Application of VSM model to document structure identification
Wei Zhang · Journal of Beijing Information Science & Technology University · 2011
By referencing VSM model,the basic structure of document component is mapped into the characteristic vector.After normalization and weight assignment,the vector similarity between the standard document component structure and the structure of the document component to be identified can be calculated,thus the identification of whole document structure can be implemented.The method achieves vector formalization of document format.It is easily calculated,and has good expansibility.Applications such as automatic format checking or document understanding can be built on that basis.