Key Points

1. 本文旨在未提供字典的情況下從 Wikipedia 抽取關係對。
2. 利用 Wiki infobox 建立關係對，再從文章中找到對應的句子。
3. 本文使用 CRF 方法利用 POS 等資訊，另一方法則用分類器利用 dependency pattern 。
4. 其結果是 CRF 表現較佳但較慢。
5. Information-extraction (IE) systems seek to distill semantic relations from natural- language text, but most systems use super- vised learning of relation-specific examples and are thus limited by the availability of training data. Open IE systems such as TextRunner, on the other hand, aim to handle the unbounded number of relations found on the Web. But how well can these open systems perform?