PLATFORM : XELDA PROJECT

XeLDA is an engine for transforming, normalizing, and extracting information from text, engineered from research in natural language at the Xerox Research Centre Europe.

XeLDA: Xerox Linguistic Development Architecture

Before information in text can be exploited by computers, it has to be recognized among all the variety of forms that it can take. Two different ways of saying the same thing have to be normalized into identical expressions so that the computer can compare them. Once the computer can accurately recognize and normalize structures in text it can perform a number of useful services on documents such as indexation, summarization, information extraction, and translation.

Linguistics researchers at the Xerox Research Centre Europe (XRCE) and at the Palo Alto Research Center (PARC) have long been leading forces in automation of natural language processing. Fruits of their laboratory researches have been harvested to feed into practical and usable products. The latest product is the Xerox Engine for Linguistic Dependent Applications (XeLDA) which profits from decades of Xerox linguistic research, providing a general-purpose natural language processing engine, packaged in an open and modular multi-platform client/server computing environment.

Introduction to XeLDA

XeLDA provides a generic engine framework for linguistic treatment of text applications, based on an open architecture which easily admits extensions integrating future Xerox-proprietary finite-state research results.
Potential applications of the XELDA engine include: comprehension aids and translation over a network, on a desktop or a portable PC, syntax checking, terminology extraction, and authoring tools in general. XELDA describes the architecture of the server, running on a Unix or Windows computer. It was designed as such a framework to provides developers and researchers with a common development architecture for the open and seamless integration of linguistic services.

Technology Background

The XeLDA toolkit is based and integrates results from the Content Analysis (CA) group at Grenoble. The Finite State Calculus Technology is the fundamental one and gives a major competitive advantage. This allows linguists to build a language model with a flexible rule base to cope with the complexity of language evolution. Based on this technology, XeLDA offers fast, powerful and accurate linguistic services.

Back to Past-Projects page.