You are here

Integrating a Natural Language Message Pre-Processor with UIMA

TitleIntegrating a Natural Language Message Pre-Processor with UIMA
Publication TypeJournal Article
Year of Publication2008
AuthorsNyberg, E, Riebling, E, Wang, RC, Frederking, R
Abstract

This paper describes the use of the Unstructured Information Management Architecture (UIMA) to integrate a set of natural language processing (NLP) tools in the RADAR system. The challenge was to define a common data model and a set of component interfaces for these tools, so that they could be integrated into a single system. The integrated system is used to pre-process each email arriving in the RADAR user’s IMAP store. We present a UIMA collection processing engine for RADAR, including a common type system for text analysis results and annotators for each of the NLP tools. The paper also includes an analysis of system performance and a discussion of the lessons learned through use the of UIMA for this integration task. 1.

URLhttp://www.richardwang.com/papers/lrec-2008.pdf