實驗室成員‎ > ‎

石琢暐

石琢暐 Cho-Wei Shih

博士班


Site: http://sites.google.com/site/choweishih/

研究方向 Research Interests

  • Image processing
  • Content management/Knowledge management
  • Opining mining/Sentiment analysis in blogosphere
  • Information Retrieval and its Applications
  • Data Mining and its Applications
  • Statistics and its Applications

研究成果 Publications

Journal

    1. Cho-Wei Shih, Tzung-Shiuan Lai, Hui-Chuan Chu, and Yuh-Min Chen (2013), "Image Completion Using Prediction Concept via Support Vector Regression," Machine Vision and Applications, 24(6), 753-768. (IF=1.549, Rank=70/247 (28.34%)).
    2. Cho-Wei Shih, Hui-Chuan Chu, and Yuh-Min Chen (2012), "The Effectiveness of Image Features Based on Fractal Image Coding for Image Annotation," Expert Systems with Applications, 39 (17), 12897-12904. (IF=1.926, Rank=15/75 (20%)).
    3. Cho-Wei Shih, Ming-Yen Chen, Hui-Chuan Chu, and Yuh-Min Chen (2012), "Enhancement of Information Seeking Using an Information Needs Radar Model," Information Processing & Management, 48(3), 524-536 (IF=1.673, Rank=39/128 (30.47%)).
    4. Cho-Wei Shih, Ming-Yen Chen, Hui-Chuan Chu, and Yuh-Min Chen (2011), "Enhancement of domain ontology construction using a crystallizing approach," Expert Systems with Applications, 38(6), 7544-7557 (IF=1.926, Rank=15/75 (20%)).
    5. Ming-Yen Chen, Chan-Hung Kuo, Cho-Wei Shih, Hui-Chuan Chu, and Yuh-Min Chen (2012), "Development of a Concept-based Information Retrieval Mechanism," Information Processing & Management (In Review)  (IF=1.673, Rank=39/128 (30.47%)) .
    6. Cho-Wei Shih, Ming-Yen Chen, Hui-Chuan Chu, and Yuh-Min Chen (2012), "Development of a Self-tuning Document Classification Using Confidence Level," Data and Knowledge Engineering (In  Review) (IF=1.722, Rank=36/128 (28.12%)).
    7. 石琢暐、朱惠娟、陳裕民 (2011)。以顧客體驗分析機制應用於服務需求之探索。臺東大學綠色科學學刊1(1),15-34。

Conference

    1. 張仕政、石琢暐、陳裕民 (2012),以順序關係強化市場機會探索模式發展。ICIM2012 第23屆資訊管理學術研討會,臺灣:高雄,2012-05.
    2. Tzung-Shiuan Lai, Cho-Wei Shih, Hui-Chuan Chu, and Yuh-Min Chen (2011), Applying Support Vector Regression to Predict Structure in Image Completion, The 2011 International Conference on Image Processing, Computer Vision, & Pattern Recognition (IPCV'11). July 18-21, 2011, USA: Las Vegas, Nevada.
    3. Cho-Wei Shih, Ming-Yen Chen, Hui-Chuan Chu, and Yuh-Min Chen (2011), Recommendation System Using Information Needs Radar Model, The World Congress on Engineering 2011 (WCE 2011): The 2011 International Conference of Data Mining and Knowledge Engineering (ICDMKE). July 6-8, 2011, UK: London. (Best Paper Award)
    4. 石琢暐、朱慧娟、陳裕民 (2011),發展顧客體驗分析機制應用於服務需求之探索。第10屆離島資訊技術與應用研討會。臺灣:臺東。2011-05。
    5. 石琢暐、陳明彥、劉德誠、陳裕民 (2010),應用本體論為基之知識表達模式於資訊系統開發,The 21st International Conference on Information Management。
    6. 石琢暐、陳明彥、陳裕民 (2010),應用三維度需求模型於資訊尋求之研究,2010資訊管理暨電子商務經營管理研討會。

博士論文 (2012)

語意導向影像管理實現方法與技術研發
Development of Enabling Methods and Technologies for Semantic-oriented Image Management

關鍵字 Keywords

影像管理, 語意導向影像管理, 影像知識模式, 影像生命週期
Semantic-oriented image management, Image management, Image knowledge model, Image life-cycle

摘要

科技發展與知識經濟加速了資訊、知識與數位內容的流通,雖有助於使用者便利地取得資訊與知識,卻也同時造成資訊過載(Information overload)之現象。資訊過載係指資訊量過多,造成使用者難以適時、確切、迅速地找到需要的資訊,因此,提供一個有效的管理模式,以協助使用者適時獲得並整合正確的資訊,實有其必要性。再者,多數使用者傾向採用影像來記錄生活中的點滴,舉凡人、事、地或物等值得記錄的時刻皆然;且影像可傳達的隱性資訊(包括情感、回憶)較聲音直覺與豐富。是故,影像內容的管理模式更顯其重要性,卻存在幾個關鍵問題:(1)缺乏一般化的影像管理模式;(2)未考量影像之特殊性;(3)忽略影像內隱的語意資訊。 
有鑑於此,本研究參考「知識管理(Knowledge management)」之模式與方法,針對影像的生命週期(Image lifecycle),提出一「語意導向影像管理模式」,再依此設計「語意導向影像管理系統架構」與開發「實現技術(Enabling technologies)」,以實現「影像知識管理」之目標。 
為實現研究目標,本研究訂立之研究項目如下: 
(1) 影像生命週期分析與設計; 
(2) 一般化影像管理模式設計; 
(3) 語意導向影像管理模式設計; 
(4) 語意導向影像管理功能定義與設計; 
(5) 語意導向影像管理系統模組開發; 
(6) 語意導向影像管理系統整合與測試; 
(7) 系統評量(實驗設計)。

Abstract

The rapid development of technology and advent of knowledge economy accelerate the circulation and access of digital content. However, it also increases the difficulty of knowledge acquirement due to information overload. Therefore, it is necessary to provide an effective management method for information gathering and integration. 
Images are widely used to record the happenings of life (including people, events, places or things) due to its ease of use and rich containment of implicit information (including emotions and memories). Therefore, image management has become a remarkable requirement, however, it still exists issues of: (a) lacking of generic image management model; (b) ignoring the characteristics of images; (c) ignoring the implicit semantics of images. 
To fulfill the requirements of image management, this research will study lifecycle of images and then propose a "semantic-oriented image management model" based on the concepts and methods of “knowledge management”. System framework and its enabling technologies will also be developed according to the proposed model. 
To achieve the objective, the research tasks include: 
(a) Design and analysis of image lifecycle, 
(b) Design of generic image management model, 
(c) Design of semantic-oriented image management model 
(d) Planning and design of system framework 
(e) Development of enabling technologies and system modules, and 
(f) Experiments and assessment.