Retrieval of Partial Documents.

Abstract

Introduction Provision of answers to informally phrased questions is a central part of information retrieval. These answers traditionally take the form of documents retrieved from a text database, but documents will often be unsatisfactory as answers. They may be large and unwieldy; the answer they represent may be diffuse, and therefore hard for the user to extract; and word-based retrieval systems may be misled by the breadth of vocabulary of a long document into believing it to be relevant. Indexing and returning parts of documents addresses these problems. We have approached the problem of partial documents in two ways. The first approach is to regard documents as an unstructured series of "pages" of text of similar length, each of which can be returned as an answer to a query. We would expect, under this approach, that any bias in the retrieval mechanism towards documents of a particular length should be eliminated. By regarding an answer to be the document from which an answer