• Publications
  • Influence
Knowledge Graph Identification
TLDR
We develop a PSL model for knowledge graph identification that both captures probabilistic dependencies between facts and enforces global constraints between entities and relations. Expand
  • 182
  • 10
  • PDF
Sparsity and Noise: Where Knowledge Graph Embeddings Fall Short
TLDR
We consider the problem of applying embedding techniques to KGs extracted from text, which are often incomplete and contain errors. Expand
  • 37
  • 2
  • PDF
Personalized explanations for hybrid recommender systems
TLDR
We build upon a hybrid probabilistic graphical model and develop an approach to generate real-time recommendations along with personalized explanations for hybrid recommender systems. Expand
  • 36
  • 2
  • PDF
Ontology-aware partitioning for knowledge graph identification
TLDR
We show that using a richer partitioning model that incorporates the ontology graph and distribution of extractions can result in order-of-magnitude speedups without reducing model performance. Expand
  • 13
  • 2
  • PDF
User Preferences for Hybrid Explanations
TLDR
We describe a hybrid recommender system built on a probabilistic programming language, and discuss the benefits and challenges of explaining its recommendations to users. Expand
  • 41
  • 1
  • PDF
Collective Entity Resolution in Familial Networks
TLDR
We present the results for both our method and the baselines and only for the positive class (co-referent entities). Expand
  • 18
  • 1
  • PDF
Disambiguating Energy Disaggregation: A Collective Probabilistic Approach
TLDR
We introduce a probabilistic framework which infers the energy consumption of individual appliances using a hinge-loss Markov random field (HL-MRF), which admits highly scalable inference. Expand
  • 10
  • 1
  • PDF
Learning Semantic Models of Data Sources Using Probabilistic Graphical Models
TLDR
We present a novel approach that efficiently searches over the combinatorial space of possible semantic models, and applies a probabilistic graphical model to identify the most probable semantic model for a data source. Expand
  • 3
  • 1
Scalable Probabilistic Causal Structure Discovery
TLDR
We develop an approach using probabilistic soft logic (PSL) that exploits multiple statistical tests, supports efficient optimization over hundreds of variables, and can easily incorporate domain knowledge and other structural constraints. Expand
  • 3
  • 1
  • PDF
Using classifier cascades for scalable e-mail classification
TLDR
Adaptive Classifier Cascades designs a policy to combine a series of base classifiers with increasing computational costs given a desired trade-off between cost and accuracy. Expand
  • 21
  • PDF