Information extraction of BBS posting based on vision feature
Jing Feng Yang · Journal of Computer Applications · 2009
Aiming to the problem of traditional automatic extraction of posting information from topic pages of Bulletin Board System(BBS),a language independent technique was proposed.The solution performed the extraction just based on the visual information of topic pages.The paper concluded the visual features of BBS postings which guided entire extraction process.It was carried out in three steps:first construct the Visual Block Tree(VBT) of the topic page,then locate the posting region in the tree,and finally extract every posting information from the posting region.Experimental results indicate that the vision feature based approach can achieve high extraction accuracy.