問答網站專家尋找機制之研究

Abstract

隨著網路科技的發展，問答網站逐漸成為重要的知識分享平台。在問答網站中，使用者可以提出任何種類的問題，然後等待其他使用者的問答。然而這樣的問答機制可能會產生一些潛在的問題。當同時有太多問題等待回答的時候，有些問題可能會被該問題的專家忽略掉，造成發問者需要等待很長的時間才能得到解答，甚至還有可能得不到適當的專家協助，因而阻礙了問答網站中的知識分享過程。 為了改善等待時間過長問題及尋找適當的專家，加速知識分享過程，我們在本研究中提出一個針對問題所屬領域找出對應領域專家的方法。我們的方法除了考慮目標問題與使用者專業知識之間的相關性外，還考量了使用者的聲望與權威。使用者聲望是經由分析使用者過去的回答行為獲得，而使用者權威則是透過鏈結分析得到。此外，我們提出問題相依之專家尋找機制，進一步考量歷史問題與目標問題的相關性來推導使用者專業知識的相關性及使用者的聲望與權威。最後我們利用奇摩知識家做為評估的資料來源，實驗結果顯示本研究所提出的方法比傳統方法能更有效的針對問題找出相對應專家。

Question answering websites are becoming an ever more popular knowledge sharing platform. On such websites people may ask any type of question and then wait for someone else to answer the question. Unfortunately, such question answering mechanisms raise some issues. Some questions may be passed over by those users in a position to answer them when the quantity of questions waiting to be solved grows quickly, and thus askers will waste a lot of time when trying to obtain answers. Even worse, askers may not obtain correct answers from appropriate experts. In this manner, knowledge sharing through question answering websites is interfered with. In this paper, we propose a novel category-based approach to effectively find experts for the category of the target question in question answering websites. Our approach considers user subject relevance, user reputation and authority of a category in finding experts. A user’s subject relevance denotes the relevance of a user’s domain knowledge to the target question. A user’s reputation is derived from the user’s past answering behavior, while user authority is derived from link analysis. Moreover, our proposed approach can be extended to develop a question dependent approach that considers the question relevance of past questions to the target question in deriving user domain knowledge, reputation and authority. We used a dataset obtained from Yahoo! Answer Taiwan to evaluate our approach. The experiment results show that our proposed methods outperform other conventional methods.