一個社會性書籤網站之探索分析

社會性書籤網站應用Web 2.0群體智慧，讓使用者利用書籤收藏自己有興趣的網路資源，並藉由標籤的標記功能管理書籤。關於社會性書籤網站的相關研究，目前已有相當多的國外研究成果發表，然而國內社會性書籤網站的相關分析卻仍然有限，因此本研究以台灣的funP推推王書籤網站為研究對象，嘗試探索台灣社會性書籤的使用概況：首先，蒐集2010年7月27日至2010年8月27日為期一個月的書籤資料為研究樣本，進行書籤之分類、書籤之標籤分佈、以及書籤標記時序等基本統計；接著，再擷取2010年7月27日至2010年8月2日為期一週的書籤資料，進行細部的使用者、標籤、網路資源等書籤三大元素的使用頻率分析與關聯分析。本研究結果發現，三大元素的使用頻率均符合齊夫定律，而三大元素的關聯關係也均符合冪次定律。換言之，使用頻率較高的少數使用者、少數標籤、以及少數網路資源主宰了多數的書籤。因此，本研究再檢視Top 20標籤，發現第一名的標籤為部落格，而其他則多為部落格相關詞彙。最後，再檢視Top 20網路資源，則確認它們多數為部落格網站。

關鍵字

社會性書籤；使用者；網路資源；標籤；齊夫定律；冪次定律；部落格

並列摘要

Applying the collective intelligence of Web 2.0, social bookmarking sites allow users to bookmark network resources of personal interest and to manage their bookmarks via tagging. Although a lot of research about foreign social bookmarking sites has been published, research on the usage of social bookmarking sites in Taiwan is still limited. Therefore, in this research we attempt to reveal the usage patterns of social bookmarking in Taiwan by investigating the social bookmarking site funP. First, we collected the bookmark data of funP from July 27, 2010 to August 27, 2010 as our research sample, and we conducted basic statistical analyses of the data, including the categorization of bookmarks, tag distribution of bookmarks, and bookmark timing of users. Then, we took the one-week bookmark data from July 27, 2010 to August 2, 2010 to perform usage-frequency and association analyses of users, tags, and URLs within bookmarks. We find that usage-frequencies of users, tags, and URLs all follow Zipf's law, while relationships between users, tags, and URLs all follow the power law. In other words, a minority of high-frequency users, tags, and URLs dominate the majority of bookmarks. Due to the discovery, we further inspected the Top 20 tags and find that the tag ”blog” has the highest usage-frequency and the rest are mostly blog-related terms. We eventually inspected the Top 20 URLs, confirming that most of them direct to blog sites.