• The Journal of the Acoustical Society of Korea
• /
• v.19 no.1
• /
• pp.4-10
• /
• 2000

The work presented in this paper is about a postprocessor, which improves the performance of automatic speech segmentation system by correcting the phoneme boundary errors. We propose a postprocessor that reduces the range of errors in the auto labeled results that are ready to be used directly as synthesis unit. Starting from a baseline automatic segmentation system, our proposed postprocessor trains the features of hand labeled results using multi-layer perceptron(MLP) algorithm. Then, the auto labeled result combined with MLP postprocessor determines the new phoneme boundary. The details are as following. First, we select the feature sets of speech, based on the acoustic phonetic knowledge. And then we have adopted the MLP as pattern classifier because of its excellent nonlinear discrimination capability. Moreover, it is easy for MLP to reflect fully the various types of acoustic features appearing at the phoneme boundaries within a short time. At the last procedure, an appropriate feature set analyzed about each phonetic event is applied to our proposed postprocessor to compensate the phoneme boundary error. For phonetically rich sentences data, we have achieved 19.9 % improvement for the frame accuracy, comparing with the performance of plain automatic labeling system. Also, we could reduce the absolute error rate about 28.6%.

• Journal of Navigation and Port Research
• /
• v.37 no.5
• /
• pp.493-498
• /
• 2013

This paper is aimed to priority evaluation of preliminary cases for IMO -IMS(International Maritime Organization- Information Management System) using fuzzy TOPSIS(Technique for Order Performance by Similarity to Ideal Solution) and AHP(Analytic Hierarchy Process). To this solve, therefore, this paper extract 24 preliminary cases and select 4 major preliminary alternative cases after analysing the structure of its alternative cases using FSM(Fuzzy Structure Modeling). Also, the weights of evaluation factors determine using AHP which able to keep the consistency when decision-makers assess. In AHP method, but, the numbers of paired comparison incerase as much as the numbers of the comparison items increase and because this evaluation have the many of vagueness, the decision of final ranking is used to fuzzy TOPSIS method which is included TOPSIS and Fuzzy Set Theory. The result are developed as order as Management of IMO Convention Information, Delivery of IMO Convention Information, Total IMO Database, Knowledge Hub of IMO Convention Information in IMO-IMS.

• The KIPS Transactions:PartD
• /
• v.13D no.6 s.109
• /
• pp.765-774
• /
• 2006

An association rule discovery, a technique to analyze the stored data in databases to discover potential information, has been a popular topic in stream data system. Most of the previous researches are concerned to single stream data. However, this approach may ignore in mining to multidimensional stream data. In this paper, we study the techniques discovering the association rules to multidimensional stream data. And we propose a AR-MS method reflecting the characteristics of stream data since make the summarization information by one data scan and discovering the association rules for significant rare data that appear infrequently in the database but are highly associated with specific event. Also, AR-MS method can discover the maximal frequent item of multidimensional stream data by using the summarization information. Through analysis and experiments, we show that AR-MS method is superior to other previous methods.

• The KIPS Transactions:PartD
• /
• v.17D no.5
• /
• pp.347-352
• /
• 2010

Automatic service identification and quality evaluation is one of key characteristics for a Service-Oriented Computing, being receiving a lot of attention from researchers in recent years. However, most researchers focus on identifying and evaluating application services and do not present methods for automatically identifying and evaluating business services from business processes. In general, the manual business service identification process by a human expert is a highly expensive and ambiguous task and may result in the service design with bad quality because of errors and misconception. We propose an automatic business service identification and quality evaluation method using Enterprise Architecture as a machine understandable knowledge-base. We verify the effectiveness of the proposed method through a case study on Department of Defense Enterprise Architecture.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.28 no.2
• /
• pp.189-196
• /
• 2010

Spatial information and smart phone technology have made innovative improvement of daily life. Spatial and geographic information are in practice for various applications. Especially, spatial information along with information and telecommunication technology could create new contents for providing services for convenient daily life. Spatial information technology, recently, is not only for acquiring location and attribute data but also providing tools to extract information and knowledge systematically for decision making. Various indoor applications have emerged in accordance with demands on daily GIS(Geographic information system). This paper aims for applying spatial information technology to support decision-making in shopping. The main contents include product database, optimal path search, shopping time expectation, automatic housekeeping book generation and analysis. Especially for foods, function to analyze information of the nutrition facts could help to improve dietary pattern and well-being. In addition, this system is expected to provide information for preventing overconsumption and impulse purchase could help economical and effective purchase pattern by analyzing propensity to consume.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.2
• /
• pp.27-34
• /
• 2010

