Algorithms for Collective Knowledge Acquisition
Abstract
Collective knowledge bases (CKBs) allow knowledge from a multitude of sources to be efficiently gathered, integrated and deployed. Markov logic provides a representation language and learning and inference algorithms for CKBs. The goal of this project was to develop solutions to several problems that need to be addressed before CKBs can be widely deployed, including: tightly coupling learning and inference, learning many levels of structure, finding complex mappings across representations, optimizing joint inference, explaining the results of inference, and accepting natural language input.
Document Details
- Document Type
- Technical Report
- Publication Date
- May 08, 2012
- Accession Number
- ADA561728
Entities
People
- Pedro Domingos
Organizations
- University of Washington