Web Image Interpretation Using Graph Grammar

Abstract

Web knowledge extraction becomes a hot topic once the invention of World Wide Web, as a result of the massive
quantity of information on the online makes it difficult to retrieve helpful information. Thanks to the varied styles
and show of information on completely different websites, it’s onerous to implement a general idea to extract
knowledge completely different websites. This paper presents a unique technique supported graph grammar to extract the constant sort of data from completely different Web sites without the requirement of coaching or adjustment. Our approach formalizes a standard Web pattern as a graph grammar. Then, based on the visual layout and
HTML DOM structure, a Web page is abstracted as a spatial graph that highlights the essential spatial relations
between data objects. According to the defined graph grammar, a spatial parsing is performed on the spatial graph
to extract structured records. We’ve evaluated our approach on twenty completely different Web sites, and achieved
the Final Score as 97.47% which shows promising performance.