Recently, lots of internet service companies provide on-line health diagnosis systems. But general persons not having expert knowledge are difficult to use, because most of the health diagnosis systems present prescriptions or dietetic treatments for diseases based on western medicine. In this paper, a self health diagnosis system of oriental medicine coinciding with physical characteristics of Korean using fuzzy ART algorithm, is proposed. In the proposed system, three high rank of diseases having high similarity values are derived by comparing symptoms presented by a user with learned symptoms of specific diseases based on treatment records using neural networks. And also the proposed system shows overall symptoms and folk remedies for the three high rank of diseases. Database on diseases and symptoms is built by several oriental medicine books and then verified by a medical specialist of oriental medicine. The proposed self health diagnosis system of oriental medicine showed better performance than conventional health diagnosis systems by means of learning diseases and symptoms using treatment records.

• Journal of Korea Technology Innovation Society
• /
• v.15 no.2
• /
• pp.369-399
• /
• 2012

Purpose: Health technology research & development is an important area to leading future. This study examined the current trends for 'oncogene' based on the research subject network to deduce a research front. Method: Papers were extracted from PubMed database using MeSH term for studies on 'oncogenes' and further categorized as papers published by Korean. Keywords were collected from all of articles. Research subject network was generated by keywords. Research subject network was analyzed by weighted degree centrality based social network analysis and transition of research subjects was analyzed by the time series. Results: On 'oncogenes', 'Genes, ras', 'Apoptosis', 'Signal Transduction' had a high degree centrality and currently 'Antineoplastic Agents', 'Prognosis', and 'Tumor Markers, Biological' were widely conducted. Conclusion: Consistency of research trend pattern was found by analyzing oncogene network with compromised to international vs. domestic trends. Analyzing keyword networks in various subject area, those will allow us to predict the research progress and propose evidence of research & developmental strategy.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.23 no.2
• /
• pp.109-132
• /
• 2012

This study examines the social network of scholars in the field of communication by using author cocitation data. A matrix containing the number of cocited documents between pairs of authors is created for social network analysis of scholars who are on the editorial board of Journal of Communication, and the networked map of the scholars is used to visualize the knowledge structure of the field by identifying groups of authors who are more central than others. In addition, the study compares the previous analysis performed in 2008 and the current analysis on the editorial board of the journal, which increased from 146 to 254 scholars in numbers. Author cocitation data was collected using Social Science Citation Index (SSCI) through the Web of Science database, and UCInet was used to create and visualize the author cocitation network and to analyze the correlation between the cocitation network and the factors that may have affected the structure of the cocitation network.

• Journal of Korean Association for Spatial Structures
• /
• v.5 no.3 s.17
• /
• pp.75-82
• /
• 2005

The conceptual structural design consists of selecting structural material and form of the building, producing a preliminary dimensional layout. The information such as height of the building use, typical live load, wind velocity, design acceleration, maximum lateral deflection, span, story height is a important factor in conceptual design phase. In this case, the knowledge solutions for past similar problems cam be used in the process of defining and finding a solution to the design problems. In this paper, the conceptual structural design method using case-based reasoning which is intended to assist engineers in the conceptual phase of the structural design of tall buildings is introduced. Inductive retrieval method and nearest-neighbor retrieval method are used for selecting structural system and similar design case, respectively.

• Geotechnical Engineering
• /
• v.10 no.3
• /
• pp.79-96
• /
• 1994

A tunnel design expert system entitled NESTED is developed using the artificial neural network. The expert system includes three neural network computer models designed for the stability assessment of underground openings and the estimation of correlation between the RMR and Q systems. The expert system consists of the three models and the computerized rock mass classification programs that could be driven under the same user interface. As the structure of the neural network, a multi -layer neural network which adopts an or ror back-propagation learning algorithm is used. To set up its knowledge base from the prior case histories, an engineering database which can control the incomplete and erroneous information by learning process is developed. A series of experiments comparing the results of the neural network with the actual field observations have demonstrated the inferring capabilities of the neural network to identify the possible failure modes and the support timing. The neural network expert system thus complements the incomplete geological data and provides suitable support recommendations for preliminary design of tunnels in rock masses.