AN IMPROVED METHOD AND SYSTEM FOR FINDING A FOCUS OF A DOCUMENT

摘要

A method and apparatus for finding the focus of a document are provided. A semantic network comprising a plurality of nodes, each representing a concept, and links, connecting the nodes, representing relations between the concepts is used. The method comprises the steps of providing a list of terms in an input document which match concepts represented by the semantic network, as well as a frequency value for each matched term indicating the number of times the term appears in the document; mapping each matched term to a referent node or to a plurality of possible referent nodes in the semantic network, and assigning weights to nodes. Assigning weights to nodes comprises carrying out the following steps for each matched term. Selecting one of the referent nodes as a start node; assigning the start node an initial weight depending on the frequency value of the term; carrying out a depth- first traversal of the semantic network, and then repeating these steps for a next referent node for the term. The depth- first traversal starts from the start node, and at each node in the traversal calculates a weight of the current node based on the weight of the previous node in the traversal and a stepping factor, with the search backtracking to previous nodes when the weight of a current node is found to be less than a predefined weight threshold. These steps are repeated for the next matched term. Then, the weights for each node from all traversals are added together and a focus of the input document is determined by identifying the node having the heaviest weight. The focus can then be used to carry out disambiguation of any ambiguous matched terms. The method can be repeated to find one or more additional foci to enable further disambiguation of terms.