PROJECT TITLE :

Named Entity Disambiguation over Texts Written in the Portuguese or Spanish Languages

ABSTRACT:

This text addresses the problem of disambiguating named entities, in text documents, towards entries during a information base like Wikipedia. The proposed approach uses supervised learning to type candidate knowledge base entries for every entity mentioned in a text, and then to classify the entry ranked in the first position as either the right disambiguation or not. We tend to present results with Portuguese and Spanish texts for a wide selection of models and configuration options. Our experiments attest to the effectiveness of supervised learning ways in this specific task, showing that out-of-the-box algorithms and comparatively easy options will achieve a high accuracy.