problems. We also evaluate the performance o...
Data broadcasting is an efficient method for disseminating data, and is widely accepted in the da... more Data broadcasting is an efficient method for disseminating data, and is widely accepted in the database applications of mobile computing environments because of its asymmetric communication bandwidth between a server and mobile clients. This requires new types of concurrency control mechanism to support mobile transactions executed in the mobile clients, which have low-bandwidths toward the server. In this paper, we propose an OCC/DTA (Optimistic Concurrency Control with Dynamic Timestamp Adjustment) protocol that can be efficiently adapted to mobile computing environments. The protocol reduces communication overhead by using client-side validation procedure and enhances transaction throughput by adjusting serialization order without violating transaction semantics. We show that the proposed protocol satisfies data consistency requirements, and simulate that this protocol can improve the performance of mobile transactions in data broadcasting environments.
Allen & Koomen' s interval planner and Dean & McDermott's time map manager (TMM) offer ... more Allen & Koomen' s interval planner and Dean & McDermott's time map manager (TMM) offer different approaches to temporal database management in planning. In this paper we present a temporal planning system that integrates ideas from both methods, and at ...
Performing a good component selection plays a critical role in the success of the final system. A... more Performing a good component selection plays a critical role in the success of the final system. Although the history of component selection process in component based software development is almost a decade old, we have found that no selection process addresses the use of previous decision experience for selecting components for similar requirements. In this research we argue that previous
5th IEEE/ACIS International Conference on Computer and Information Science and 1st IEEE/ACIS International Workshop on Component-Based Software Engineering,Software Architecture and Reuse (ICIS-COMSAR'06), 2006
Having a variety of good characteristics against other pattern recognition techniques, the multil... more Having a variety of good characteristics against other pattern recognition techniques, the multilayer perceptron (MLP) has been used in many applications. But, it is known that the error backpropagation (EBP) algorithm that the MLP uses in learning has the defect that requires relatively long learning time. In order to increase learning speed it is very effective to use the online-based learning methods, which update the weight vector of the MLP pattern by pattern, because the learning data for pattern recognition contain high redundancy. A typical online EBP algorithm applies the fixed learning rate for each update of the weight vector. Though a large amount of speedup with the online EBP can be obtained just by choosing the appropriate fixed rate, fixing the rate has the inefficiency that doesn’t fully utilize the instant updates of the online mode. And, although the patterns come to be divided into the learned and the unlearned during learning process and the learned have no need...
ACIS International Conference on Computer and Information Science, 2005
Rational Unified Process (RUP) provides a component-based development process which is use-case d... more Rational Unified Process (RUP) provides a component-based development process which is use-case driven, architecture-centric, and iterative and incremental. This paper describes our experience of RUP application to the development of a web-based project management system, which had been peformed by a five-member team for one year. The paper introduces how we applied RUP in the development of the system and
Pacific Rim International Conference on Artificial Intelligence, 2002
The MLP-based speaker verification system can have to provide fast speaker enrollment process as ... more The MLP-based speaker verification system can have to provide fast speaker enrollment process as well as high speaker recognition rate and quick speaker verification process. The multilayer perceptron (MLP) presents higher pattern recognition rate without assuming underlying density distribution against the existing parametric pattern recognition methods, and enables rapid operation by sharing internal parameters between various models. Among a variety
We present a novel approach for computing link-based similarities among objects accurately by uti... more We present a novel approach for computing link-based similarities among objects accurately by utilizing the link information pertaining to the objects involved. We discuss the problems with previous link-based similarity measures and propose a novel approach for computing link based similarities that does not suffer from these problems. In the proposed approach each target object is represented by a vector. Each element of the vector corresponds to all the objects in the given data, and the value of each element denotes the weight for the corresponding object. As for this weight value, we propose to utilize the probability of reaching from the target object to the specific object, computed using the "Random Walk with Restart" strategy. Then, we define the similarity between two objects as the cosine similarity of the two vectors. In this paper, we provide examples to show that our approach does not suffer from the aforementioned problems. We also evaluate the performance o...
Data broadcasting is an efficient method for disseminating data, and is widely accepted in the da... more Data broadcasting is an efficient method for disseminating data, and is widely accepted in the database applications of mobile computing environments because of its asymmetric communication bandwidth between a server and mobile clients. This requires new types of concurrency control mechanism to support mobile transactions executed in the mobile clients, which have low-bandwidths toward the server. In this paper, we propose an OCC/DTA (Optimistic Concurrency Control with Dynamic Timestamp Adjustment) protocol that can be efficiently adapted to mobile computing environments. The protocol reduces communication overhead by using client-side validation procedure and enhances transaction throughput by adjusting serialization order without violating transaction semantics. We show that the proposed protocol satisfies data consistency requirements, and simulate that this protocol can improve the performance of mobile transactions in data broadcasting environments.
Allen & Koomen' s interval planner and Dean & McDermott's time map manager (TMM) offer ... more Allen & Koomen' s interval planner and Dean & McDermott's time map manager (TMM) offer different approaches to temporal database management in planning. In this paper we present a temporal planning system that integrates ideas from both methods, and at ...
Performing a good component selection plays a critical role in the success of the final system. A... more Performing a good component selection plays a critical role in the success of the final system. Although the history of component selection process in component based software development is almost a decade old, we have found that no selection process addresses the use of previous decision experience for selecting components for similar requirements. In this research we argue that previous
5th IEEE/ACIS International Conference on Computer and Information Science and 1st IEEE/ACIS International Workshop on Component-Based Software Engineering,Software Architecture and Reuse (ICIS-COMSAR'06), 2006
Having a variety of good characteristics against other pattern recognition techniques, the multil... more Having a variety of good characteristics against other pattern recognition techniques, the multilayer perceptron (MLP) has been used in many applications. But, it is known that the error backpropagation (EBP) algorithm that the MLP uses in learning has the defect that requires relatively long learning time. In order to increase learning speed it is very effective to use the online-based learning methods, which update the weight vector of the MLP pattern by pattern, because the learning data for pattern recognition contain high redundancy. A typical online EBP algorithm applies the fixed learning rate for each update of the weight vector. Though a large amount of speedup with the online EBP can be obtained just by choosing the appropriate fixed rate, fixing the rate has the inefficiency that doesn’t fully utilize the instant updates of the online mode. And, although the patterns come to be divided into the learned and the unlearned during learning process and the learned have no need...
2008 Third International Conference on Convergence and Hybrid Information Technology, 2008
The purpose of this research is to solve the problem of time log processing. Processing the activ... more The purpose of this research is to solve the problem of time log processing. Processing the activity time is an overhead for the personal software process (PSP) users. We have proposed an approach to address the limitations arising from extracting the PSP time log from the time log generated by speech recognition sensor. Our approach categorizes the problem on the
Uploads
inproceedings by Ho-jin Choi
Papers by Ho-jin Choi