透過您的圖書館登入
IP:18.191.93.12
  • 學位論文

資料探勘技術在牆面觸控展示系統互動行為分析之應用

Applying Data Mining Techniques to Analyzing Interaction Behavior in Touch Display Systems

指導教授 : 黃有評

摘要


觸控式技術越來越成熟,但如何提升觸控系統與使用者之間的互動性,讓使用者在操作上更加便利,是目前在研究人機介面設計的重點方向。本論文以大型牆面當作觸控螢幕為互動環境,設計一套互動式觸控展示系統,可將展示內容投影至牆面,透過光學式觸控技術,可直接用手在投影畫面上操作物件。在互動設計上,本研究以無線射頻辨識技術登入系統,再以行動條碼、使用行為分析及圖片推薦等模組來設計展示系統,以提升展示系統之互動性。以無線射頻辨識技術登入系統可有效管理系統使用狀況,行動條碼可提供展示系統與手持裝置之間互動,使用行為分析及圖片推薦模組是用來收集使用者在牆面螢幕上的觸控事件,並將資訊儲存至後端資料庫,接著使用時間資料探勘技術挖掘資料間的關聯法則,利用此結果來分析及推薦使用者資訊,可提升軟體設計品質及互動性。另外,在互動式介面的設計上,以分割視窗作為概念,將畫面分割為四個視窗,並以連連看互動式小遊戲為例,讓使用者在玩樂中學習,不僅加深使用者對資訊的印象,也能加深與使用者的互動。本論文除提出一套完整的大型牆面觸控展示系統外,亦分析使用者行為資料探勘之結果,以驗證所提系統之有效性。

並列摘要


Touch technology is blooming quickly. Improving the interactions between users and the touch system requires intensive study on the human-computer interface design. A touch display system basing on large wall as a projection medium is presented in this study. Using the technology of optical touch, the users can directly operate the objects projected on the screen. In the proposed interactive design, users present their own RFID tags to login the system. QR code, user behavior analysis, picture recommendation and other modules are integrated into the display system to enhance the interactivity. Using RFID to login the system can effectively manage the interactive system. QR code provides a medium for bridging the display system with handheld devices. User behavior analysis and picture recommendations are used to collect users’ touch behavior on clicking and dragging events on the screen. Those touch events are further analyzed in the back-end database by temporal data mining to extract interesting association rules. The proposed interaction interface is to separate the display screen into 4 sub-windows that provide multiple functions for users to browse simultaneously. Each sub-window has its own function and the upper right sub-window is designed for picture matching game for the purpose of learning through games. The visitors can memorize the contents of exhibitions through games. A practical and large touch display system is presented as well as users’ interactive behavior is analyzed by data mining to verify the effectiveness of the proposed system.

參考文獻


[15] 陳彥良、趙書榮、陳禹辰,「幾個快速挖掘關聯規則的資料探勘方法」,電子商務學報,第五卷,第二期,2003,第1-10頁。
[2] E. Hornecker, ““I Don’t Understand It Either, But It Is Cool” - Visitor Interactions with A Multi-Touch Table in A Museum,” in Proc. of Third IEEE Int. Workshop on Horizontal Interactive Human Computer Systems, Amsterdam, Nederland, Oct. 2008, pp.113-120.
[4] S.-Y. Wu and Y.-L. Chen “Mining Nonambiguous Temporal Patterns for Interval-based Events,” IEEE Trans. on Knowledge and Data Engineering, vol. 19, no. 6, June 2007, pp.742-758.
[5] C.-M. Chen, C.-M. Hong and C.-C. Chang, “Mining Interactive Social Network for Recommending Appropriate Learning Partners in A Web-based Cooperative Learning Environment,” in Proc. of 2008 IEEE Conf. on Cybernetics and Intelligent Systems, Chengdu, China, Sep. 2008, pp.642-647.
[7] Y.-P. Huang, Y.-T. Chang and F.E. Sandnes, “RFID-based Interactive Learning in Science Museums,” in Proc. of the 5th Int. Conf. on Ubiquitous Intelligence and Computing, Oslo, Norway, June 2008, pp.700-712.

延伸閱讀