• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.3D
• /
• pp.409-416
• /
• 2006

Real-time traveler information disseminated through Variable Message Signs (VMS) is known to have effects on driver route choice decisions. In the past, many studies have attempted to optimize the system performance using VMS message content as the primary control variable of driver route choice. This research proposes a VMS information provision optimization model which searches the best combination of VMS message contents and display sequence to minimize the total travel time on a highway network considered. The driver route choice models under VMS information provision are developed using a stated preference (SP) survey data in order to realistically capture driver response behavior. The genetic algorithm (GA) is used to find the optimal VMS information provision strategies which consists of the VMS message contents and the sequence of message display. In the process of the GA module, the system performance is measured using micro traffic simulation. The experiment results highlight the capability of the proposed model to search the optimal solution in an efficient way. The results show that the traveler information conveyed via VMS can reduce the total travel time on a highway network. They also suggest that as the frequency of VMS message update gets shorter, a smaller number of VMS message contents performs better to reduce the total travel time, all other things being equal.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.536-543
• /
• 2023

Speaker diarization, which labels for "who spoken when?" in speech with multiple speakers, has been studied on a deep neural network-based end-to-end method for labeling on speech overlap and optimization of speaker diarization models. Most deep neural network-based end-to-end speaker diarization systems perform multi-label classification problem that predicts the labels of all speakers spoken in each frame of speech. However, the performance of the multi-label-based model varies greatly depending on what the threshold is set to. In this paper, it is studied a speaker diarization system using single-label classification so that speaker diarization can be performed without thresholds. The proposed model estimate labels from the output of the model by converting speaker labels into a single label. To consider speaker label permutations in the training, the proposed model is used a combination of Permutation Invariant Training (PIT) loss and cross-entropy loss. In addition, how to add the residual connection structures to model is studied for effective learning of speaker diarization models with deep structures. The experiment used the Librispech database to generate and use simulated noise data for two speakers. When compared with the proposed method and baseline model using the Diarization Error Rate (DER) performance the proposed method can be labeling without threshold, and it has improved performance by about 20.7 %.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.49 no.5
• /
• pp.77-83
• /
• 2012

The importance of mobility management is becoming to be one of the upcomming issues to be addressed to provide the converged services and the convergence of the heterogeneous network environments. In this paper, the new user mobility management scheme which can be utilized to model the user's mobility behaviors for interworking with heterogeneous overlay convergent networks under the time-varying radio propagation environment has been proposed. Thus user mobility management scheme based on user mobility model is considered in order to optimize the dwell time of users in the overlay convergent networks. This Mobile IP user mobility management will be very useful to model the user mobility behaviors and can be used to estimate the signaling traffic and frequency spectrum demands for massive data transfer for the heterogeneous overlay convergent networks.

• International Commerce and Information Review
• /
• v.15 no.3
• /
• pp.411-430
• /
• 2013

Overseas marketing before the conclusion of a contract is a huge obstacle to enter the global market because it has needed the company's maximum capacities which are some cost, time and human resources. Thus it required a reestablishment about the education fields and research schemes. Due to the advancements in ICT and internet, a subject of e-trade is becoming a critical issue with a subject of a practice of international trade. However, since the mid-2000s, e-trade research articles and educational materials such as textbooks, research papers are being gradually reduced. Therefore the purpose of this study, from an oversea marketing point of view among the various e-trade fields, is to measure an education performance and an academic research scheme. And this study will suggest direction of improvement about research scheme and educational performance in the overseas marketing. According to the result, to establish the education and research scheme about SEM(Search Engine Marketing), SEO(Search Engine Optimization) and SNA(Social Network Analysis) which are introduced in the industry among the education and research field related the e-trade is urgent. And some subjects need a capstone-design reconcile theory and practice.

• Journal of Korean Tunnelling and Underground Space Association
• /
• v.16 no.3
• /
• pp.321-339
• /
• 2014

This paper concerns the development of a knowledge-based tunnel design system within the framework of artifical neural networks(ANNs). The system is aimed at expediting a routine tunnel design works such as computation of segment lining body forces and stability analysis of selected cross section. A number of sub-modules for computation of segment lining body forces and stability analysis were developed and implemented to the system. It is shown that the ANNs trained with the results of 3D numerical analyses can be generalized with a reasonable accuracy, and that the ANN based tunnel design concept is a robust tool for tunnel design optimization. The details of the system architecture and the ANNs development are discussed in this paper.

