Understanding and Quantifying Creativity in Lexical Composition
Why do certain combinations of words such as "disadvantageous peace" or "metal to the petal: appeal to our minds as interesting expressions with a sense of creativity, while other phrases such as "quiet teenager", or "geometrical base" not as much? We present statistical explorations to understand the characteristics of lexical compositions that give rise to the perception of being original, interesting, and at times even artistic. We first examine various correlates of perceived creativity based on information theoretic measures and the connotation of words, then present experiments based on supervised learning that give us further insights on how different aspects of lexical composition collectively contribute to the perceived creativity.

Dataset, binary labels (user labeled)

Word pairs creativity scores (user labeled)

Questions or suggestions are welcome! contact point:

Polina Kuznetsova, Email:

Motivational Examples: Information Measures

Distribution Transformation as a result of Composition:

Motivational Examples: Vector Space Models (red: not creative, blue: creative)

Word Pair Vectors Visualization with Predictions (red: not creative, blue: creative)


Understanding and Quantifying Creativity in Lexical Composition.
Polina Kuznetsova, Jianfu Chen and Yejin Choi.
Empirical Methods in Natural Language Processing (EMNLP), 2013


  author    = {Polina Kuznetsova, Jianfu Chen and Yejin Choi},
  title     = {Understanding and Quantifying Creativity in Lexical Composition},
  booktitle = {The 2013 Conference on Empirical Methods on Natural Language Processing},
  month     = {October},
  year      = {2013},
  address   = {Seattle, USA},
  publisher = {Association for Computational Linguistics},