• Title/Summary/Keyword: 탐색과정

Search Result 3,568, Processing Time 0.026 seconds

Automatic Recognition of Korean Broadcast News Using Flexible Vocabulary Recognition Models (가변 어휘 인식 모델을 이용한 한국어 방송 뉴스 음성의 인식)

  • 유하진
    • Proceedings of the Acoustical Society of Korea Conference
    • /
    • 1998.08a
    • /
    • pp.70-73
    • /
    • 1998
  • 본 논문에서는 한국어 방송 뉴스 인식 시스템에 관하여 기술한다. 인식 실험 과정에서는 실제로 방송된 음성을 인식하였으나, 인식을 위한 음향 모델은 본 연구소에서 갭라한 고립단어 인식용 가변 어휘 인식모델을 이용하였다. 가변 어휘 인식기는 방송 음성의 연속 문장을 이용하지 않고, 음향학적으로 고르게 분포된 고립 단어를 이용하여 학습되었다. 본 연구에서는 한국어의 특성상 문장이 영어권과 같이 단어 단위가 아닌 어절로 나누어 지는 점을 고려하여, 다양한 형태의 사전 표제어를 대상으로 실험하였다. 또한 탐색과정의 초기단계에 장거리 언어모델을 사용함으로써 인식 오류를 줄일 수 있었다.

  • PDF

A Comparison of the Form-Finding Method of Tensegrity Structures (텐세그리티 구조물의 형상탐색 기법 비교)

  • Lee, Seunghye;Lee, Jaehong
    • Journal of the Computational Structural Engineering Institute of Korea
    • /
    • v.27 no.4
    • /
    • pp.313-320
    • /
    • 2014
  • A tensegrity structure consists of a set of continuous cables in tension and a set of discontinuous struts in compression. The tensegrity structure can be classified into self-stressed and pre-stressed pin-jointed structure. A key step in the design of tensegrity structures is the determination of their equilibrium configuration, known as form-finding. In this paper, three effective methods are presented for form-finding of tensegrity structures. After performing form-finding process, a set of force density and corresponding topology results can be obtained. Then the force density method combined with a genetic algorithm is adopted to uniquely define a single integral feasible set of force densities. Numerical examples are presented that demonstrate the excellent performance of the algorithms.

A Study of Information Searching Behaviors Caused by Time pressure of Researchers (연구자의 시간압박감에 따른 정보탐색행태에 관한 연구)

  • Hong Ki-Churl
    • Journal of the Korean Society for Library and Information Science
    • /
    • v.31 no.3
    • /
    • pp.209-237
    • /
    • 1997
  • The main purpose of this study is to find the difference of information searching behaviours consist of information pursuit cognition and searching character by time pressure of researchers. The majar findings and result by time pressure of researchers are summarized as follows : In terms of information retrieval systems, according to the time pressure, there is a significant difference in recall, precision and diversity. In terms of main used material, there is a significant difference in a book, a research report, and a Journal. In terms of resource types, there is a significant difference in both record and human resource. Also, in terms of searching behaviours, there is a difference to spend time for information collection and information searching strategy, and to decide information searching method and degree of searching.

  • PDF

The Methods for Improvement of MPEG Picture Quality using the Characteristics of Pixels in Block and Inter-Block Correlations (블록내 화소특성 및 블록간 상관성을 이용한 MPEG 화질 개선 방법)

  • Kyeong Hwan Lee
    • Journal of Korea Multimedia Society
    • /
    • v.5 no.1
    • /
    • pp.28-37
    • /
    • 2002
  • In this paper, we propose new methods used in MPEG codec that can improve the picture quality degradations of the reconstructed images, such as blocking artifacts, without additional computations. At first, a new 2-step MAD motion search is presented in the motion search and compensation course to reduce the computations and the errors of block boundary pixels. And we also present an overlapped 2-step MAD motion search that use the pixels beyond the block boundary using the reduced computations. In DCT/quantization course for intra-blocks, 1-D DCT predictive quantization and pixel difference predictive quantization, those use the adjacent pixel sets of the Previously reconstructed blocks and improve the inter-block continuity, are presented. As simulation results, proposed methods shows better picture qualities by reducing blocking artifacts than that of the conventional MPEG.

  • PDF

Multi-Level Motion Estimation Algorithm Using Motion Information in Blocks (블록 내의 움직임 정보를 이용한 다단계 움직임 예측 알고리즘)

  • Heak Bong Kwon
    • Journal of Korea Multimedia Society
    • /
    • v.6 no.2
    • /
    • pp.259-266
    • /
    • 2003
  • In this paper, we propose a multi-level block matching algorithm using motion information in blocks. In the proposed algorithm, the block-level is decided by the motion degree in the block before motion searching procedure, and then adequate motion searching performs according to the block-level. This improves computational efficiency by eliminating the unnecessary searching Process in no motion or low motion regions, and brings more accurate estimation results by deepening motion searching Process in high motion regions. Simulation results show that the proposed algorithm brings the lower estimation error about 20% MSE reduction with the fewer blocks pet frame and the operation number was reduced to 56% compared to TSSA and 98% compared to FS -BMA with constant block size.

