Interactive Knowledge Capture in the New Millennium: How the Semantic Web Changed Everything -Yolanda Gil (USC InformationSciencesInstitute) Presenter: Rajat Jain USC Viterbi School of Engineering
Semantic Web The term was coined by Tim Berners-Lee for a web of data that can be processed by machines
Content for presentation • Introduction :New challenges in Semantic Web • Early Works • Discussion on Challenges • Conclusion
New Challenges in the Semantic Web One User Many Users User must be trained No time for training One System Many distributed subsystems One knowledge base Many knowledge bases Many reasoners One reasoner Expert user Varying quality and coverage
Early Work : The knowledge acquisition bottleneck • User teaching a computer how to perform the aspects of the task that he or she wanted to delegate • Knowledge acquisition tool developed for a specific type of problem solving task What is required ? A single knowledge acquisition system
More proactive and could less burden to the user More knowledge acquisition system can reason different kinds of gaps Better it can hypothesize how those gaps could be filled • Maintaining a knowledge base consistent Goal: Need a methodology to assess the quality of knowledge acquisition systems.
Challenge: How to acquire common knowledge from many knowledge Contributors ? • Many people willing to provide knowledge to computers How to keep volunteers engaged in contributing, and developed techniques to validate the knowledge acquired ?
Accommodate natural language statements • User interfaces that facilitated the collection of specific types of knowledge from users, including process knowledge and argumentation structures Working on new frameworks to create knowledge resources that incorporate alternative views within a community of contributors
Challenge : No training required Teaching a computer a task training requires • But user prefer to communicate knowledge in a manner that is natural to them • 90 million end users without having programming expertise will be interested in creating applications by 2012 Research ongoing: To develop an approach to learn procedures from human tutorial instruction given in natural language
Challenge: Functioning with Limited Knowledge Unsophisticated users can share knowledge about complex processes Gaps in the knowledge system • Ability to understand what it knows • To identify the missing knowledge Results into Goal: To provide a computer Investigation ongoing: Meta-reasoning capabilities to detect gaps and inconsistencies
Challenge: Distributed Problem Solving • Today knowledge and reasoning are physically and organizationally distributed Distributed Data sources Software and tools to process data Data repositories and Software repositories are distributed and managed separately from the workflow system
Solution: Develop matchers • Problem: Finding relevant components in distributed architecture Matchers Relevant agents To automate Natural language description Wings is a semantic workflow system that assists scientist with the design of computational experiments.
Challenge: Provenance and Trust • Knowledge is obtained from open sources results into unknown rights How to determine whether to trust a particular piece of information ?
Factors considered • Content Trust over Entity Trust • Identify factors that affect trust information origins • Models to derive source reputation • Tracking the origin of processed information • Provenance of knowledge base axioms • Semantic workflow representations to reason about metadata properties
Conclusion • Semantic Web has influenced the area of interactive knowledge capture • Demand from Internet users increases the challenges and impact on research in interactive knowledge capture
Questions Thank You