Text Similarity Cumulative model and algorithm research for dynamic multi-document summarization

In the Internet era, the effective organization of dynamic evolution network information, which improves the accessing efficiency, is an urgent need to solve the key issues. This paper describes dynamic evolution of network information, identifies and analyzes the document collection on the same topic in different stages. In order to construct dynamic of evolution Content differences, we present one dynamic multi-document summarization model, which is Text Similarity Cumulative Method model. On this basis, three efficient dynamic sentence weighting methods and one sentence selection method were proposed, some experiments were conducted on the test data of Update Summarization in TAC2008, the result show effectiveness.