Hybrid Text Summarization Method based on the TF Method and the Lead Method

This paper describes a hybrid text summarization method based on a TF-based sentence extraction method and a LEAD sentence extraction method. The LEAD method is known to be effective than other methods for document summarization of newspapers in lower summarization (output-to-input) ratio. In order to combine the LEAD method with the TF method, we used a rectangular distribution function that determines the importance of sentences according to their position in a document. With our method, the importance of a sentence is determined by multiplying the TF-based score and the distribution function. We conducted open test evaluation using the formal run test data of sentence extraction sub-task in NTCIR-2 Workshop TSC task (30 newspaper articles). The proposed method was tested by the average values of F-measure for 10%, 30%, and 50% summaries, and proved 34.1% for TF method, 39.1% for LEAD method, and 42.4% for the proposed method.