Open information extraction from the web

TLDR

Traditional IE systems target narrow, pre‑specified relations in small, homogeneous corpora, and extending them to new domains requires manual rule creation that scales linearly with the number of target relations. This work proposes Open IE, a data‑driven approach that extracts a broad set of relational tuples without human input, and presents TEXTRUNNER, a scalable implementation that assigns probabilities to tuples and supports efficient querying. The authors evaluate TEXTRUNNER on a 9‑million‑page web corpus, comparing its performance to the state‑of‑the‑art KNOWITALL system. TEXTRUNNER achieves a 33 % error reduction, extracts orders of magnitude more facts within the same time as KNOWITALL, and yields 11 million high‑probability tuples comprising over 1 million concrete facts and more than 6.5 million abstract assertions.

Abstract

Traditionally, Information Extraction (IE) has focused on satisfying precise, narrow, pre-specified requests from small homogeneous corpora (e.g., extract the location and time of seminars from a set of announcements). Shifting to a new domain requires the user to name the target relations and to manually create new extraction rules or hand-tag new training examples. This manual labor scales linearly with the number of target relations. This paper introduces Open IE (OIE), a new extraction paradigm where the system makes a single data-driven pass over its corpus and extracts a large set of relational tuples without requiring any human input. The paper also introduces TEXTRUNNER, a fully implemented, highly scalable OIE system where the tuples are assigned a probability and indexed to support efficient extraction and exploration via user queries. We report on experiments over a 9,000,000 Web page corpus that compare TEXTRUNNER with KNOWITALL, a state-of-the-art Web IE system. TEXTRUNNER achieves an error reduction of 33% on a comparable set of extractions. Furthermore, in the amount of time it takes KNOWITALL to perform extraction for a handful of pre-specified relations, TEXTRUNNER extracts a far broader set of facts reflecting orders of magnitude more relations, discovered on the fly. We report statistics on TEXTRUNNER's 11,000,000 highest probability tuples, and show that they contain over 1,000,000 concrete facts and over 6,500,000 more abstract assertions.