2 DOCUMENT PROCESSING MODULE AS MENTIONED BEFORE, THE NUMBER OF DOC...

2.2 Document Processing Module

As mentioned before, the number of documents returned by The document processing module in QA systems is also the information retrieval system may be very large. Paragraph commonly referred to as paragraph indexing module, where filtering can be used to reduce the number of candidate the reformulated question is submitted to the information documents, and to reduce the amount of candidate text from retrieval system, which in turn retrieves a ranked list of each document. The concept of paragraph filtering is based on relevant documents. The document processing module the principle that the most relevant documents should contain usually relies on one or more information retrieval systems to the question keywords in a few neighboring paragraphs, rather gather information from a collection of document corpora than dispersed over the entire document. Therefore, if the which almost always involves the World Wide Web as at least keywords are all found in some set of N consecutive one of these corpora [7]. The documents returned by the paragraphs, then that set of paragraphs will be returned, information retrieval system is then filtered and ordered. otherwise, the document is discarded from further processing. Therefore, the main goal of the document processing module is to create a set of candidate ordered paragraphs that contain