System and method for indexing documents and retrieving similar document based on link transition count distance calculations
AssigneeNtt Docomo, Inc.
InventorsTakehiro Nakayama, Daisuke Torii
Life of this patent
Estimating the term…
Abstract
In an index generating device, a similarity calculating unit calculates access similarities indicating similarities of access histories between documents, based on history information indicating the access histories to the respective documents by users, and a similar document specifying unit specifies a similar document similar to a given document as a retrieval target, based on the access similarities or the like. A retrieval index generating unit generates a retrieval index for the given document as the retrieval target from words appearing in a document set consisting of the similar document and the given document as the retrieval target.
Claims12
An index generating system for generating a retrieval index for document retrieval, comprising:
- a memory that stores history information indicating access histories to respective documents by users, the history information including frequency information wherein access frequencies to the respective documents by the respective users are correlated with identification information of the users;
- and
- a processor that
- calculates access similarities indicating similarities of the access histories between the documents, based on the frequency information stored in the memory;
- analyzes a link structure among documents on a hypertext system in which the documents are interlinked and calculates a link transition count distance between the documents by counting a number of link transitions between the documents;
- specifies a similar document similar to a given document as a retrieval target by comparing the access similarities with a similarity determination threshold to specify the similar document, and specifies, as a document group, documents within a link transition count distance not more than a predetermined distance from the given document as the retrieval target and compares the access similarities between the documents in the document group and the given document as the retrieval target with the similarity determination threshold to specify the similar document;
- and
- generates a retrieval index for the given document as the retrieval target from words appearing in a document set consisting of the given document as the retrieval target and the similar document.
An index generating method for generating a retrieval index for document retrieval, comprising:
- storing, in a memory, history information indicating access histories to respective documents by users, the history information including frequency information wherein access frequencies to the respective documents by the respective users are correlated with identification information of the users;
- calculating, by a processor, access similarities indicating similarities of the access histories between the documents, based on the frequency information stored in the memory
- analyzing, by the processor, a link structure among documents on a hypertext system in which the documents are interlinked and calculating a link transition count distance between the documents by counting a number of link transitions between the documents;
- specifying, by the processor, a similar document similar to a given document as a retrieval target by comparing the access similarities with a similarity determination threshold to specify the similar document, and specifying, as a document group, documents within a link transition count distance not more than a predetermined distance from the given document as the retrieval target and comparing the access similarities between the documents in the document group and the given document as the retrieval target with the similarity determination threshold to specify the similar document;
- and
- generating, by the processor, a retrieval index for the given document as the retrieval target from words appearing in a document set consisting of the given document as the retrieval target and the similar document.
Dependencies are inferred from the claim text.