透過您的圖書館登入
IP:3.149.26.176
  • 期刊

結合多使用者量化指標分析之數位內容推薦技術

A Recommendation Technique of Combining Multi-User Based Quantitative Indicators for Digital Content

摘要


隨著數位化生活時代的來臨,各部落格網站所擁有的數位內容加倍的成長,網站經營者如何在巨量資料中找尋到使用者想要看的部落格文章,是一個重要的議題。本文提出一個完整的個人化數位內容推薦技術,以內容關聯分析為基礎並結合使用者喜好度、社群緊密度與文章新鮮度三種使用者量化指標,實際應用於部落格網站。實務成果發現,結合多使用者量化指標分析的數位內容推薦技術,不僅可以幫助網站提升使用者滿意度與黏著度,更可以增加文章類別的覆蓋率。此結果表示本文所提的技術,可以有效的協助數位內容服務網站提昇良性的使用者數位閱讀經驗。

並列摘要


With the coming of digital reading era, blog website has also being emerging. It is important to help users find what they want to read in blog website. This paper presents a personalized recommendation technique considering user-based analysis and combining three types quantitative indicators for digital content(user preference correlation analysis, social relation analysis and freshness analysis for digital content).In practice result, combined with quantitative indicators of multi-user digital content recommendation technology not only helps improve user satisfaction and adhesion degree with the blog website, but can increase the coverage of article category. This result indicates the proposed technique can effectively help to enhance users’ digital reading experience.

延伸閱讀