Requirement Analysis of Knowledge Acquisition from Texts, Deliverable ID 5.1

The goal of this document is to present the requirements for Knowledge Ac-quisition from text in the context of the X-Media project. Acquisition in X-Media will be performed via Automatic Document Annotation based on Information Extraction from text. The requirements analysis here presented focuses on key limitations of the existing IE technology with respect to the project's goals, namely limited capability of coping with large-scale (both considering the size of corpora and that of the ontology and of the KB), lack of portability of acqui-sition models, limited or no user interaction methodology and limited use of background knowledge.