• Publications
  • Influence
Type-Sensitive Knowledge Base Inference Without Explicit Type Supervision
TLDR
We propose an unsupervised typing gadget, which enhances top-of-the-line base models for KBC (DistMult, Complex) with two type-compatibility terms between entity-relation pairs, and combine the signals in a novel manner. Expand
  • 16
  • 4
  • PDF
Limited Attention and Centrality in Social Networks
TLDR
We propose a new measure of centrality - limited-attention version of Bonacich's Alpha-centrality - that models the effect of limited attention on epidemic diffusion. Expand
  • 19
  • PDF
Call me maybe: understanding nature and risks of sharing mobile numbers on online social networks
TLDR
We collected 76,347 unique mobile numbers posted by 85,905 users on Twitter and Facebook and analyzed 2,997 numbers, prefixed with +91. Expand
  • 8
  • PDF
Knowledge-Guided Linguistic Rewrites for Inference Rule Verification
TLDR
A corpus of inference rules between a pair of relation phrases is typically generated using the statistical overlap of argument-pairs associated with the relations (e.g., PATTY, CLEAN). Expand
  • 8
  • PDF
Mitigating the Effect of Out-of-Vocabulary Entity Pairs in Matrix Factorization for KB Inference
TLDR
This paper analyzes the varied performance of Matrix Factorization (MF) on the related tasks of relation extraction and knowledge-base completion, which have been unified recently into a single framework ofknowledge-base inference (KBI). Expand
  • 4
  • PDF
Knowledge Base Completion: Baseline strikes back (Again)
TLDR
We use 1-N scoring to compute a fullsoftmax (exact probability), instead of approximating it by using negative sampling, for improving performance of older models to match those of recent SOTA models. Expand
  • 3
  • PDF
Temporal Knowledge Base Completion: New Algorithms and Evaluation Protocols
TLDR
We present TIMEPLEX, a novel time-aware KBC method, that also automatically exploits the recurrent nature of some relations and temporal interactions between pairs of relations. Expand
Joint Matrix-Tensor Factorization for Knowledge Base Inference
TLDR
We perform an extensive evaluation to compare popular KB inference models across popular datasets in the literature. Expand