• Title/Summary/Keyword: 기술군집

Search Result 975, Processing Time 0.438 seconds

Binary Visual Word Generation Techniques for A Fast Image Search (고속 이미지 검색을 위한 2진 시각 단어 생성 기법)

  • Lee, Suwon
    • Journal of KIISE
    • /
    • v.44 no.12
    • /
    • pp.1313-1318
    • /
    • 2017
  • Aggregating local features in a single vector is a fundamental problem in an image search. In this process, the image search process can be speeded up if binary features which are extracted almost two order of magnitude faster than gradient-based features are utilized. However, in order to utilize the binary features in an image search, it is necessary to study the techniques for clustering binary features to generate binary visual words. This investigation is necessary because traditional clustering techniques for gradient-based features are not compatible with binary features. To this end, this paper studies the techniques for clustering binary features for the purpose of generating binary visual words. Through experiments, we analyze the trade-off between the accuracy and computational efficiency of an image search using binary features, and we then compare the proposed techniques. This research is expected to be applied to mobile applications, real-time applications, and web scale applications that require a fast image search.

Speech Synthesis using Diphone Clustering and Improved Spectral Smoothing (다이폰 군집화와 개선된 스펙트럼 완만화에 의한 음성합성)

  • Jang, Hyo-Jong;Kim, Kwan-Jung;Kim, Gye-Young;Choi, Hyung-Il
    • The KIPS Transactions:PartB
    • /
    • v.10B no.6
    • /
    • pp.665-672
    • /
    • 2003
  • This paper describes a speech synthesis technique by concatenating unit phoneme. At that time, a major problem is that discontinuity is happened from connection part between unit phonemes, especially from connection part between unit phonemes recorded by different persons. To solve the problem, this paper uses clustered diphone, and proposes a spectral smoothing technique, not only using formant trajectory and distribution characteristic of spectrum but also reflecting human's acoustic characteristic. That is, the proposed technique performs unit phoneme clustering using distribution characteristic of spectrum at connection part between unit phonemes and decides a quantity and a scope for the smoothing by considering human's acoustic characteristic at the connection part of unit phonemes, and then performs the spectral smoothing using weights calculated along a time axes at the border of two diphones. The proposed technique removes the discontinuity and minimizes the distortion which can be occurred by spectrum smoothing. For the purpose of the performance evaluation, we test on five hundred diphones which are extracted from twenty sentences recorded by five persons, and show the experimental results.

Clustering Analysis of Effective Health Spending Cost based on Kernel Filtering Techniques (커널필터링 기법을 이용한 건강비용의 효과적인 지출에 관한 군집화 분석)

  • Jung, Yong Gyu;Choi, Young Jin;Cha, Byeong Heon
    • Journal of Service Research and Studies
    • /
    • v.5 no.2
    • /
    • pp.25-33
    • /
    • 2015
  • As Data mining is a method of extracting the information based on the large data, the technique has been used in many application areas to deal with data in particular. However, the status of the algorithm that can deal with the healthcare data are not fully developed. In this paper, One of clustering algorithm, the EM and DBSCAN are used for performance comparison. It could be analyzed using by the same data. To do this, EM and DBSACN algorithm are changing performance according to the variables in Health expenditure database. Based on the results of the experimental data, We analyze more precise and accurate results using by Kernel Filtering. In this study, we tried comparison of the performance for the algorithm as well as attempt to improve the performance. Through this work, we were analyzed the comparison result of the application of the experimental data and of performance change according to expansion algorithm. Especially, Collects data from the various cluster using the medical record, it could be recommended the effective spending on medical services.

Feature-based Gene Classification and Region Clustering using Gene Expression Grid Data in Mouse Hippocampal Region (쥐 해마의 유전자 발현 그리드 데이터를 이용한 특징기반 유전자 분류 및 영역 군집화)

  • Kang, Mi-Sun;Kim, HyeRyun;Lee, Sukchan;Kim, Myoung-Hee
    • Journal of KIISE
    • /
    • v.43 no.1
    • /
    • pp.54-60
    • /
    • 2016
  • Brain gene expression information is closely related to the structural and functional characteristics of the brain. Thus, extensive research has been carried out on the relationship between gene expression patterns and the brain's structural organization. In this study, Principal Component Analysis was used to extract features of gene expression patterns, and genes were automatically classified by spatial distribution. Voxels were then clustered with classified specific region expressed genes. Finally, we visualized the clustering results for mouse hippocampal region gene expression with the Allen Brain Atlas. This experiment allowed us to classify the region-specific gene expression of the mouse hippocampal region and provided visualization of clustering results and a brain atlas in an integrated manner. This study has the potential to allow neuroscientists to search for experimental groups of genes more quickly and design an effective test according to the new form of data. It is also expected that it will enable the discovery of a more specific sub-region beyond the current known anatomical regions of the brain.

Semantic Structure Represented in College Presidents' Welcome Greetings Using Network Analysis : Daegu & Gyeongbuk Provinces (연결망 분석을 활용한 대학 총장 인사말의 의미론적 구조: 대구·경북 지역을 중심으로)

  • Son, Ji-Hoon;Kim, Jae-Hun;Park, Han-Woo
    • The Journal of the Korea Contents Association
    • /
    • v.21 no.9
    • /
    • pp.24-33
    • /
    • 2021
  • This study examined a semantic relationship expressed in college presidents' welcome greetings in order to explore the promotion strategies and future direction of universities in Daegu & Gyeongbuk provinces in South Korea. Greetings were collected from university websites as of September, 2020. According to word frequency analysis, "everyone," "welcome," and "visiting" were mostly used in the headlines. In the body texts, "college" and "education" were frequently paired. While the two- & three-year colleges focus on industrial and technical capabilities, four-year universities tend to emphasize educational excellence and academic research performance. This study is valuable in that it understands the direction that universities in Daegu and North Gyeongsang Province put forward amid the decreasing school-age population and the changing social environment.

