Xerox Incremental Parser

Fine-Grained Linguistic Analysis For Contextual Entity Recognition And Fact Extraction.

Technology Description

Extracting the meaning from unstructured content is an ever growing need. It calls for computational text analysis that goes far beyond simple keyword search.

XIP is an on-the-fly rule compiler with very powerful formalism, and an API to integrate its syntacic and semantic text parsing functionalities into end-user applications. It comes with an interface for grammer developers to build upon XRCE's XIP grammars.

XIP English Parsing Capabilities

The XIP English grammar can extract the following:
  • Chunks - verb phrases (VP), noun phrases (NP), prepositional phrases (PP), etc.
  • Dependencies - subject / object, passive / active, etc.
  • Named entities - people, organizations, locative & temporal expressions
  • Semantic roles - destination, purpose, duration
  • Communicative intentions - paradigm shifts, etc.
  • Co-references

Features & Benefits For Developers

  • More expressive formalism - Can represent and manipulate a larger number of linguistic features, express abstract rules, and access contect (incl. in other sentences)
  • Declarative - Rules and engine are distinct
  • Modular - Linguistic descriptions are organized in modules. One multi-purpose system for various end-user solutions
  • Versatile, multi-functional, language-independent and incremental - New rules can build upon existing ones
  • Full XML architecture
  • Computational efficiency - ca. 2,000 words per second in English

Intellectual Property Summary

Xerox Intellectual Property includes patents, patent applications, and know-how.

For Licensing Information

To learn more about licensing the Xerox Incremental Parser technology.