Lect1 - INFS 740 Database Programming for the...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: INFS 740 Database Programming for the World Wide Web Professor Larry Kerschberg Department of Computer Science George Mason University 1 1 Outline of Lecture •  •  •  •  •  Modern Enterprise InformaHon Systems Role of Metadata in InformaHon Systems Role of XML in Modern InformaHon Systems Meta ­Data Standards Conclusions 2 Modern Enterprise InformaHon Systems •  InformaHon Systems Architectures –  SOA  ­ Service ­Oriented Architectures –  Mashups Rule! –  Intelligent Middleware  ­ Internet ­based three ­ Hered systems •  Web server, ApplicaHon Server, Database Server –  Preference is to access data in mulHple heterogeneous databases, combine the data and create value ­added informaHon. 3 Problems with SOA Architectures •  Scalability and Performance –  SOAP ­based Web Services depend on XML message size. –  Cohen notes in his FastSOA book: “At the intersecHon of XML and SOA are significant scalability and performance problems that are found using my test methodology and solved by using naHve XML technology, including XQuery and naHve XML databases.” 4 SOAP Service Scalability Index 5 SOA Concerns for IT Managers 6 Composite Applica.ons •  Composite ApplicaHons create value by delivering new views, transformaHons, and persistent queriable stores of data, connected through a business process workflow to a presentaHon interface such as a web browser. •  First, composite applicaHons leave data as is, that is, they do not require data providers to transform the data; • Second, composite applicaHons, once created, may be treated as the database itself; • Third, as organizaHons respond to change, the composite applicaHons deliver a way to store and reuse all the data, including the metadata about the database and the applicaHon itself. 7 Data Services Rather than extract the data from databases, the SOA approach creates Data Services to link exisHng data in new and valuable composites. Data Services are a flexible way to bridge the gap between exisHng data and the needs of a dynamic business process. Also called Enterprise ApplicaHon IntegraHon, and workflow management. 8 Role of Metadata in InformaHon Systems •  We have metadata everywhere! •  RelaHonal Database Schemas – Data DefiniHon Language (DDL) describes relaHons, a`ributes, data types, formats, etc. •  Object and RelaHonships in UML, including meta ­models; •  ScienHfic data – provenance and lineage. 9 XML Introduc.on •  XML hails from markup languages used for document formabng; •  Appeared in the late 1990s; •  Used to describe the structure of documents; •  Used to “free” data from relaHons and into semi ­structured form. •  Enabled Message ­Oriented Middleware (MOM) to allow apps to exchange informaHon; • Facilitated Remote Procedure Calls, and this lead to Web Services (SOAP) and later to SOA and business process languages (BPEL) and the Enterprise Service Bus (ESB). •  Other interesHng protocols are RSS ­Feeds (Really Simple SyndicaHon) and Podcasts (Hme ­ shifed content via subscripHons). 10 Example of a Business Process The Business Process ExecuHon Language (BPEL) can be used to specify acHviHes related to a business process that composes several Web Services. The figure shows the orchestraHon of a Loan ApplicaHon workflow, and how BPEL denotes the states of the workflow in XML form. 11 XML Impact on Data • RelaHonal Databases dominated the marketplace in the 1980 ­1990s; •  They sHll dominate, but other data models are supported • Object ­relaHonal databases • XML Databases • MulHdimensional databases •  SQL is the standard relaHonal query language based on relaHonal calculus; • XML data is semi ­structured and not relaHonal; •  XML Databases are a combinaHon of XML Documents and the XML Query Language (XQuery); •  The XML Data Model appears to be more intuiHve to organize complex data. 12 XML Impact on Business •  XML is used in business to exchange contracts and documents; supplanHng EDI systems. •  Complex Data ­Interchange Schemas. Industries have formed standards bodies to create and adopt XML document interchange. • STAR BOD Schema for the Auto Industry h`p://www.starstandard.org/ •  HL7 Schema in the Health Care Industry h`p://www.hl7.org/ • E ­Business Standards h`p://ebxml.org/ • OASIS (UN ­Sponsored) h`p://www.oasis ­open.org/home/ index.php •  XML is used to define business protocols and document exchanges to fulfill standard business processes. •  Dynamic Documents with embedded XML and links to latest informaHon. 