Formalization and Implementation of Summary Generation

Abstract

This paper presents a formalization of the summary generation process, which takes a set of observed facts as input and generates its summary in the form of natural language text. We think human summalization process consists of the following four phrases : (1) to reconstruct the original semantic structure in human mind from observed facts, (2) to extract the characteristic facts from the reconstructed structure, (3) to build the surface structure in the macroscopic view and finally, (4) to generate natural language text from the surface structure. Our system generates newspaper-like text in the baseball game domain, and the implementation is also presented.