querying semantically tagged documents on the web.pdf

Querying Semantically Tagged Do cuments on the Web  Ziv BarYossef y Yaron Kanzaz Yakov Kogan z Werner Nutt x Yehoshua Sagiv z Abstract We have develop ed and implemented QUESTa system for QUErying Semantically Tagged do cuments on the WorldWide Web The advent of new markup languages like xml makes it likely that do cuments on the Web are not simple html do cuments but contain in addition ob jects that represent the semantic structure of the do cument When such do cuments are available more p owerful metho ds to access and query in formation on the Web will b ecome p ossible We describ e the system QUEST which is designed to query do cuments in the markup language ohtml and to generate such do cuments as query results ohtml combines html and ob jects of a semistructured data mo del The new features of QUEST are it can b e used to query a combination of hyp ertext and ob ject mo del the results of queries are ohtml pages and thus of the same typ e as the data b eing queried QUEST implements a new approach for querying semistructured data that allows answers to b e incomplete Our exp erience in querying semantic do cuments on the Web can b e useful for the de sign and implementation of query languages for xml This pap er provides an overview of the QUEST system and of the exp erience gained from its design and implementation Intro duction The enormous growth in the usage of the WorldWide Web as a way to access information suggests that the Web will so on b ecome a more databaselike information source than it is to day The main obstacle for the evolution of the Web into one giant database is that most of the information accessible thro