The Information Workbench -
Download as PPTX, PDF0 likes160 views
The document discusses the Information Workbench, a platform for interacting with the Web of Data across its entire lifecycle. It integrates heterogeneous data sources, supports user-generated content, and addresses provenance, search, exploration, visualization, and publishing. The platform provides a hybrid search combining structured queries and keywords across structured and unstructured sources. It features a continuous, personalized living user interface with widgets and multiple interaction paradigms. A demo is available online and the platform aims to be an open, customizable solution for building end-user applications across domains.
The Information Workbench -
- 1. The Information Workbench Interacting with the Web of Data Peter Haase, Andreas Eberhart Thanh Tran Ulrich Walter Günter Ladwig Sebastian Godelet Andreas Wagner Tobias Mathäß Lei Zhang Claudiu Dragulin Rudi Studer
- 2. The Information Workbench • Addressing the entire lifecycle of interacting with the Web of Data – Integration of data sources – Content generation by the end user – Provenance – Search and Exploration – Visualization – Publishing • Integrated management of heterogeneous User- Wikipedia data sources generated DBpedia, Yago – Structured and unstructured – Published and user-generated Earthquake (Data.gov) – Static and dynamic – Open domain Structured Dynamic
- 3. Data Sources in the Application • Entire English Wikipedia • Data from Linked Open Data – DBpedia – YAGO –… • Data from Data.gov (US Government) – E.g. live data about earthquakes • Many more
- 4. Semantic Search • Hybrid Search: Structured queries combined with keywords across structured and unstructured data sources • Query interpretation: Translation of keywords into hybrid queries • Keyword search combined with faceted search: Iterative refinement process based on keywords and operations on facets
- 5. Living UI • Continuous, seamless and personal user experience across domains • Widget-based user interface • Multiple paradigms for interaction: browsing, visualization, editing, knowledge acquisition • Mashups with external sources • Automated selection of widgets based on available data • Customization and personalization
- 7. Conclusions • The meaning of data has to play a central role. – Integrated management of unstructured and structured / semantic data – Semantics exploited throughout the complete lifecycle of interaction with the data • Various, heterogeneous information sources – Management of real-life data from various sources, provenance – Heterogenous in: structured/unstructured, static / dynamic, published / user- generated • The application has to be an end-user application, i.e. an application that provides a practical value to domain experts. – Open world, open domain, provides value to end users across domains – Can be tailored to specific domains – Also applicable to enterprise scenarios: E.g. in Data Center Management
- 8. Conclusions • The application provides an attractive and functional Web interface • Widget-based, living UI • Web 2.0-like interaction • Functionality goes beyond pure information retrieval. The results should be as accurate as possible. • Addressing the complete lifecycle of the interaction with the data • Novel paradigms for search, enabling precise answers to complex information needs against hybrid data • There is a use of dynamic data • Integration of real-time, live-data sources • Multi-media documents are used in some way • Web 2.0-syle mashups with external sources, such as Youtube, Twitter • Scalability • Large unstructured corpus (incl. Wikipedia), large subset of LOD
- 9. Thank You!
- 10. Platform for Application Building • Custom providers for legacy and enterprise data sources • Easily extensible from the backend to the UI • Deployment on Cloud Infrastructures • Open Source release planned