Back to results
US 2003/0172058 A1Published application
Document similarity calculation apparatus, clustering apparatus, and document extraction apparatus
AssigneeFujitsu Limited
InventorIsao Namba
Life of this patent
Estimating the term…
Abstract
An input section inputs a document set. A normalization section calculates a similarity as a relative value between documents, with respect to combinations of a plurality of documents, in the document set. The normalization section employs the tf·idf method. In tf·idf method, a document vector and a significance of a word included in the document is used to perform normalization to convert each similarity to an absolute value.
Claims14
A document similarity calculation apparatus comprising:
- a similarity calculation unit which respectively calculates a similarity as a relative value between documents, with respect to combinations of a plurality of documents, using a document vector and a significance of a word included in the document;
- and
- a conversion unit which converts each similarity calculated by the similarity calculation unit to an absolute value by normalization.
A document similarity calculation method comprising:
- a similarity calculation step of respectively calculating a similarity as a relative value between documents, with respect to combinations of a plurality of documents, using a document vector and a significance of a word included in the document;
- and
- a conversion step of converting each similarity calculated at the similarity calculation step to an absolute value by normalization.
A document similarity calculation program for allowing a computer to function as:
- a similarity calculation unit which respectively calculates a similarity as a relative value between documents, with respect to combinations of documents, using a document vector and a significance of a word included in the document;
- and
- a conversion unit which converts each similarity calculated by the similarity calculation unit to an absolute value by normalization.
A clustering apparatus comprising:
- a similarity calculation unit which respectively calculates a similarity as a relative value between documents, with respect to combinations of a plurality of documents, using a document vector and a significance of a word included in the document;
- a conversion unit which converts each similarity calculated by the similarity calculation unit to an absolute value by normalization;
- and
- a clustering unit which executes clustering of a plurality of documents, based on the similarity of the absolute value.
1 more element
A clustering method comprising:
- a similarity calculation step of respectively calculating a similarity as a relative value between documents, with respect to combinations of a plurality of documents, using a document vector and a significance of a word included in the document;
- a conversion step of converting each similarity calculated at the similarity calculation step to an absolute value by normalization;
- and
- a clustering step of executing clustering of a plurality of documents, based on the similarity of the absolute value.
1 more element
A clustering program for allowing a computer to function as:
- a similarity calculation unit which respectively calculates a similarity as a relative value between documents, with respect to combinations of a plurality of documents, using a document vector and a significance of a word included in the document;
- a conversion unit which converts each similarity calculated by the similarity calculation unit to an absolute value by normalization;
- and
- a clustering unit which executes clustering of a plurality of documents, based on the similarity of an absolute value.
1 more element
A document extraction apparatus comprising:
- a similarity calculation unit which respectively calculates a similarity as a relative value between documents, with respect to combinations of a plurality of documents, using a document vector and a significance of a word included in the document;
- a conversion unit which converts each similarity calculated by the similarity calculation unit to an absolute value by normalization;
- a clustering unit which performs clustering of a plurality of documents, based on the similarity of the absolute value;
- a cluster sort unit which sorts the results of the clustering, using the number of documents constituting each cluster as a key;
- a representative document selection unit which selects a representative document from each cluster, with respect to the sorted results;
- and
- an output unit which outputs the representative documents in order corresponding to the sorted results.
4 more elements
A document extraction method comprising:
- a similarity calculation step of respectively calculating a similarity as a relative value between documents, with respect to combinations of a plurality of documents, using a document vector and a significance of a word included in the document;
- a conversion step of converting each similarity calculated at the similarity calculation step to an absolute value by normalization;
- a clustering step of performing clustering of a plurality of documents, based on the similarity of an absolute value;
- a cluster sort step of sorting the results of clustering, using the number of documents constituting each cluster as a key;
- a representative document selection step of selecting a representative document from each cluster, with respect to the sorted results;
- and
- an output step of outputting the representative documents in order corresponding to the sorted results.
4 more elements
A document extraction program for allowing a computer to function as:
- a similarity calculation unit which respectively calculates a similarity as a relative value between documents, with respect to combinations of a plurality of documents, using a document vector and a significance of a word included in the document;
- a conversion unit which converts each similarity calculated by the similarity calculation unit to an absolute value by normalization;
- a clustering unit which performs clustering of a plurality of documents, based on the similarity of the absolute value;
- a cluster sort unit which sorts the results of the clustering, using the number of documents constituting each cluster as a key;
- a representative document selection unit which selects a representative document from each cluster, with respect to the sorted results;
- and
- an output unit which outputs the representative documents in order corresponding to the sorted results.
4 more elements
Dependencies are inferred from the claim text.