13 Enterprise Knowledge CreaHon and DistribuHon •  •  •  •  •  •  Acquire data and informaHon from mulHple, possibly heterogeneous sources, IntegraHon of informaHon, tagging of informaHon with semanHc tags, Create intellectual property (IP) with valued ­added processing, Protect IP products, processes and resources, Share knowledge with partners, Distribute IP products to customers and partners. 14 E ­Business Knowledge CreaHon KM XML 0101010101 0101010101 Generate Information Information Integration Indexing (XML Meta Tags) Capture Information: • Federated Databases • Web Searching • Intelligent Agents • Knowledge Rovers • XML Messages • Email Messages Process & Manage Information Knowledge Management KM Materialize Extract Information: • Data Mining • Decision Support • AI • Data Warehousing Risk Management Publish & Share Convert To Knowledge XML Public Domain & WWW Knowledge Base Security Concerns? Push Publishing? XML RosettaNet SOAP WIDL ebXML Customers XML Web Services XML XML Role Knowledge Management Role Suppliers & Business Partners KM E-Business Data Acquisition and Knowledge Creation with XML as the Enabler Courtesy of Mr. Gus Jabbour 15 Role of Metadata in Knowledge Management •  Metadata is data-about-data and is used to describe the a`ributes of a resource.     Metadata is used in several KM acHviHes: search, discovery, documentaHon, refinement, and disseminaHon. These acHviHes may be carried out by human end ­users or their (human or automated) agents. •  •  Metadata is needed in the Internet context to enhance precision of informaHon retrieval. Metadata may be embedded within a document (metatags) or they may be external to the document. 16 Metadata Standards IniHaHves •  •  •  •  •  XML (eXtensible Markup Language) Dublin Core Metadata IniHaHve for Digital Libraries, Dublin Core is an internaHonal iniHaHve hosted by OCLC W3C  ­ RDF, (WWW ConsorHum) Resource DescripHon Framework W3C  ­ SemanHc Web, DAML+OIL. Web Services 17 Content Indexing and Tagging of InformaHon Resources •  •  Research in automaHc classificaHon at OCLC includes the Scorpion Project for Dewey Decimal ClassificaHon. Commercial products from Autonomy and Convera: •  •  Use Bayesian Networks and Neural Networks to formulate concepts automaHcally, not just keyword extracHon. Use text mining to correlate related concepts found in heterogeneous documents. •  •  AutomaHc tagging will help analysts to create knowledge and link back to original sources. DARPA Agent Markup Language (DAML) program is creaHng a tool set for markup of SemanHc Web ontologies and services. 18 Dublin Core (DC) Metadata IniHaHve •  Simplicity – the DC is intended to be usable by non ­ catalogers as well as resource descripHon specialists. •  SemanHc Interoperability – diverse descripHon models hinder sharing and understanding across disciplines. •  InternaHonal Consensus – parHcipants are from all over the world. •  Extensibility – may be extended to include more specialized structure and semanHcs. •  Metadata Modularity on the Web – brings Digital Library perspecHve to encoding metadata on the WWW. 19 Dublin Core Metadata Types Content Title Subject DescripHon Type Source RelaHon Coverage 20 Intellectual Property Creator Publisher Contributor Rights Instan.a.on Date Format IdenHfier Language Conclusions •  XML is the lingua franca of the Web; •  XML markup languages and standards which lead to data services, data interoperability, web services, new informaHon architectures; •  XML provides more flexibility in handling semi ­structured heterogeneous data. •  We will be studying various aspects of XML in this course. 21 ...
View Full Document

This note was uploaded on 10/24/2009 for the course INFS 740 taught by Professor Brodsky,a during the Fall '08 term at George Mason.

Ask a homework question - tutors are online