Back to results
US 7,406,459 B2Granted
Concept network
AssigneeMicrosoft Corporation
InventorsZheng Chen, Shengping Liu, Wei-Ying Ma
Life of this patent
Estimating the term…
Abstract
A concept network that can be generated in response to a user query. In one embodiment, the concept network is being used with a search tool. The search tool searches a plurality of data storage locations. Each data storage location is arranged with a node. Certain ones of the nodes are connected by at least one link. The concept network selects a portion of certain ones of the nodes based on the link, wherein the at least one link is used for content purposes.
Claims24
At least one computer readable medium having thereon at least one computer executable module for constructing concept networks comprising:
- a search tool for a plurality of data storage locations, each data storage location being arranged into a node, certain nodes being connected by at least one link, the at least one link comprising at least one hyperlink having a structure comprising a plurality of hyperlink blocks and each link being useable to gain access to at least one of the plurality of data storage locations;
- a first concept network that considers a portion of certain ones of the plurality of nodes based on the at least one link, wherein:
- the first concept network is constructed based on at least one hyperlink block of the structure of the at least one hyperlink;
- and
- the first concept network is merged with a second concept network based on a measure of mutual information between the first concept network and the second concept network, the measure of mutual information comprising calculating a mutual information MI for a word pair wi, wj in accordance with:
- MI(wi, wj)=Pr(wi, wj)log(Pr(wi, wj))/(Pr(w i)Pr(wj))
- MI(wi, wj)=Pr(wi, wj)log Pr(wi, wj)/Pr(wi)Pr(wj)
- wherein Pr(wi) corresponds to a probability that a first word wi of the word pair appears in a data structure, Pr(wj) corresponds to a probability that a second word wj, of the word pair appears in the data structure, and Pr(wi,wj) corresponds to a probability that both words wi, wj, of the word pair appear in the data structure;
- and
- a user interface configured to, at least, enable interaction with the search tool and display at least one search result based on the first concept network.
7 more elements
A computerized system for constructing concept networks comprising:
- a search tool for a plurality of data storage locations, each data storage location including a plurality of nodes;
- a first concept network constructed based on structural aspects of hyperlinks referencing the data storage locations, the structural aspects comprising that at least one of the hyperlinks comprises a plurality of hyperlink blocks, and wherein the first concept network is merged with a second concept network based on a measure of mutual information between the first concept network and the second concept network, the measure of mutual information comprising calculating a mutual information MI for a word pair Wi, wj in accordance with:
- MI(wi, wj)=Pr(wi, wj)log(Pr(wi, wj))/(Pr(w i)Pr(wj))
- wherein Pr(wi) corresponds to a probability that a first word wj of the word pair appears in a data structure, Pr(wj) corresponds to a probability that a second word
- MI(wi, wj)=Pr(wi, wj)log Pr(wi, wj)/Pr(wi)Pr(wj)
- wj, of the word pair appears in the data structure, and Pr(wi, wj) corresponds to a probability that both words wi, wj of the word pair appear in the data structure;
- and
- a user interface configured to, at least, enable interaction with the search tool and display at least one search result based on the concept network.
5 more elements
A computerized method of constructing a concept network comprising:
- analyzing a set of links referencing a set of data storage locations, the set of links comprising at least one hyperlink having a structure comprising a plurality of hyperlink blocks;
- arranging the set of data storage locations into a first set of concept network nodes;
- arranging a further concept network node into the first set of concept network nodes based on at least one hyperlink block of the structure of the at least one hyperlink;
- merging a second set of concept network nodes with the first set of concept network nodes based on a measure of mutual information between the first set of concept network nodes and the second set of concept network nodes, the measure of mutual information comprising calculating a mutual information MI for a word pair wi, wj in accordance with:
- MI(wi, wj)=Pr(wi, wj)log(Pr(wi, wj))/(Pr(w i)Pr(wj))
- MI(wi, wj)=Pr(wi, wj)log Pr(wi, wj)/Pr(wi)Pr(wj)
- wherein Pr(wi) corresponds to a probability that a first word wi, of the word pair appears in a data structure, Pr(wj) corresponds to a probability that a second word wj, of the word pair appears in the data structure, and Pr(wi, wj) corresponds to a probability that both words wi, wj of the word pair appear in the data structure;
- and
- in response to at least one user indication, storing data corresponding to at least one aspect of the concept network on a computer readable medium.
6 more elements
Dependencies are inferred from the claim text.