Method of building a geo-tree
First Claim
Patent Images
1. A method of building a geo-tree structure, wherein a geo-tree corresponds to a logical tree having nodes comprising geographic information, comprising:
- using a plurality of first media items each having at least media, a geographic tag, and a contextual tag to assign geographic information to a second media item having at least media and a contextual tag but without a geographic tag bycollecting the plurality of first media items;
associating the geographic tags of the plurality of first media items with the contextual tags of the plurality of first media items, wherein the contextual tags of the plurality of first media items and the second media item include words or terms describing content of the media of the respective one of the plurality of first media items and the second media items, and wherein at least one same word or term of the contextual tags of the plurality of first media items are associated with a plurality of geographic tags;
clustering the collected first media items based on the association of the geographic tags of the plurality of first media items with the contextual tags of the plurality of first media items;
classifying the clusters using at least one of the words or terms of the contextual tags of the plurality of first media items;
building a geo-tree with each classified cluster being associated with a node on the tree, wherein the nodes of the geo-tree comprise geographic information;
storing the geo-tree on a computer readable medium; and
assigning the second media item to a cluster in the geo-tree based on words or terms in the contextual tag associated with the second media item matching a word or term of one of the classified clusters so that the second media item is associated with the geographic information of the node of the assigned cluster.
1 Assignment
0 Petitions
Accused Products
Abstract
A method of building a geo-tree includes collecting a plurality of data items having geographic and contextual tags, associating the geographic tags with the contextual tags, clustering the collected data items based on the contextual tags, classifying the clusters based on the contextual tags, building a geo-tree with each classified cluster being associated with a node on the tree and storing the geo-tree.
26 Citations
17 Claims
-
1. A method of building a geo-tree structure, wherein a geo-tree corresponds to a logical tree having nodes comprising geographic information, comprising:
using a plurality of first media items each having at least media, a geographic tag, and a contextual tag to assign geographic information to a second media item having at least media and a contextual tag but without a geographic tag by collecting the plurality of first media items; associating the geographic tags of the plurality of first media items with the contextual tags of the plurality of first media items, wherein the contextual tags of the plurality of first media items and the second media item include words or terms describing content of the media of the respective one of the plurality of first media items and the second media items, and wherein at least one same word or term of the contextual tags of the plurality of first media items are associated with a plurality of geographic tags; clustering the collected first media items based on the association of the geographic tags of the plurality of first media items with the contextual tags of the plurality of first media items; classifying the clusters using at least one of the words or terms of the contextual tags of the plurality of first media items; building a geo-tree with each classified cluster being associated with a node on the tree, wherein the nodes of the geo-tree comprise geographic information; storing the geo-tree on a computer readable medium; and assigning the second media item to a cluster in the geo-tree based on words or terms in the contextual tag associated with the second media item matching a word or term of one of the classified clusters so that the second media item is associated with the geographic information of the node of the assigned cluster. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
11. A user equipment comprising:
-
a transceiver for communicating data; a memory for storing a plurality of first media items each having at least media, a geographic tag, and a contextual tag; and a processor for; using the first plurality of media items to assign geographic information to a second media item having at least media and a contextual tag but without a geographic tag by associating the geographic tags of the first plurality of media items with the contextual tags of the plurality of first media items, wherein the contextual tags of the plurality of first media items and the second media item include words or terms describing content of the media of the respective one of the plurality of first media items and the second media items, and wherein at least one same word or term of the contextual tags of the plurality of first media items are associated with a plurality of geographic tags; clustering the collected first media items based on the association of the geographic tags of the plurality of first media items with the contextual tags of the first plurality of media items; classifying the clusters using at least one of the words or terms of the contextual tags; building a geo-tree structure, wherein a geo-tree corresponds to a logical tree having nodes comprising geographic information, with each classified cluster being associated with a node on the tree; and assigning the second media item to a cluster in the geo-tree based on words or terms in the contextual tag associated with the second media item matching a word or term of one of the classified clusters so that the second media item is associated with the geographic information of the node of the assigned cluster. - View Dependent Claims (12, 13, 14, 15)
-
-
16. A server comprising:
-
a communication interface for communicating data; a memory for storing a plurality of first media items each having at least media, a geographic tag, and a contextual tag; and a processor for; using the first plurality of media items to assign geographic information to a second media item having at least media and a contextual tag but without a geographic tag by associating the geographic tags of the first plurality of media items with the contextual tags of the plurality of first media items, wherein the contextual tags of the plurality of first media items and the second media item include words or terms describing content of the media of the respective one of the plurality of first media items and the second media items, and wherein at least one same word or term of the contextual tags of the plurality of first media items are associated with a plurality of geographic tags; clustering the collected first media items based on the association of the geographic tags of the plurality of first media items with the contextual tags of the first plurality of media items; classifying the clusters using at least one of the words or terms of the contextual tags; building a geo-tree structure, wherein a geo-tree corresponds to a logical tree having nodes comprising geographic information, with each classified cluster being associated with node on the tree; and assigning the second media item to a cluster in the geo-tree based on words or terms in the contextual tag associated with the second media item matching a word or term of one of the classified clusters so that the second media item is associated with the geographic information of the node of the assigned cluster. - View Dependent Claims (17)
-
Specification