English

Interactive Semantic Featuring for Text Classification

Computation and Language 2016-06-27 v1 Machine Learning

Abstract

In text classification, dictionaries can be used to define human-comprehensible features. We propose an improvement to dictionary features called smoothed dictionary features. These features recognize document contexts instead of n-grams. We describe a principled methodology to solicit dictionary features from a teacher, and present results showing that models built using these human-comprehensible features are competitive with models trained with Bag of Words features.

Keywords

Cite

@article{arxiv.1606.07545,
  title  = {Interactive Semantic Featuring for Text Classification},
  author = {Camille Jandot and Patrice Simard and Max Chickering and David Grangier and Jina Suh},
  journal= {arXiv preprint arXiv:1606.07545},
  year   = {2016}
}

Comments

presented at 2016 ICML Workshop on Human Interpretability in Machine Learning (WHI 2016), New York, NY

R2 v1 2026-06-22T14:33:13.686Z