A Study on Research Paper Classification Using Keyword Clustering (키워드 군집화를 이용한 연구 논문 분류에 관한 연구)

  • Lee, Yun-Soo;Pheaktra, They;Lee, JongHyuk;Gil, Joon-Min
    • KIPS Transactions on Software and Data Engineering
    • /
    • v.7 no.12
    • /
    • pp.477-484
    • /
    • 2018
  • Due to the advancement of computer and information technologies, numerous papers have been published. As new research fields continue to be created, users have a lot of trouble finding and categorizing their interesting papers. In order to alleviate users' this difficulty, this paper presents a method of grouping similar papers and clustering them. The presented method extracts primary keywords from the abstracts of each paper by using TF-IDF. Based on TF-IDF values extracted using K-means clustering algorithm, our method clusters papers to the ones that have similar contents. To demonstrate the practicality of the proposed method, we use paper data in FGCS journal as actual data. Based on these data, we derive the number of clusters using Elbow scheme and show clustering performance using Silhouette scheme.

Segmentation of Target Objects Based on Feature Clustering in Stereoscopic Images (입체영상에서 특징의 군집화를 통한 대상객체 분할)

  • Jang, Seok-Woo;Choi, Hyun-Jun;Huh, Moon-Haeng
    • Journal of the Korea Academia-Industrial cooperation Society
    • /
    • v.13 no.10
    • /
    • pp.4807-4813
    • /
    • 2012
  • Since the existing methods of segmenting target objects from various images mainly use 2-dimensional features, they have several constraints due to the shortage of 3-dimensional information. In this paper, we therefore propose a new method of accurately segmenting target objects from three dimensional stereoscopic images using 2D and 3D feature clustering. The suggested method first estimates depth features from stereo images by using a stereo matching technique, which represent the distance between a camera and an object from left and right images. It then eliminates background areas and detects foreground areas, namely, target objects by effectively clustering depth and color features. To verify the performance of the proposed method, we have applied our approach to various stereoscopic images and found that it can accurately detect target objects compared to other existing 2-dimensional methods.

Metaproteomics in Microbial Ecology (메타프로테오믹스의 미생물생태학적 응용)

  • Kim, Jong-Shik;Woo, Jung-Hee;Kim, Jun-Tae;Park, Nyun-Ho;Kim, Choong-Gon
    • Korean Journal of Microbiology
    • /
    • v.46 no.1
    • /
    • pp.1-8
    • /
    • 2010
  • New technologies are providing unprecedented knowledge into microbial community structure and functions. Even though nucleic acid based approaches provide a lot of information, metaproteomics could provide a high-resolution representation of genotypic and phenotypic traits of distinct microbial communities. Analyzing the metagenome from different microbial ecosystems, metaproteomics has been applied to seawater, human guts, activated sludge, acid mine drainage biofilm, and soil. Although these studies employed different approaches, they elucidated that metaproteomics could provide a link among microbial community structure, function, physiology, interaction, ecology, and evolution. These approaches are reviewed here to help gain insights into the function of microbial community in ecosystems.

An Empirical Study and Policy Implications Regarding Correlations of Korean Small Businessman's Perception of Systematization Using Cluster Analysis (한국 소상공인의 조직화 인식도 상호관계에 관한 실증적 연구와 정책적 시사점 : 군집분석을 이용한 접근)

  • Suh, Geun-Ha;Lee, Kwang-No;Yoon, Sung-Wook
    • Journal of the Korea Academia-Industrial cooperation Society
    • /
    • v.12 no.3
    • /
    • pp.1157-1164
    • /
    • 2011
  • In this study, association of small business is divided into four groups: Franchise, Joint Brand, Industry Association and Registered Retailer. Cluster analysis is taken to find what kind of strategic considerations associated small businesses choose when they set up new strategies. The results show that there are some differences in the perception of association, effects of association and final performance of management by gender, academic background, and age. The data also find three clusters: price competitive, marketing competitive and neither group. Implications of this study is that government should focus more on not only improving infrastructures of self-businesses but also associating small businesses, modernizing managerial systems in the future.

Components Clustering for Modular Product Design Using Network Flow Model (네트워크 흐름 모델을 활용한 모듈러 제품 설계를 위한 컴포넌트 군집화)

  • Son, Jiyang;Yoo, Jaewook
    • Journal of the Korea Academia-Industrial cooperation Society
    • /
    • v.17 no.7
    • /
    • pp.263-272
    • /
    • 2016
  • Modular product design has contributed to flexible product modification and development, production lead time reduction, and increasing product diversity. Modular product design aims to develop a product architecture that is composed of detachable modules. These modules are constructed by maximizing the similarity of components based on physical and functional interaction analysis among components. Accordingly, a systematic procedure for clustering the components, which is a main activity in modular product design, is proposed in this paper. The first phase in this procedure is to build a component-to-component correlation matrix by analyzing physical and functional interaction relations among the components. In the second phase, network flow modeling is applied to find clusters of components, maximizing their correlations. In the last phase, a network flow model formulated with linear programming is solved to find the clusters and to make them modular. Finally, the proposed procedure in this research and its application are illustrated with an example of modularization for a vacuum cleaner.