1 / 12

Topic Maps 2008 Enterprise Knowledge Integration Using Topic Maps

Topic Maps 2008 Enterprise Knowledge Integration Using Topic Maps. Heimo Hänninen*. * a founding member of WSSSL. Goals (other than “go try Topic Maps”…). Improved user experience (consistency) One stop shopping for information Better search based on harmonized metadata.

modesta-curtis
Download Presentation

Topic Maps 2008 Enterprise Knowledge Integration Using Topic Maps

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Topic Maps 2008Enterprise Knowledge Integration Using Topic Maps Heimo Hänninen* * a founding member of WSSSL Topic Maps 2008

  2. Goals (other than “go try Topic Maps”…) • Improved user experience (consistency) • One stop shopping for information • Better search based on harmonized metadata • Improved integration model • Reusing business logic (service) • Simplifies application development Heimo Hänninen / Topic Maps 2008

  3. Business viewpoint: Pain points in current portal(Portal supports users in operating and maintaining their telecom network) • Users complain they cannot find the information they need • Search is poor • Users are used to more advanced functionalities on consumer portals and expect to have them in our portal as well • Inflexibility of the architecture to provide • Different views (dynamic views based on context) • Linking capabilities – linking is mostly hard-coded • Portal does not guide the users in any way • High development costs because of inflexible system architecture • one-to-one integration model • no business logic reuse (redundancy) Heimo Hänninen / Topic Maps 2008

  4. Business goals 1/2 • Increased customer satisfaction and productivity through content discovery • Not only the users can find the information but they will discover information that they did not know of • Enhanced user experience: information is presented in a user-friendly way and information access methods are user-friendly • Seamless user journeys from topic to topic, irregardless of “content domain” and owning organization • Customization: user sees only their products (but can also access full product portfolio if so wish) • Personalization: user can indicate which products (and releases) they are working on and will see personalized views • Filtering and sorting capabilities to drill down to desired information • Guiding the user to find the essential data rather than trying to provide all the possible data • Intelligent search (ontology-savvy search) Heimo Hänninen / Topic Maps 2008

  5. Business goals 2/2 • Process benefits • People who produce the contents for distribution do not have to be aware of all possible content that can be linked to their content – linking patterns can be automated • Navigation can be automatically populated • System benefits • Data integration and mappings • No point-to-point integrations Heimo Hänninen / Topic Maps 2008

  6. Knowledge Integration System evolution a lot more months 3 months 18 months PoC “Dynamo” Portal app. “PIC” Portal Wide, Other applications JSP App Web Apps Web Apps App. X presentation application logic REST REST as needed Enterprise Ontology & Glossary Service Enterprise Service KI logic EAI EAI manual messages XML/ASCII messages XML/ASCII data sources Heimo Hänninen / Topic Maps 2008

  7. Cost savings by integration model and service approach point-to-point many-to-one App 1 App 2 App n App 1 App 2 App n biz logic integration integration 0.5*n kEur n kEur n kEur 4*n kEur KI Service biz logic 4*n kEur any EAI integration integration EAI EAI EAI integration analysis, integration analysis, integration PDM CMS PDM CMS CRM Heimo Hänninen / Topic Maps 2008

  8. Selecting technology to build KI application • complete In-house: build application on top of RDBMS • mid effort In-house: application & XML abstraction on RDBMS or XML-DBMS • lower effort In-house: application on “KI savvy engine/ technology”  way to go 1 2a 2b 3a 3b 3c Service Layer KI Application (Java, Python…) KI Application (Java, Python…) KI App KI App KI App KI Application (Java, Python…) API API XML JDK API XML DB-engine Topic Maps Engine RDF Engine XML API XML API Oracle RDF-layer -DBC SQL SQL SQL SQL SQL Xquery Oracle Spatial XML-DB RDBMS RDBMS RDBMS RDBMS Heimo Hänninen / Topic Maps 2008

  9. Lessons learnt • Topic Maps are flexible yet easy to understand and model • PSI type of concept is mandatory in a large company • Back-end integrations are always more laborious than you think • policies, security, data ownership, pipelines, mapping, data quality… • Invest in services, create interfaces as needed • Colliding interests – gap between EA and business: • EA focuses on governance – top down • Innovations are made in grassroots projects to fulfill biz needs – bottom up • Iterative mode of work, make mistakes early, find right people • Creating a KI system takes two years – maturing process and changing mind set takes a decade • Working ontology is a bottom-up effort, yet must keep in synch with corporate guidelines • Demand SW vendors to support “Subject-centric computing” Heimo Hänninen / Topic Maps 2008

  10. Predicting future of semantic technology – Blue Ray or HD DVD dilemma ? • Google “study” - how unscientific is that?  • RDF vs. Topic Maps = 10:1 • RDF: google for "RDF software“ =10 800 hits • 1st: Welcome to RDF Software, home of the Structural Pest Control System for Windows  • Topic Maps: google for: "Topic Maps software“ = 938 hits • at Microsoft 20:1 • "RDF" site: microsoft.com = 1140 hits • "Topic Maps" site:microsoft.com =59 hits • at IBM 22:1 • "RDF" site:ibm.com =5 350 hits • "Topic Maps" site:ibm.com = 248 hits • at Oracle 450:1 • "RDF" site:oracle.com =3 600 hits • "Topic Maps" site:oracle.com = 8 hits • … or perhaps some day a semantic gadget by Heimo Hänninen / Topic Maps 2008

  11. Ontology technology choicesLIFE IS HARD, BUT WE FEEL GOOD Heimo Hänninen / Topic Maps 2008

  12. … most likely both Topic Maps and RDF will have their place in the sun Thank you and questions, please Heimo Hänninen / Topic Maps 2008

More Related