In the three models discussed below, a document has a weight associated with each index term. This document weight is a measure of the degree to which the document is characterized by that term. Without loss of generality, we assume that document weights for all index terms lie in the range [0, 1]. This is less restrictive than in the standard Boolean model, which limits the values to the extremes of the range, namely 0 and 1.
To retrieve documents relevant to a given query, we need to calculate the query-document similarity for documents in the collection. The query-document similarity is an attempt to predict the relevance of a document to the query. In the following subsections, we consider each model and its method for calculating similarity.
In the three models discussed below, a document has a weight associated with each index term. This document weight is a measure of the degree to which the document is characterized by that term. Without loss of generality, we assume that document weights for all index terms lie in the range [0, 1]. This is less restrictive than in the standard Boolean model, which limits the values to the extremes of the range, namely 0 and 1.To retrieve documents relevant to a given query, we need to calculate the query-document similarity for documents in the collection. The query-document similarity is an attempt to predict the relevance of a document to the query. In the following subsections, we consider each model and its method for calculating similarity.
การแปล กรุณารอสักครู่..
