學術研討會資訊推薦系統建置

Abstract

鑒於學術研究者對於未來學術研討會的資訊需求迫切，希望能建置一個收集並推薦這方面資訊的系統，將即時、正確的資訊提供給研究者。本研究設計實做一個以內容導向式(content-based approach)的推薦系統(Scholarship Conference Information Recommend System)SCIRS，基於過去學者曾參與研討會資訊，利用文字探勘(text mining)及網頁探勘(web mining)的技術，搜尋網路上將舉辦的研討會資訊，篩選相關的資訊，搜尋相關度高的研討會資訊，作為推薦的標的。 本推薦系統除了使用Apache Lucene開發的全文檢索引擎，並結合自動化程式，定時擷取網路上研討會新資訊，檢索相關資訊再提供給學術研究者，提供一個便捷的取得研討會資訊管道。

In light of urgent demands on future conference information required by researchers, this study aims to design and implement a system which can collect accurate and timely information for them and meet their requirements. By reviewing related literature, this study built SCIRS (Scholarship Conference Information Recommendation System) in terms of content-based approach. The developed system adopts the text mining and web mining techniques to search the conference information which will held in the near future and relevant to their research fields. Therefore, it could help researchers to filter and index related information as recommendation. In addition to the context-search engine developed by Apache Lucene, the developed system also integrates the automated update engine which will access the Internet and update conference information into the SCIRS on schedule. We hope that SCIRS could provide researchers with a simple and efficient way to acquire conference information.