GRAPHICAL MODELS FOR REPRESENTING TEXT DOCUMENTS FOR COMPUTER ANALYSIS

摘要

In a method of representing a text document in a graphical model, a document containing a plurality of ordered words is received and a graph data structure for the document is generated. The graph data structure includes a plurality of nodes and edges, each node representing an individual word in the document, and each edge identifying the number of times two nodes are within a predetermined distance from each other. The graph data structure is stored in the information store.