Modeling and Querying Web Data A Survey

1
Modeling and Querying Web DataA Survey

• By Li Lu

2
Overview

• Introduction
• Data Representation for Querying the Web
• Modeling and Querying the Web
• Summary and Future

3
Introduction

• Background
• The most common techniques used in searching
  information from the Web are based on sending
  information retrieval requests to index servers.
• Use web query techniques to locate, filter and
  present web information.
• Challenges
• Difficult to build a common model for Web.
• Hard to extract information from web data.

4
Data Representation for Querying the Web

• Graph Data Models
• Based on a labeled graph in which the nodes
  represent web pages, edges represent links
  between web pages, and the labels on the edges
  can be attribute names
• Capable of express navigational queries over the
  graph structure.
• Semistructured Data Models
• Based on labeled directed graphs. There is no
  restriction on the number of edges that can go
  out from a given node, or on the type of
  attribute value.
• Be able to query the schema or the labels on the
  edges of the graph

5
Data Representation for Querying the Web (cont.)
A Hypertree Containing a Publications Database
(WebOQL) AM98
6
Data Representation for Querying the Web (cont.)

• Semantic Web Data Models
• Semantic Web is a Web whose content can be
  annotated by metadata and be processed
  automatically by machines.
• The formulation of semantic assertions of
  semantic Web is based on Resource Description
  Framework (RDF) model LS99, which can be viewed
  as a partially labeled directed graph.
• They have the ability to exploit the semantics of
  the Web content and can provide better query
  result than their counterpart that based on the
  content and structure of the Web data.

7
Data Representation for Querying the Web (cont.)
An Example RDF Graph WWW1
8
Modeling and Querying the Web

• Query Languages for Graph Representation of
  Website
• The query languages combine both the
  content-based queries and structure-based
  queries. Therefore, they are able to formulate
  regular path expression queries and to express
  navigational queries over the graph structure.
• WebSQL MMM97, W3QL KS95, WebLog LSS96
• Example WebSQL MMM97

9
Modeling and Querying the Web (cont.)

• WebSQL MMM97
• Model of Web as a relational database with two
  virtual relations Document and Anchor.
• Documenturl, title, text, type, length, modif
• Anchorbase, href, label
• To map onto the graph structure of the WWW, each
  document in the Document relation is mapped to a
  node object in the graph and each hypertext link
  between two documents in Anchor relation is
  represented by a link object.

10
Modeling and Querying the Web (cont.)

• Sample query FLM98 to find a list of tuples of
  the form (d1, d2, label), where d1 is a document
  stored at local site, d2 is a document stored
  somewhere else, and d1 points to d2 by a link
  labeled label. Suppose all the local documents
  are reachable from www.mysite.start.
• SELECT d.url, e.url, a.label
• FROM Document d SUCH THAT
• www.mysite.start ? d,
• Document e SUCH THAT d gt e,
• Anchor a SUCH THAT a.base d.url
• WHERE a.href e.url

11
Modeling and Querying the Web (cont.)

• Query Languages for Semi-Structured
  Representation of Website
• To discover the implicit structure within the
  semistructured Web data and then recast the Web
  data to fit into the discovered structure
• WG-Log CDPT98, ULIXES and PENELOPE AMM97a,
  WebOQL AM98
• Example WebOQL AM98

12
Modeling and Querying the Web (cont.)

• WebOQL AM98
• Introduced a hypertree data structure. Hypertree
  is an ordered arc-labeled tree with two kinds of
  arcs, internal arcs and external arcs. Internal
  arcs are used to indicate structured objects and
  external arcs are used to indicate hyperlinks
  among objects. Arcs are labeled with records.

A Hypertree Containing a Publications Database
(WebOQL) AM98
13
Modeling and Querying the Web (cont.)

• Represent web pages by hypertree and mapping
  function. Mapping function is used to map URLs to
  corresponding hypertrees. The hypertree and
  mapping function are also called schema and
  browsing function of the Web respectively.
• Sample query FLM98 to extract the title and
  URL of the full version of papers authored by
  Smith from the csPapers database.
• SELECT y.Title, y.Url
• FROM x in csPapers, y in x
• WHERE y.Authors Smith

14
Modeling and Querying the Web

• Query Languages for Semantic Web
• Semantic web is a web whose content can be
  annotated by metadata and be processed
  automatically by machines.
• Semantic query has the ability to exploit the
  semantics of the Web content.
• RQL KACPS02, SquishQL MSR02 , TRIPLE
  SBAHKW02.

15
Summary and Future

• Summary
• Web data models are divided into three main
  categories graph data model, semistructured data
  model and semantic web data model.
• Based on these data models, Web query languages
  are also classified into three primary groups.
• Future
• To develop techniques to manipulate dynamic pages
  could be beneficial to Web query application and
  it may be a promising direction for future
  research.
• To combine the query result from different
  resource on the Web, especially the result from
  both structured and unstructured data sources
  also pose some challenges for future research.

16
References

• AM98 G. Arocena, A. Mendelzon, WebOQL
  Restructuring Documents, Databases, and Webs,
  Proc. ICDE'98, Orlando, Florida, Feb. 1998.
• CDPT98 S. Comai, E. Damiani, R. Posenato, L.
  Tanca, A Schema-based Approach to Modeling and
  Querying WWW Data, Proc. of FQAS'98, Roskilde,
  May 1998, LNAI 1495.
• AMM97a P. Atzeni, G. Mecca, P. Merialdo, To
  Weave the Web, International Conference on Very
  Large Data Bases (VLDB'97), Athens, Greece,
  August 26-29, 1997, pages 206-215.
• FLM98 D. Florescu, A. Levy, A. Mendelzon,
  Database Techniques for the World-Wide Web A
  Survey, SIGMOD Record 27, 3 (1998), 59-74.
• KACPS02 G. Karvounarakis, S. Alexaki, V.
  Christophides, D. Plexousakis, M. Scholl, RQL A
  Declarative Query Language for RDF, WWW2002, May
  2002, Honolulu, Hawaii.
• KS95 D. Konopnicki and O. Shmueli,
  W3QS A query system for the World Wide Web, In
  Proc. of the Int. Conf. on Very Large Data Bases
  (VLDB), pages 54-65, Zurich, Switzerland, 1995.
• LSS96 L. V. S. Lakshmanan, F. Sadri, L.
  N. Subramanian, A declarative language for
  querying and restructuring the Web, In Proc. of
  the sixth International Workshop on Research
  Issues in Data Engineering, RIDE96, New Orleans,
  February 1996.
• MM97 A. O. Mendelzon, T.
  Milo, Formal Models of Web Queries, Proceedings
  of the Sixteenth ACM Symposium on Principles of
  Database Systems, 134-143, 1997.
• MMM97 A. Mendelzon, G. Mihaila, T. Milo,
  Querying the world wide web, International
  Journal on Digital Libraries, 1(1)54-67, 1997.

17

• MSR02 L. Miller, A. Seaborne,
  A. Reggiori, Three Implementations of SquishQL,
  a Simple RDF Query Language, Proceedings of 1st
  International Semantic Web Conference. ISWC2002,
  Sardinia, Italy, June 9-12, 2002
• SBAHKW02A. Sheth, C. Bertram, D. Avant, B.
  Hammond, K. Kochut, Y. Warke, Semantic Content
  Management for Enterprises and the Web, IEEE
  Internet Computing, July/August 2002, pp.80-87,
  2002.
• WWW1 http//www.amk.ca/talks/semweb-intro,
  Introduction to the Semantic Web and RDF