Contact:
Enego Agirre
Goal
This project aims to add meaning, knowledge and reasoning to current interface technologies. Specifically, KNOW will provide novel natural language interpretation and reasoning capabilities to current multilingual computer applications. In a significant step towards full Natural Language Understanding (NLU), KNOW will enhance current interfaces by automatically increasing the depth and breadth of the linguistic knowledge in existing multilingual resources, and by devising improved conceptbased Natural Language Processing (NLP) technologies that use those resources for advanced reasoning and inference.
In order to improve the existing wordbased Natural Language Processing technologies, KNOW will develop full syntactic parsers and semantic interpreters (including word sense disambiguation systems and semantic role labellers) for the languages involved in the project.
Furthermore, KNOW will supply a common conceptual structure, thus facilitating knowledge management of document content. This common conceptual repository will be shared across the project languages (English, Spanish, Catalan and Basque), thereby representing a truly multi and crosslingual knowledge base. Our aim is to develop efficient automatic deduction techniques and powerful inference implicitly/ explicitly knowledge represented into the knowledge base.
Current interface applications that can deal with meaning are restricted to limited domains. KNOW will open the way for a new generation of broadcoverage unrestricteddomain conceptbased language understanding applications with capabilities that significantly surpass those currently available. KNOW will demonstrate the feasibility of these technologies in two applications: Cross Lingual Information Retrieval and Question/Answering on three multimodal databases from EFE, Argazkipress and Zientzia eta teknika in all project languages.