Projects

Archimedes: A Probabilistic Master Knowledge Base System

The Archimedes project aims at building a probabilistic master knowledge base system by combining novel system components and algorithms that we are designing and building at UF. In the context of the Archimedes project, we pursue a spectrum of research directions we are exploring at the UF Data Science Research (DSR) group including: query-driven and scalable statistical inference, probabilistic data models, state-parallel and data parallel data analytics framework, multimodal (e.g., text, image) information extraction, and KB schema enrichment. This line of research on supporting large-scale automatically extracted knowledge bases is of high impact for many application domains from medical informatics to ecology. We have received funding from industry as well as federal government including DARPA, EMC, Amazon and Google. Other related projects include DeepDive from Stanford, YAGO from Max Planck Institute, NELL from CMU as well as WikiData/Freebase and Google Knowledge Vault.

ProbKB	A probabilistic knowledge base constructed from web extractions and probabilistic inference.
DBlytics/MADLib	A statistical analysis library for relational databases and massively parallel processing systems.
Archer	An efficient query-driven algorithm that decreases computation costs of machine-learning, data-mining and other algorithms without sacrificing quality.
CAMeL	A crowdsource-based probabilistic database for the purpose of cleaning and maintaining automatically populated knowledge bases.
SigmaKB	A knowledge base that combines schema from multiple knowledge bases to produce a more powerful query answering system.
Rose	A system for knowledge extraction from electronic health records and knowledge exchange between machine intelligence and human.
SMARTeR	Knowledge-base enhanced information retrieval systems. Topic modeling applications in “survey clustering” and “relevance ranking of documents” in eDiscovery.
VITA	Multimodal knowledge extraction and fusion.