Research on mathematical model for automatic summarization

Abstract

Automatic summarization is need of the era. The definitions, features, functions and classifications of summary are discussed in the paper. The problem of automatic summarization is described using a mathematic method and the solution is proposed. The solution makes use of meta-knowledge to describe the composition of the summary and help to calculate the semantic distance between summary and source document. According to these, the mathematical model of automatic summarization is established. It is concluded that the best summary is the summary that has the minimum sum of all the meta-knowledge weight among all the summaries. It is proposed that how to get meta-knowledge aggregate and their weight are the key problems in the model. There is also a discussion of how to select output from all the best summaries in the paper. The mathematic description of automatic summarization will be useful for the application of automatic summarization and the summary quality assessment.