A Dual-Attention Neural Network for Pun Location and Using Pun-Gloss Pairs for Interpretation

Identification of Homographic Pun Location for Pun Understanding

This paper introduces a novel framework for homographic pun location identification. Two observations, the existence of the support term and the preferred positions of the pun in context, are considered as crucial hints for pun identification. We first nominate the pun candidates, and then select the most probable one based on various strategies. Experimental results show the effectiveness of o...

Pun and Non-Pun Humor in Second-Language Learning

We believe that the use of humor by conversational software agents would significantly improve secondlanguage learning. Puns help students to learn about homophony, and to distinguish between words with similar sounds or spellings; idiom-based jokes provide memorable contexts for idioms, which are notoriously difficult to memorize; and scalar humor introduces the student to common comparisons i...

Computational Mechanisms for Pun Generation

Computer pun-generators have so far relied on arbitrary semantic content, not linked to the immediate context. The mechanisms used, although tractable, may be of limited applicability. Integrating puns into normal text may involve complex search.

Phonological Pun-derstanding

Many puns create humor through the relationship between a pun and its phonologically similar target. For example, in “Don’t take geologists for granite” the word “granite” is a pun with the target “granted”. The recovery of the target in the mind of the listener is essential to the success of the pun. This work introduces a new model for automatic target recovery and provides the first empirica...

Adjusting Sense Representations for Word Sense Disambiguation and Automatic Pun Interpretation