Document searching method for retrieval of document from database, involves giving weight according to set of components representative of frequency of sequence, and selecting document according to result of comparison of vectors

摘要

The method involves comparing a vector (Y) that represents an input document (DOC) with another vector (V1) that represents a group of documents. An identifier of a sequence of letters is provided in the group of documents. A list of sequence of letters is generated from another group of documents. The weight is given according to a set of components representative of frequency of the sequence in the two groups of documents. One document (D) is selected according to a result of comparison of the vectors. Independent claims are also included for the following: (1) a device for searching a document (2) a computer program comprising instructions for performing a method for searching the document.