• Journal of the Korean Society for Precision Engineering
• /
• v.29 no.12
• /
• pp.1376-1381
• /
• 2012

Optical devices are used extensively in the field of information network. Increasing demand for optical device, optical interconnection has been a important issue for commercialization. However many problems exist in the interconnection between optical device and optical fiber, and in the case of the multi-channel, problems of the optical alignment and optical array arise. For solving the alignment and array problem of optical device and the optical fiber, we fabricated fiber alignment and array by using imprint technology. Achieved higher precision of optical fiber alignment and array due to fabricating using imprint technology. The silicon stamp with different depth was fabricated using the conventional photolithography. Using the silicon stamp, a nickel stamp was fabricated by electroforming process. We conducted imprint process using the nickel stamp with different depth. The optical alignment and array by fabricating the patterns of optical device and fiber alignment and array using imprint process, and achieved higher precision of decreasing the dimensional error of the patterns by optimization of process. The fabricated optical interconnection of PLC device was measured 3.9 dB and 4.2 dB, lower than criteria specified by international standard.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.2617-2622
• /
• 2003

The pH neutralization process has long been taken as a representative benchmark problem of nonlinear chemical process control due to its nonlinearity and time-varying nature. For general nonlinear processes, it is difficult to control with a linear model-based control method so nonlinear controls must be considered. Among the numerous approaches suggested, the most rigorous approach is the dynamic optimization. However, as the size of the problem grows, the dynamic programming approach is suffered from the curse of dimensionality. In order to avoid this problem, the Neuro-Dynamic Programming (NDP) approach was proposed by Bertsekas and Tsitsiklis (1996). The NDP approach is to utilize all the data collected to generate an approximation of optimal cost-to-go function which was used to find the optimal input movement in real time control. The approximation could be any type of function such as polynomials, neural networks and etc. In this study, an algorithm using NDP approach was applied to a pH neutralization process to investigate the feasibility of the NDP algorithm and to deepen the understanding of the basic characteristics of this algorithm. As the global approximator, the neural network which requires training and k-nearest neighbor method which requires querying instead of training are investigated. The global approximator requires optimal control strategy. If the optimal control strategy is not available, suboptimal control strategy can be used even though the laborious Bellman iterations are necessary. For pH neutralization process it is rather easy to devise an optimal control strategy. Thus, we used an optimal control strategy and did not perform the Bellman iteration. Also, the effects of constraints on control moves are studied. From the simulations, the NDP method outperforms the conventional PID control.

• IMMUNE NETWORK
• /
• v.15 no.2
• /
• pp.83-90
• /
• 2015

Evaluation and screening of vaccines against tuberculosis depends on development of proper cost effective disease models along with identification of different immune markers that can be used as surrogate endpoints of protection in preclinical and clinical studies. The objective of the present study was therefore evaluation of subcutaneous model of M.tuberculosis infection along with investigation of different immune biomarkers of tuberculosis infection in BALB/c mice. Groups of mice were infected subcutaneously with two different doses : high ($2{\times}10^6CFU$) and low doses ($2{\times}10^2CFU$) of M.tuberculosis and immune markers including humoral and cellular markers were evaluated 30 days post M.tuberculosis infections. Based on results, we found that high dose of subcutaneous infection produced chronic disease with significant (p<0.001) production of immune markers of infection like $IFN{\gamma}$, heat shock antigens (65, 71) and antibody titres against panel of M.tuberculosis antigens (ESAT-6, CFP-10, Ag85B, 45kDa, GroES, Hsp-16) all of which correlated with high bacterial burden in lungs and spleen. To conclude high dose of subcutaneous infection produces chronic TB infection in mice and can be used as convenient alternative to aerosol models in resource limited settings. Moreover assessment of immune markers namely mycobacterial antigens and antibodies can provide us valuable insights on modulation of immune response post infection. However further investigations along with optimization of study protocols are needed to justify the outcome of present study and establish such markers as surrogate endpoints of vaccine protection in preclinical and clinical studies in future.

• Journal of Digital Contents Society
• /
• v.19 no.6
• /
• pp.1161-1167
• /
• 2018

The development of machine learning has enabled machines to perform delicate tasks that only humans could do, and thus many companies have introduced machine learning based translators. Existing translators have good performances but they have problems in number translation. The translators often mistranslate numbers when the input sentence includes a large number. Furthermore, the output sentence structure completely changes even if only one number in the input sentence changes. In this paper, first, we optimized a neural machine translation model architecture that uses bidirectional RNN, LSTM, and the attention mechanism through data cleansing and changing the dictionary size. Then, we implemented a number-processing algorithm specialized in number translation and applied it to the neural machine translation model to solve the problems above. The paper includes the data cleansing method, an optimal dictionary size and the number-processing algorithm, as well as experiment results for translation performance based on the BLEU score.

• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.11
• /
• pp.453-462
• /
• 2017

OpenDaylight is an SDN (Software Defined Networking) open source framework, which is popular in network fields recently. This paper analyzes the performance of a controller cluster architecture by focusing on distributed datastore and Raft leader election algorithm. In addition, we propose an enhanced version of Raft algorithm in order to improve the performance of distributed datastore by distributing shard leaders over controller cluster. This paper compares the conventional Raft algorithm with the proposed version of the Raft algorithm. Moreover, we compare the performance of distributed datastore according to shard roles such as leader and follower. Experimental results show that Shard leaders provide better performance than followers and Shard updating requests need to be distributed over multiple controllers. So, by using proposed version of Raft algorithm, controller performance can be improved. The details of the experiment results are cleary described.