Hybrid systems for information extraction and question answering

Abstract

Information Extraction, Summarization and Question Answering all manipulate natural language texts and should benefit from the use of NLP techniques. Statistical techniques have till now outperformed symbolic processing of unrestricted text. However, Information Extraction and Question Answering require by far more accurate results of what is currently produced by Bag-Of-Words approaches. Besides, we see that such tasks as Semantic Evaluation of Text Entailment or Similarity-As required by the RTE Challenge, impose a much stricter performance in semantic terms to tell true from false pairs. We will speak in favour of a hybrid system, a combination of statistical and symbolic processing with reference to a specific problem, that of Anaphora Resolution which looms large and deep in text processing.