Advances in Computer Science and Engineering
Volume 6, Issue 2, Pages 199 - 204
(May 2011)
|
|
CONSTRUCTING CANDIDATE ATTRIBUTE DYNAMICALLY FOR INFORMATION EXTRACTION
Bo-Hyun Yun
|
Abstract:
The candidate attribute technique for the information extraction makes it possible to extract the information besides predefined template attribute. Because conventional information extraction methods are to extract the predefined template attribute, they do not template attribute dynamically. In order to meet this problem, we present the method that selects the issued topic with named entities in the document and performs the grouping by the clustering method, and then it recommends the topic of the group as the candidate attributes of the template. |
| Keywords and phrases: information extraction, candidate attribute recommendation, template. |
| |
| Number of Downloads: 310 | Number of Views: 857 |
|