  • PDF

Analysis of Characteristics of Scientific Inquiry Problem Finding Process in Small Group Free Inquiry (소집단 자유 탐구에서 과학적 탐구 문제 발견 과정의 특징 분석)

  • Cheon, Myeongki;Lee, Bongwoo
    • Journal of The Korean Association For Science Education
    • /
    • v.38 no.6
    • /
    • pp.865-874
    • /
    • 2018
  • The purpose of this study is to explore the process of inquiry problem finding in high school students' small group free-inquiry. For this purpose, 91 second grade high school students took part in small group free-inquiry. We conducted interviews with students (48 students in 15 groups) who were relatively successful in the inquiry performed for one semester (about 4 months). Based on the results of the interviews, we analyzed the characteristics of the inquiry problem finding through the steps and strategies in the inquiry problem finding process. The main results are as follows: First, in the inquiry problem finding process, steps such as selecting keyword, presenting an inconvenience, presenting a question, and finding an inquiry problem were found, and in particular, the process of selecting the keyword that correspond to the subject of inquiry, such as the material and situation of inquiry, is very important step in inquiry problem finding. Second, the strategies that students used in the process of finding inquiry problem included searching information, review of prior research, sharing of knowledge and experience, linking and extension of knowledge and experience, environmental awareness, expert consultation, discussion of suitability, elaboration, etc. Third, finding an inquiry problem was relatively easy in the inquiry for finding out problems (i.e. inconvenience) in everyday life and investigating ways to solve them. Fourth, the review of prior researches through the internet was useful in the process of selecting keyword and elaboration. Fifth, the factors that students consider when selecting one of several candidate inquiry problems are feasibility, real-life applicability, and economic condition. Sixth, the current affairs had a positive impact on the inquiry problem finding. Based on the above results, we discussed some ways to increase students' inquiry problem finding ability.

Assessment of Landslide Susceptibility in Jecheon Using Deep Learning Based on Exploratory Data Analysis (데이터 탐색을 활용한 딥러닝 기반 제천 지역 산사태 취약성 분석)

  • Sang-A Ahn;Jung-Hyun Lee;Hyuck-Jin Park
    • The Journal of Engineering Geology
    • /
    • v.33 no.4
    • /
    • pp.673-687
    • /
    • 2023
  • Exploratory data analysis is the process of observing and understanding data collected from various sources to identify their distributions and correlations through their structures and characterization. This process can be used to identify correlations among conditioning factors and select the most effective factors for analysis. This can help the assessment of landslide susceptibility, because landslides are usually triggered by multiple factors, and the impacts of these factors vary by region. This study compared two stages of exploratory data analysis to examine the impact of the data exploration procedure on the landslide prediction model's performance with respect to factor selection. Deep-learning-based landslide susceptibility analysis used either a combinations of selected factors or all 23 factors. During the data exploration phase, we used a Pearson correlation coefficient heat map and a histogram of random forest feature importance. We then assessed the accuracy of our deep-learning-based analysis of landslide susceptibility using a confusion matrix. Finally, a landslide susceptibility map was generated using the landslide susceptibility index derived from the proposed analysis. The analysis revealed that using all 23 factors resulted in low accuracy (55.90%), but using the 13 factors selected in one step of exploration improved the accuracy to 81.25%. This was further improved to 92.80% using only the nine conditioning factors selected during both steps of the data exploration. Therefore, exploratory data analysis selected the conditioning factors most suitable for landslide susceptibility analysis and thereby improving the performance of the analysis.

An Empirical Study on the Information Search Effort by Information Ambiguity Effects (정보의 모호성이 정보탐색 노력에 미치는 영향요인에 관한 연구)

  • Yoon, Jung-Hyeon
    • Journal of the Korean Society for information Management
    • /
    • v.20 no.3
    • /
    • pp.17-30
    • /
    • 2003
  • The purpose of this study is to identify the relationships among information search activities, of which perceived risk, cognitive dissonance, and involvement play a role. A survey of 155 students who have a recent experience on information search activities. Total six hypotheses the research already reported on information search effort, thereby giving decision maker a richer understanding of information search behavior.

Searching Behavior of Scientists and Technologists Based on Their Information Needs (과학기술분야 연구자의 정보요구별 탐색행태 분석)

  • Han Seung-Hee;Lee Jee-Yeon
    • Proceedings of the Korean Society for Information Management Conference
    • /
    • 2006.08a
    • /
    • pp.17-22
    • /
    • 2006
  • 본 연구에서는 질문지법을 이용하여 과학기술분야 연구자들의 정보요구를 파악하고 이를 바탕으로 정보요구별 탐색행태를 분석하였다. 분석결과, 특정 정보요구에 대해 비공식 커뮤니케이션 채널이 정보탐색에 중요한 영향을 미치는 것으로 나타났으며, 탐색과정에서 이용자가 느끼는 정보부족의 원인에 대해서는 정보요구별로 어느 정도 차이가 있는 것이 확인되었다. 반면, 정보탐색환경에서 연구자들이 중요하게 생각하는 요소는 정보요구별로 큰 차이가 없는 것으로 나타났다. 정보요구를 바탕으로 과학기술분야 연구자의 탐색행태 분석결과를 적용함으로써 이용자 특성에 따라 개별화된 과학기술 정보 서비스를 기대할 수 있다.

  • PDF

Bibliographic Information and Reformulation (서지정보 구조와 재탐색:온라인 목록을 중심으로)

  • 곽철완
    • Journal of the Korean Society for information Management
    • /
    • v.13 no.1
    • /
    • pp.103-117
    • /
    • 1996
  • The purpose of this study is to identify how users react with the search options and bibliographic information during search reformulation on the online catalog. Two online catalogs(N0TIS system and Dynix system) were selected and undergraduate students were recruited as the participants for this study. The study shows that the search options and bibliographic information affected the type of the query reformulation and the re-choice of the search options during search.

  • PDF