The eXtended Pattern Definition Language (XPDL) is a PolyAnalyst’s proprietary rule language for information extraction.

Information Extraction is the task of automatically extracting meaningful information, such as company and people names, product models, stock prices, customer complaints, from unstructured texts. XPDL enables users to develop extraction rules for various kind of information and can be applied in a wide range of tasks, such as social media and customer feedback analysis, patient record analytics, competitive intelligence.

At the core of XPDL is PDL, PolyAnalyst query language that provides rich features to analyze unstructured texts. XPDL provides a shell over PDL language to organize PDL queries into a hierarchy and capture complex patterns.

The next sections describe XPDL syntax and capabilities. To facilitate the quick start, example rules are available in the section Appendix.