本研究旨在實測ASReview系統性文獻回顧工具於教育領域文獻篩選之適用性與操作流程。隨著人工智慧技術逐步導入文獻篩選與證據統合研究，ASReview憑藉主動式學習與機器學習模型，在加速文獻探討方面展現高度潛力。本文以實作導向的方式，詳細說明ASReview的操作介面、流程設計、智慧篩選機制及其技術限制。實測結果顯示，ASReview能顯著提升在標題與摘要層級進行大規模初篩的效率，有效地降低研究者的人工負荷。然而，本系統於多使用者合作、全文評閱支援、操作介面友善度與即時進度掌握等層面仍面臨挑戰，且需具備基本的程式操作能力，對非技術背景使用者而言存在進入門檻。綜合建議，ASReview適合應用於個人或小型研究團隊進行文獻初步篩選階段，並可搭配其他工具補足全文篩選流程，作為教育領域導入AI文獻篩選的實務參考依據。

This study aims to evaluate the applicability and operational workflow of ASReview, an AI-assisted tool for conducting systematic literature screening in the field of education. With the growing integration of artificial intelligence into evidence synthesis and review processes, ASReview has drawn increasing attention for its potential to accelerate literature screening through active learning and machine learning models. Adopting a hands-on approach, this article presents a detailed walkthrough of ASReview’s interface, operation procedures, intelligent screening mechanisms, and system limitations. Results indicate that ASReview can significantly enhance efficiency during large-scale title and abstract screening, thereby reducing researchers’ manual workload. However, several challenges remain, including limited support for multi-user collaboration, full-text screening, interface intuitiveness, and real-time progress tracking. In addition, the system requires basic programming knowledge, which may pose a barrier for users without technical backgrounds. Based on these findings, we recommend that ASReview be prioritized for individual or small-team preliminary screening and supplemented by other tools for full-text review—offering practical guidance for incorporating AI tools into literature screening in educational research contexts.