Entity Mention Detection using a Combination of Redundancy-Driven Classifiers

摘要: We present an experimental framework for Entity Mention Detection in which two different classifiers are combined to exploit Data Redundancy attained through the annotation of a large text corpus, as well number Patterns extracted automatically from same corpus. In order recognize proper name, nominal, and pronominal mentions we not only information given by recognized within corpus being annotated, but also occurring external unannotated The system was first evaluated Evalita 2009 evaluation campaign obtaining good results. current version is used applications: on one hand, it LiveMemories project, aims at scaling up content extraction techniques towards very scale multimedia sources. On other annotate corpora, such Italian Wikipedia, thus providing easy access syntactic semantic both Natural Language Processing Information Retrieval communities. Moreover web service available going be integrated into TextPro suite NLP tools.