Kanishka Misra

Purdue University
0000-0002-7261-5691
https://kanishka.xyz
Publications14
h-index2
Citations9
Highly Influential Citations0
Author pages are created from data sourced from our academic publisher partnerships and public sources.

Recommended Authors

  • Benjamin Van Durme
  • 200 Publications • 6,024 Citations
  • Tal Linzen
  • 77 Publications • 1,732 Citations
  • Samuel R. Bowman
  • 86 Publications • 7,975 Citations
  • Tatiana R. Ringenberg
  • 10 Publications • 29 Citations
  • Publications
  • Influence
Chat Analysis Triage Tool: Differentiating contact-driven vs. fantasy-driven child sex offenders.
Investigating crimes against children, specifically sexual solicitations, are complicated because not all offenders are contact-driven, meaning they want to meet the minor for sex in the physical… Expand
  • 3
Exploring BERT’s sensitivity to lexical cues using tests from semantic priming
TLDR
We present a case study analyzing the pre-trained BERT model with tests informed by semantic priming. Expand
  • 2
  • PDF
Not So Cute but Fuzzy: Estimating Risk of Sexual Predation in Online Conversations
TLDR
The sexual exploitation of minors is a known and persistent problem for law enforcement. Expand
  • 1
Exploring Automatic Identification of Fantasy-Driven and Contact-Driven Sexual Solicitors
  • 1
Finding Fuzziness in Neural Network Models of Language Processing
TLDR
In this paper, we test the extent to which models trained to capture the distributional statistics of language show correspondence to fuzzy-membership patterns. Expand
L1 Influence on Content Word errors in Learner English Corpora: Insights from Distributed Representation of Words
Authorship Analysis of Online Predatory Conversations using Character Level Convolution Neural Networks
TLDR
We present an authorship attribution model that trains on a corpus of online conversations involving predators and non-predators, and perform subsequent analysis of the message representations. Expand
  • 1
Exploring Lexical Relations in BERT using Semantic Priming
A Sentiment Based Non-Factoid Question-Answering Framework
TLDR
This paper investigates whether emotions should be considered for non-factoid “how” Question-Answering systems with the eventual goal of enabling the system to retrieve answers in a more emotionally intelligent way. Expand
  • 1
Exploring Lexical Irregularities in Hypothesis-Only Models of Natural Language Inference
TLDR
We perform chi-square test for independence between NLI labels and proto-roles on each of the train, dev, and test sets in SPR. Expand
...
1
2
...