知識分享推薦機制應用於線上論壇社群之研究

Abstract

網際網路 (Internet)和網路應用服務 (Web Application)的發展，使得網路上分享的知識變得複雜且多元。線上論壇社群由於使用者間的社會關係、文章循序的呈列結構和使用者互相討論的公開性、使其成為知識管理的利器之一。本篇研究提出一個以結合文字語意分析、社會網路關係及角色與專業知識分析的線上論壇知識推薦機制，並且描述此推薦如何應用在FreeAdvice網站上。此推薦機制分為兩部份：文章推薦及專家推薦。當使用者在推薦清單裡沒有找到符合的結果時，此推薦機制採用馬可夫鏈模型概念，從使用者的社會關係裡推薦出最具意願性且知識程度足夠解答使用者問題的專家。從實驗結果來看，我們從發展的社群論壇網路中發現許多的意涵並實證了社會網路關係對知識分享的重要性。實驗評估顯示我們提出的線上論壇知識分享推薦機制具有高度的可行性及潛力。

Nowadays, online forums have become a useful tool for knowledge management for the decades of flourishing in Web-based technology. This study proposes a synthetical recommendation mechanism which includes experts and articles recommendation and combines with Profession and Reliability (PR), Social Intimacy and Popularity (SIP), Semantic Similarity (SS), and social network based Markov Chain model (SN-MC), for knowledge sharing in the online forum communities. The process of recommendation includes two phases, relevant discussion thread recommending and helpful expert finding. The advantage of the proposed mechanism is its overall consideration of aspects in knowledge sharing. Accordingly, our experimental results show that with the support of the proposed recommendation mechanism, the requesters in forum can easily find the similar discussion threads to prevent spamming the same discussion. Besides, if the requesters can’t find the qualified discussion threads, this mechanism provides a relatively efficient and active way to find the appropriate experts.