APRCOIE: An open information extraction system for Chinese

Abstract

Open information extraction (OIE) is critical in natural language processing. Compared to the flourishing development of OIE systems in English, very few high-quality Chinese OIE systems are publicly available. APRCOIE is a system that conducts open information extraction for Chinese text data. Differing from traditional rule-based approaches and learning-based methods, the new system innovatively employs automated methods to explore the nature of complex Chinese literal characteristics, thus generating a large number of extraction rules and subsequently establishing the extraction model. The feature of low resource demand makes APRCOIE easy to deploy in various real applications.

Keywords

• Open information extraction
• Tensor computing
• Automated pattern generation
• Rule-based OIE
• Learning-based OIE