中央研究院 資訊科學研究所

活動訊息

友善列印

列印可使用瀏覽器提供的(Ctrl+P)功能

Connection Discovery Through User Shared Images From Social Media - From Analytics, System to Applications

:::

Connection Discovery Through User Shared Images From Social Media - From Analytics, System to Applications

  • 講者許丕文 教授 (Hong Kong Uni. of Sci. & Tech., Hong Kong)
    邀請人:陳昇瑋
  • 時間2016-06-17 (Fri.) 11:00 ~ 12:00
  • 地點資訊所新館106演講廳
摘要

Billions of user shared images in social media are seamlessly generated by individuals everyday, and this form of user shared content is widely accessible by others due to its intrinsic nature for sharing in online/mobile social networks. From our research results, using user shared images is proved as an effective alternative to improve the service relevance for users by discovering user connections, such as recommendations for following/followee-relationship, friendship and information relevant to user interests on social media platforms. 

 

In this talk, we will share our research results that investigated over 8.5 million user shared images from 8 social networks with different content sharing mechanisms from different origins, with over 20 million online connections. A phenomenon is observed across these social networks that connected users tend to share visually similar images, regardless of the origins and content sharing mechanisms of the social networks. Through this phenomenon, novel practical analytics using user shared images are developed as effective and more accessible alternatives to techniques using user annotated tags and social graphs for effectively discovering user connections and communities. To the best of our knowledge, this is the first attempt to prove and characterize such phenomenon by massive user shared images, and propose practical analytics and systems to discover user connections for various applications. 

 

BIO

許丕文教授目前擔任香港科技大學電子及計算機工程系的助理教授,英國劍橋大學的訪問研究院士。同時他成立了亞洲第一所社交媒體研究中心(香港科技大學NIE社交媒體研究中心), 致力於所有社交媒體數據分析及新一代社交媒體設計及開發。許丕文教授作為世界經濟論壇全球議程理事會(社交媒體) 的成員,參與與其它政府及商家探討關鍵的社交媒體問題。