Region labeling method and device of data documents

Abstract

A region labeling method of data documents and a device thereof are provided. The region labeling method includes: obtaining a tree structure, which has a plurality of nodes including a plurality of administrative division names and iconic names with a hierarchical relationship therebetween; receiving the data document and retrieving at least one keyword from the data document; comparing the at least one keyword with the nodes to find a first node matching the at least one keyword; and labeling the first node and at least one father node of the first node to the data document.

What is claimed is: 
     
         1 . A region labeling method of a data document, the region labeling method comprising:
 obtaining a tree structure comprising a plurality of nodes that comprise a plurality of administrative area names and iconic names having a hierarchical relationship therebetween;   receiving the data document and retrieving at least one keyword from the data document;   comparing the at least one keyword with the plurality of nodes to find a first node matching the at least one keyword; and   labeling the first node and at least one father node related to the first node to the data document.   
     
     
         2 . The region labeling method according to  claim 1 , wherein a step of obtaining the tree structure comprises:
 obtaining a regional name;   determining whether the regional name belongs to a second node of the tree structure; and   adding the regional name to the tree structure,   wherein the regional name comprises the administrative area name and the iconic name.   
     
     
         3 . The region labeling method according to  claim 1 , wherein a step of retrieving the at least one keyword from the data document further comprises:
 retrieving the at least one keyword from a heading or a text of the data document; and   retrieving the at least one keyword from a source of the data document,   wherein the source of the data document comprises a place of occurrence related to the data document and a location of a provider of the data document.   
     
     
         4 . The region labeling method according to  claim 3 , wherein a step of comparing the at least one keyword with the plurality of nodes to find the first node matching the at least one keyword further comprises:
 defining a corresponding weighting for the at least one keyword respectively.   
     
     
         5 . The region labeling method according to  claim 4 , wherein the step of comparing the at least one keyword with the plurality of nodes to find the first node matching the at least one keyword further comprises:
 using the weighting corresponding to the at least one keyword as an order reference in comparing the at least one keyword with the plurality of nodes.   
     
     
         6 . The region labeling method according to  claim 1 , wherein the step of comparing the at least one keyword with the plurality of nodes to find the first node matching the at least one keyword further comprises:
 searching the tree structure to determine whether the first node exists in the tree structure, wherein the administrative area name or the iconic name included in the first node is identical to one of the at least one keyword; and   finding the at least one father node related to the first node in the tree structure.   
     
     
         7 . The region labeling method according to  claim 1 , wherein a step of labeling the first node and the at least one father node related to the first node to the data document further comprises:
 establishing a plurality of corresponding tags according to the first node and the at least one father node related to the first node; and   connecting the tags and the data document to complete the labeling, and storing the data document in a storage database.   
     
     
         8 . A region labeling device of a data document, the region labeling device comprising:
 a classifying unit obtaining a tree structure via a network, wherein the tree structure comprises a plurality of nodes that comprise a plurality of administrative area names and iconic names having a hierarchical relationship therebetween;   a retrieving unit receiving the data document via the network and retrieving at least one keyword from the data document;   a comparing unit coupled to the retrieving unit and comparing the at least one keyword with the plurality of nodes to find a first node matching the at least one keyword;   a labeling unit coupled to the comparing unit and labeling the first node and at least one father node related to the first node to the data document; and   a storage database coupled to the classifying unit, the retrieving unit, the comparing unit, and the labeling unit to store the tree structure and the data document.   
     
     
         9 . The region labeling device according to  claim 8 , wherein the classifying unit obtains the tree structure by:
 obtaining a regional name;   determining whether the regional name belongs to a second node of the tree structure; and   adding the regional name to the tree structure,   wherein the regional name comprises the administrative area name and the iconic name.   
     
     
         10 . The region labeling device according to  claim 8 , wherein the retrieving unit retrieves the at least one keyword from the data document further by:
 retrieving the at least one keyword from a heading or a text of the data document; and   retrieving the at least one keyword from a source of the data document, wherein the source of the data document comprises a place of occurrence related to the data document and a location of a provider of the data document.   
     
     
         11 . The region labeling device according to  claim 10 , wherein the comparing unit defines a corresponding weighting for the at least one keyword respectively. 
     
     
         12 . The region labeling device according to  claim 11 , wherein the comparing unit uses the weighting corresponding to the at least one keyword as an order reference in comparing the at least one keyword with the plurality of nodes. 
     
     
         13 . The region labeling device according to  claim 8 , wherein the comparing unit further executes:
 searching the tree structure to determine whether the first node exists in the tree structure, wherein the administrative area name or the iconic name included in the first node is identical to one of the at least one keyword; and   finding the at least one father node related to the first node in the tree structure.   
     
     
         14 . The region labeling device according to  claim 8 , wherein the labeling unit further executes:
 establishing a plurality of corresponding tags according to the first node and the at least one father node related to the first node; and   connecting the tags and the data document to complete the labeling, and storing the data document in the storage database.