Difference between revisions of "Chinese Text Filtering based on Domain Keywords Extracted from Wikipedia"
(Starting a page: Chinese Text Filtering based on Domain Keywords Extracted from Wikipedia) |
(wikilinks) |
||
| Line 1: | Line 1: | ||
| − | '''Chinese Text Filtering based on Domain Keywords Extracted from Wikipedia''' - scientific work related to Wikipedia quality published in 2013, written by Xiang Wang, Hu Li, Yan Jia and SongChang Jin. | + | '''Chinese Text Filtering based on Domain Keywords Extracted from Wikipedia''' - scientific work related to [[Wikipedia quality]] published in 2013, written by [[Xiang Wang]], [[Hu Li]], [[Yan Jia]] and [[SongChang Jin]]. |
== Overview == | == Overview == | ||
| − | Several machine learning and information retrieval algorithms have been used for text filtering. All these methods have a common ground that they need positive and negative examples to build user profile. However, not all applications can get good training documents. In this paper, authors present a Wikipedia based method to build user profile without any other training documents. The proposed method extracts keywords of a special category from Wikipedia taxonomy and computes the weights of the extracted keywords based on Wikipedia pages. Experiment results on Chinese news text dataset SogouC show that the proposed method achieves good performance. | + | Several machine learning and [[information retrieval]] algorithms have been used for text filtering. All these methods have a common ground that they need positive and negative examples to build user profile. However, not all applications can get good training documents. In this paper, authors present a [[Wikipedia]] based method to build user profile without any other training documents. The proposed method extracts keywords of a special category from Wikipedia taxonomy and computes the weights of the extracted keywords based on Wikipedia pages. Experiment results on Chinese news text dataset SogouC show that the proposed method achieves good performance. |
Revision as of 01:04, 9 June 2019
Chinese Text Filtering based on Domain Keywords Extracted from Wikipedia - scientific work related to Wikipedia quality published in 2013, written by Xiang Wang, Hu Li, Yan Jia and SongChang Jin.
Overview
Several machine learning and information retrieval algorithms have been used for text filtering. All these methods have a common ground that they need positive and negative examples to build user profile. However, not all applications can get good training documents. In this paper, authors present a Wikipedia based method to build user profile without any other training documents. The proposed method extracts keywords of a special category from Wikipedia taxonomy and computes the weights of the extracted keywords based on Wikipedia pages. Experiment results on Chinese news text dataset SogouC show that the proposed method achieves good performance.
