• Speech Sciences
• /
• v.9 no.1
• /
• pp.39-47
• /
• 2002

This paper proposes an effective method to remedy the acoustic modeling problem inherent in the usual log-normal Parallel Model Composition intended for achieving robust speech recognition. In particular, the Gaussian kernels under the prescribed log-normal PMC cannot sufficiently express the corrupted speech distributions. The proposed scheme corrects this deficiency by judiciously selecting the 'fairly' corrupted component and by re-estimating it as a mixture of two distributions using data-driven PMC. As a result, some components become merged while equal number of components split. The determination for splitting or merging is achieved by means of measuring the similarity of the corrupted speech model to those of the clean model and the noise model. The experimental results indicate that the suggested algorithm is effective in representing the corrupted speech distributions and attains consistent improvement over various SNR and noise cases.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.16 no.2
• /
• pp.128-147
• /
• 1991

We propose a dynamic incidence matrix (DIM) for reflecting states and time conditions of a timed Petri net (TPN) explicitly. Since a DIM consists of a conventional incidence matrix, two time-related vectors and two state-related vectors, we can get the advantages inherent in the conventional incidence matrix of describing a static structure of a system as well as another advantage of expressing time dependent state transitions. We introduce an algorithm providing the DIM with a state transition mechanism. Because the algorithm is, in fact, an algorithmic model for discrete event simulation of TPN models, we provide a theoretical basis of model transformation of a TPN model into a DEVS(Discrete Event system Specification) model. By executing the algorithm we can carry out performance analysis of computer communication protocols which are represented TPN models.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.1101-1104
• /
• 2002

In order to send the information certainly via the network against the packet lost caused by hardware troubles or limitation of packet transferring, we must construct reliable network infrastructure. However, it is difficult to construct comfortable network early if we construct rely on the prediction or the experience through a lot of troubles. In this paper, we propose the method to construct reliable network infrastructure based on the computer network simulation. This simulation is based on the percolation model. Percolation model is known as the model that represents connections. We gave some simulations for the various network topologies: the square lattice network, the cubic lattice network, and the full connection type network.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.8
• /
• pp.101-106
• /
• 2017

In this paper, we propose a new teaching-learning model with two-step feedback on programming language learning, which is a basic preliminary learning for programming. Programming learning is aimed at improving problem solving skills and thinking by experiencing problem solving through programming. For programming, the learner must know how to work with the computer and what to do with it. To do this, concrete thinking should be established and described in an accurate programming language. In recent, most studies have focused on the effects of programming learning and have not studied the effects of education on language itself. Therefore, in this study, the teaching-learning model for programming language education is presented and applied to the field, and the results are compared with the existing instructional-teaching model.

• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.539-541
• /
• 1990

A new neural network model, based on the Hopfield's crossbar associative network, is presented and shown to be an effective tool for the NP-Complete problems. This model is applied to a class of layer assignment problems for VLSI routing. The results indicate that this modified Hopfield model improves stability and accuracy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.10
• /
• pp.1770-1782
• /
• 2011

A brain-computer interface (BCI) is a communication system that translates brain activity into commands for computers or other devices. In other words, BCIs create a new communication channel between the brain and an output device by bypassing conventional motor output pathways consisting of nerves and muscles. This is particularly useful for facilitating communication for people suffering from paralysis. Due to the low bit rate, it takes much more time to translate brain activity into commands. Especially it takes much time to input characters by using BCI-based typewriters. In this paper, we propose a brain-operated typewriter which is accelerated by a language prediction model. The proposed system uses three kinds of strategies to improve the entry speed: word completion, next-syllable prediction, and next word prediction. We found that the entry speed of BCI-based typewriter improved about twice as much through our demonstration which utilized the language prediction model.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.214-216
• /
• 2005

디지털 방송이 시작되면서 시청자가 선택할 수 있는 채널은 200여 개로 늘어났다. 지금처럼 리모컨으로 채널을 돌려가며 보거나 원하는 TV 프로그램을 찾기란 거의 불가능해진 것이다. 이러한 다채널 다매체 시대에 원하는 프로그램 시청을 도와줄 수 있는 프로그램 가이드 시스템의 필요성이 증가하게 되었고, 더 나아가 TV를 시청하는 각 개인의 선호도를 반영하는 것이 요구되었다. 본 논문에서는 r-order Markov Model을 이용한 개인화된 전자 TV 프로그램 추천 시스템을 제안한다. Markov Model은 시간이 지남에 따라 시청하는 프로그램의 변화를 모델링하기 위한 방법으로 사용하였다. 이 시스템은 시청자의 선호 프로그램을 예측하기 위해서 r-order Markov Model을 제안하는 것뿐만 아니라 TV 시청자의 프로그램 선호를 예측하기 위한 모델들을 적용하였다. 실험 결과는 Markov Model이 추천에 대한 높은 정확성을 제공할 수 있다는 것을 보여준다.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.12
• /
• pp.219-226
• /
• 2016

In this paper, we propose an efficient repository model for online software education. The software education of app development consists of 7 stages: coding & debugging, submit, collaboration, review, validate, deployment, certification. Proposed repository model supports all 7 stages efficiently. In the coding & debugging stage, the students repeat coding and debugging of source. In the submit stage, the output of previous process such as source codes, project, and videos, are uploaded to repository server. In the collaboration stage, other students or experts can optimize or upgrade version of source code, project, and videos stored in the repository. In the review stage, mentors can review and send feedbacks to students. In the validate stage, the specialists validate the source code, project, and the videos. In the deployment stage, the verified source code, project, and videos are deployed. In the certification stage, the source code, project, and the videos are evaluated to issue the certificate.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.9
• /
• pp.81-86
• /
• 2018

Software defect severity is very important in projects with limited historical data or new projects. But general software defect prediction is very difficult to collect the label information of the training set and cross-project defect prediction must have a lot of data. In this paper, an unclassified data set with defect severity is clustered according to the distribution ratio. And defect severity-based prediction model is proposed by way of labeling. Proposed model is applied CLAMI in JM1, PC4 with the least ambiguity of defect severity-based NASA dataset. And it is evaluated the value of ACC compared to original data. In this study experiment result, proposed model is improved JM1 0.15 (15%), PC4 0.12(12%) than existing defect severity-based prediction models.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4763-4775
• /
• 2020

This research aims to develop a deep learning-based method that automatically detects and segments the defects on railroad surfaces to reduce the cost of visual inspection of the railroad. We developed our segmentation model by modifying a fully convolutional network model [1], a well-known segmentation model used for machine learning, to detect and segment railroad surface defects. The data used in this research are images of the railroad surface with one or more defect regions. Railroad images were cropped to a suitable size, considering the long height and relatively narrow width of the images. They were also normalized based on the variance and mean of the data images. Using these images, the suggested model was trained to segment the defect regions. The proposed method showed promising results in the segmentation of defects. We consider that the proposed method can facilitate decision-making about railroad maintenance, and potentially be applied for other analyses.