Humor generation can be viewed as a special case of automatic text generation, and current humor generation systems can be categorized to two major approaches:

1.	Neural network based systems

2.	Template-based systems


Types Of Humor Generation Systems

To compare different joke generation systems, some basic quality measures are needed. However, until now a standard method of evaluating humor generation systems are still missing. Most approaches use humorousness of the output as an evaluation criterion, and let actual human rate the jokes. The main challenge of this method is that it’s highly subjective. As an alternative, Binsted and Ritchie (1997) suggest that to evaluate jokes based on whether people can tell it apart from a human-generated joke.

Evaluation Of Humor Generation Systems

Humor Generation is the task of creating jokes using algorithms, especially deep neural network models recently. However, until today, there is still no system that can generate jokes like human.

University of Notre Dame

(1) Standard neural networks: real estate, online advertising

(2) Convolutional neural networks: applied in image recognition including photo tagging

(3) Recurrent Neural Networks: applied in sequence data including speech recognition, machine translation

(4) custom/hybrid neural networks: complex applications such as autonomous driving

Applications of neural networks in supervised learning

Amin, M., & Burghardt, M. (2020, December). A survey on approaches to computational humor generation. In Proceedings of the The 4th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (pp. 29-41).

Learn Before

Related

Learn After