以知識流為基礎之文件推薦:整合個人化與群體式推薦方法

Abstract

推薦系統被用來改善資訊過載的問題，而其主要的概念為依據使用者的喜好來進行推薦。在一個知識密集的環境中，知識工作者需要存取與任務相關的知識(文件)來執行任務，而工作者對文件的參考行為可以知識流的方式呈現，用以代表工作者在不同時間點，由於資訊需求的改變所造成的參考行為演化。根據工作者的資訊需求，文件推薦方法可以推薦適當的文件，以主動地支援工作者的工作。然而，大部分傳統的推薦方法沒有考慮工作者的知識流，而且忽略其他大多數具有相似知識流的群體工作者之資訊需求。由於從工作者過去的參考行為來取得個人的資訊需求會有所遺漏，而群體的資訊需求可以反映工作者過去的參考行為中，所遺漏的部分個人資訊需求，並可補充工作者的個人需求。因此，本研究方法採用混合式方法，將以知識流為基礎的群體式推薦方法與傳統推薦方法結合，方法中考量群體的觀點來補足個人觀點之不足。藉由整合兩種方法，來平衡兩個方法之間的權重並取得更準確的推薦。最後在實驗結果中顯示所提出的方法比傳統推薦方法有較高的準確度。

Recommender systems can be used to improve the information overloading problem and help workers to identify interested knowledge based on their preferences. In a knowledge-intensive environment, knowledge workers need to access task-related codified knowledge (documents) to perform their tasks. A worker’s referencing behavior on document access can be modeled as a knowledge flow (KF) to represent the evolution of his/her information needs over time. Document recommendation methods can proactively support knowledge workers in task performance by recommending appropriate documents to suit their information needs. However, most traditional recommendation methods do not consider workers’ knowledge flows and the majority of information needs of a group of workers with similar knowledge flows. The group’s information needs may reflect a part of an individual worker’s information needs, which cannot be derived from his/her past referencing behavior, and can complement a worker’s personal needs. Thus, we take the group perspective into consideration to offset the drawback of personal perspective by hybrid approaches, which combine the KF-based group recommendation method (KFGR) with traditional recommendation methods. The proposed hybrid methods can balance the tradeoff between the group-based and personalized methods by integrating the merits of both methods. The experiment results show that the proposed methods can enhance the quality of recommendation compared with the traditional recommendation methods.