透過您的圖書館登入
IP:3.145.186.173
  • 學位論文

手持裝置上個人化網路影音連結隨選系統

Personalized Web-based Media Links on Demand System for Mobile Devices

指導教授 : 李秀惠

摘要


隨著具有無線上網功能的手持裝置日漸普及,透過手機上網觀看影音資訊已經是相當可行的新體驗。大都市中的WiFi無線網路涵蓋廣泛,幾乎可以隨時隨地上網,在郊區也可以透過新一代的3G通訊協定上網,所以現在到處都可以用智慧型手機上網。但是受限於手機狹小的螢幕,瀏覽網站內容相當的不方便,不僅要找到影音資源困難,要打入影音連結的網址更是不容易。除此之外,現有的手機上網收看影音服務大多需要收費,而且影音內容更受限於提供網站,可供選擇的內容不夠多樣化,使用者仍無法有便利的手機上網收看影音內容的管道。 但是網路上存在的影音連結相當豐富,無論是新聞影音、電影預告以及熱門音樂影片等,都有許多網站提供公開的連結。如果可以設計一個系統,自動抓取並整理分類這些影音連結,而後依照使用者個人的喜好,把連結以簡單的清單方式傳送到使用者的手持裝置上,就可以在手機上欣賞豐富且免費的多媒體資源了。因此我們設計了這個系統,能夠自動抓取網路上現有的各種影音連結,並且分類整理後,提供給手持裝置的使用者。 除此之外,我們還設計了使用者回饋機制,能夠從使用者的反應當中去學習使用者的喜好,讓使用者可以很快的點選到想看的連結,省去在手持裝置上找尋的時間,以及定義出各項影音連結的熱門程度,來達到個人化的線上影音連結隨選系統,並於最後的實驗結果證明本論文中個人化的機制,的確有助於讓使用者能更快的點擊到感興趣的連結。

並列摘要


Playing videos on mobile devices is amazing. With the growth of mobile development, nowadays cellular phones are capable of playing multimedia contents and accessing to the Internet. The way mobile users access the Internet is via the WiFi or 3G functionality of their mobile phones. At the present time, wireless network signal is almost everywhere in the city. Taipei, for example, is claimed to be the first 100% wireless network coverage city in the world. Moreover, 3G protocol also becomes popular so that mobile users can easily access the Internet at anywhere. But due to the limit of small screen size of most mobile phones, how to get the demanded multimedia content from the Internet such as free news videos, movie trailers and music videos is still confusing the user. Another problem is that it’s inconvenient to key in words to search for interesting videos. In this thesis we design a personalized web-based multimedia on demand system for wireless mobile devices that provides a convenient approach for the user to easily get their interesting contents from the web. Our system not only crawls and extracts video links from the web, but also categorizes and scores them by our algorithms, and generates simple and user-friendly hyper-linked video lists to the user’s mobile screen. Also, our system can record and learn from users’ browsing history, and provides video links that are closed to each user’s preferences. We design a user feedback module that can improve the preciseness of users’ preferences. So the mobile users won’t need to have patience with typing in the long and ugly URLs and scrolling the web page up and down to find videos anymore. We aim at making a revolution in the way people watch web-based videos on their mobile devices.

參考文獻


[JH 2003] Cho, J. and Garcia-Molina, H. “Effective page refresh policies for web crawlers,” ACM Transactions on Database Systems 28(4), 2003.
[HMI 2003] Keiichiro Hoashi, Kazunori Matsumoto, Naomi Inoue. “Personalization of user profiles for content-based music retrieval based on relevance feedback,” Proceedings of the eleventh ACM international conference on Multimedia MULTIMEDIA '03, 2003.
[Wikipedia] http://en.wikipedia.org/
[JH 2000] Cho, J. and Garcia-Molina, H. “Synchronizing a database to improve freshness,” In Proceedings of ACM International Conference on Management of Data (SIGMOD), 117-128, 2000.
[CL 2006] Sheung-On Choy, Andrew K. Lui. “Web Information Retrieval in Collaborative Tagging Systems,” Proceedings of the 2006 IEEE/WIC/ACM International Conference on Web Intelligence WI '06, 2006.

延伸閱讀